Fig. 1

Fig. 2

The authors of the very first entry in the catalogue, which treats of this sculpture, do not seem to be familiar with such textual reference to cloud gods or to their significance. Thus they identify the male divinities of the cloud as apsaras [5] . This is indeed a big problem. If the authors cannot distinguish female apsaras from the flat chested male divinities, I wonder how it would be possible to handle the other complexities of art historical study. In the same entry, they argue that the sculpture should be dated to the 5th to 6th century instead of the previously accepted 9th century. The main point of their argument is based on the stylistic similarity of the lotus shown in the nativity sculpture as a pedestal on which the newborn Siddhārtha is standing and the same flower held by the Gana Baha Padmapāṇi. Since the latter can be dated ca. 550 the authors express their opinion that the nativity sculpture also belongs to about the fifth or sixth century. But stylistic study is not that simple. The lotus employed for a pedestal and the lotus held by divinities should not be treated as the same. Compare the lotus pedestal of Gana Baha Padmapaṇi with the lotus he holds (fig. 2). The difference is huge. The pedestal is treated here rudimentarily, rendering only the pericarp of the lotus decorated with vertical linear pattern around its edge, whereas the lotus held by the god is rendered much more elaborately and naturalistically. This means the sixth century Nepalese artist was familiar with the naturalistic treatment of the lotus but it was not used for a pedestal of a Buddhist deity at that time. Such usage compares with that of the beads and flame motif. It appeared for the first time in 467 A. D. when it was used for the flaming edge of Viṣṇu’s shield in the famous Tilaganga Viṣṇu image (fig. 3). But this motif became part of the nimbus only after the seventh century. Thus it becomes clear that the fifth century date for the nativity sculpture is not based on a logical explanation. The languorous, elongated body of Maya and her diaphanous sāri differ widely from the dwarfish proportion, stiffness, and rudimentary treatment of the sāri in the Tilaganga relief (fig. 4) but bears some similarities with those features seen in the twelfth or thirteenth century bracket figures from Uku Baha (fig. 5). The abstract space between the crossed legs of Maya and the Uku Baha bracket figures is almost identical. The pleated middle section of the sāri in both examples cascades down from the waist and goes over the left leg in similar fashion before it terminates into the flower bud like end. Thus the 9th century date given by Stella Kramrisch in her seminal 1964 catalogue remains unchanged. [6]

Fig. 3	Fig. 4	Fig. 5

The significance of the waterholes in the valley can be understood properly only if we give an attention to the fact that the main source of water is believed to be rain and the mother sky itself is considered to be a big hole, mahābila. Newars originally believed that the sky is mother, which sharply contrasts with Indo-European concept of father sky. [15] Rainmaking rituals are performed around these waterholes during droughts and some of these water sources are named after the Ākāśagańgā, the milky-way which is viewed as celestial water in both Sanskrit literature and Newar tradition. [16] The Svayambhūpurāna prescribes worshiping the waterholes (jalotpannarandhrāni) on the full moon day of Śrāvana month, when the copious monsoon rain is expected. [17] The waterholes are believed to be the reminiscent of the primordial lake of the Kathmandu valley. Both Hindus and Buddhists agree that the main deity of the waterholes is Guhyeśvarī, and worship her as their tutelary lineage deities, variously called istadevatā, kuladevatā or degudya. However, the Newar Buddhist tradition of the valley prefers to identify her with the dharmodayā triangle. This is evidenced by a brief statement found in the Svayambhūpurāna, where the symbolic representation of this goddess is described as a yoni like dharmodayā (yonyākārena samjātā … dharmodayāsvarūpena trailokye ca prakhyāpitā). [18] Such vulvial symbols are often represented in stone relief. An seventeenth century inscription found at the western section of the Cilamco stūpa in Kiritpur tells us that such stone relief was known to the Buddhist community of the valley not exactly as dharmodayā but as dharmadhātu shrine. This is understandable because in esoteric Buddhist literature the word dharmadhātu is synonymous with dharmodayā. For instance, Hevajratantra Tīkā (folio 33b) explains that “dharmodayā is dharmadhātu because the worldly phenomenon or the noble dharma [directly] rises from here.” [19]

Such dharmadhātu shrines are very different from the dharmadhātu mandalas which are round and laid flat on plinth-like high structures. The dharmadhātu shrines, on the other hand, are vertical stone slabs with pointed arch adorned with the Garuda or kīrtimukha motif flanked by two makaras (fig. 8). Locally these shrines are also known as torana, “a gate” not only because the shrines appear like the gates of Newar palaces and temples but also because the inverted triangle symbolizes both the vulva and a celestial gate. The author of the Samvarodayatantra 2.25 clearly states that dharmodayā is both a yoni and a gate (dharmodayā-yoni-dvārānām). This concept goes back to the early Vedic period. The Atharvaveda 10.2.31, 10.8.43 for instance, describes the human body with nine apertures as an unconquerable city (ayodhyā) with nine gates.

Fig. 8

Although the gate like Dharmadhātu shrines are seen all over the valley, one finds them also on the valley’s great stūpas including Svayambhū stūpa, Tukan Bahal stūpa in Kathmandu and Pulan Syangu stūpa at the western slope of Svayambhū hill. The famous Dhando stūpa of Chabhil is also surrounded by multiple dharmadhātu shrines of this type. Earlier versions of the shrines are well illustrated in a drawing published in Hodgson's article [20] almost two centuries ago (fig. 8). In a sense the great stūpas of the valley are not only the abode of the five Tathāgatas but also the shrine of the aboriginal mother goddess of waterholes represented by the dharmodayā/dharmadhātu symbol. This is not surprising because the Svayambhū himself emanated from this deity’s waterhole.

There is an explanation for why this goddess of the dharmodayā/dharmdhātu is known as Khagānanā, "Bird Faced." The Candamahārosanatantra (15. 16) and its commentary by Kumāracandra inform us that in esoteric Buddhist literature the word khagamukha “the face or the bill of a bird” means female genitalia. [21] Since the term khagamukha is synonymous with khagānana in Sanskrit, it become abundantly clear that the mother goddess was known as “Bird faced” because of her association with yoni.

The yoni symbol is represented in South Asian art in many different ways. It is true that in some example the yoni is depicted almost like the bill of a bird. [22] To my knowledge, such example is not available in Nepal. But in this country the yoni symbol is depicted almost always within a torana. According to Newar legend the bird depicted on top of the torana or dharmadhātu shrine is not exactly Garuda of Sanskrit literature but a mythical bird Chepu who controls celestial and terrestrial water that appear in the form of male and female serpents. The authors of the Tantric Buddhist texts also deny the identity of the bird with Garuda. Thus they use a technical word kramaśīrsa "he head of the succession" as a synonym for the apex of the torana. In a description of the elaborate torana of a mandala painting, the author of the Pindīkrama, refers to this bird as a kramaśīrsa paksinī "female bird perched at the apex of a [torana]." [23] Although we do not know exactly where the Pindīkrama was written perhaps in Nepal, this particular reference to female gender of the bird is apparently the initial step toward the Nepalese interpretation of the bird-faced female deity in association with torana and dhormodayā.

Fig. 9

Art historically, however, the bird on the torana is the metamorphosis of the ancient kīrtimukha motif symbolizing four atmospheric directions and center. In my previous work, I have explained that although the legend of Chepu or female bird is not directly related to such artistic development, its symbolic association with atmospheric phenomenon has remained intact in the valley even in the late seventeenth century. The entire shrine with the kīrtimukha or the bird motif on the apex of torana is part of the iconography of the tutelary lineage deities worshiped by Newars before and after monsoonal rain. [24] Hindu tutelary deities are also represented in similar shrines. In those cases the shrines are not designated as dharmodayā or dharmadhātu but simply a torana. Thus we have good reason to believe that not only the symbolic identity of the yoni with the celestial gate, but also the popularity of the analogy between the yoni and the bird seems to be the main reasons that the Buddhists in Nepal prefer to represent the symbol of yoni/dharmodayā within the torana.

Fig. 10

At the Southeast corner of Svayambhū stūpa there is an eighteenth century shrine decorated with the Newar style torana, the dharmodayā/dharmadhātu, with the Chepu bird on top, but with two round openings instead of an inverted triangle as expected (fig. 9). That it is simply a different version of the typical shrine is borne out by a seventeenth century painting in the collection of the Cleveland Museum of Art (1973.69) (fig. 10). As Bangdel has noticed, it depicts a similar shrine, which seems to be that of the istadevatā mentioned in the inscription at the bottom of the painting. It is highly possible that this tutelary deity is Khagānanā, an identity that depends upon the family tradition of the donors mentioned in the inscription. It is interesting to observe that such a torana shrine closely resembles Tibetan gau, a portable shrine, which in turn is the cognate of the empty niches of the early monolithic caityas. The opening of the gau is known to the Tibetans as sgo-khuin “doored space,” undoubtedly because the shrine is no other than a torana.

The nineteenth century Nepali Pundit Amrtānanda, who helped Brain Houghton Hodgson to study Newar Buddhism, was familiar with the torana shrine attached to the Svayambhū stūpa. He describes it briefly in his unpublished work Dharmasamgrahakośa and expresses his view that the shrine may represent Manjuśrī, Vairocana or Vajradhātvīśvarī. Since he could not identify it exactly, he concluded that this empty shrine actually represents śūnyatā. More recently in 1978, when the famous Buddhist Pundit Hemaraja Shakya wrote his book on Svayambhū stūpa in Newari language he identified the shrine as Vajradhātvīśvarī. [25] This identification is based on Amrtānanda’s work but without referring to other possibilities mentioned in the source. Since then local informants and tourist guides, with great enthusiasm, have been explaining the significance of the shrine to visitors. Depending on such information, Huntington found no problem identifying the shrine as Vajradhātvīśvarī, which helped him to show her identity with Prajnāpāramitā and the relation between the Svayambhū stūpa and Vairocana sādhana of Guhyasamāja. [26] The authors of the catalogue discuss the shrine repeatedly and have published its photographs in two different occasions. But they did not make any attempt to read the short inscription given at the bottom of the shrine, which can be translated as follows:

“In the year of 872 (A. D. 1752) on the fifth day of the bright half of the Āsādha month [this] divine Dharmadhātu … is commissioned by the donor Gyanapati.”

Clearly this is a dharmadhātu shrine, a typical feature on the stūpas of the valley. Here one can argue that both dharmadhātu and Vajradhātvīśvarī symbolize śūnyatā; therefore the shrine actually belongs to Vajradhātvīśvarī, whose significance in turn is based on the fact that she is no other than Vajravārāhī. Such traditional approach, which is based on synthesis instead of analysis, is not helpful because it prevents us from detecting the amalgamation of pre-Buddhist religious elements into the esoteric Buddhist pantheon. In fact identification of two divinities bearing different names always indicates a new layer of development.

Fig. 11

At the beginning of the nineteenth century there was a similar dharmadhātu shrine at the northeastern corner of the Svayambhū stūpa as testified by the Laksacaitya painting, dated 1809, in the Avery Brundage collection at the Asian Art Museum of San Francisco (B61 D10+) where one is clearly represented (fig. 11). As we know from the Svayambhūpurāna originally this northeast corner of the stūpa was characterized by a celestial hole (vivara). [27] Apparently a main purpose of Gyanapati’s commission of a dharmadhātu in this southeast corner was to create symmetricality. Compare Hodgson’s drawing of Chabahil stūpa (fig. 8). It is also important to note that the second representation of Svayambhū stūpa in the lower section of the Laksacaitya painting shows two Dharmdhātu shrines flanking Aksobhya’s temple. Thus the round hole at the bottom of Gyanapati’s dharmadhātu shrine apparently echoes the original hole at the northeast corner. Furthermore the stylized lotus petals around the hole in the dharmadhātu shrine immediately recalls the typical shrine of goddess Khagānanā at Hmasinga. In the Svayambhūpurāna [28] this goddess of a fresh water spring, is described as jalaskandha “a mass of water” and in the same text the primordial lake Kalīhrada of the Kathmandu valley is designated as dharmadhātu. [29] Clearly in Newar Buddhism the word dharmdhātu symbolizes not only śūnyatā but also water. The hidden logicality behind this association of śūnyatā with water is based on the view that śūnyatā is the sky, which is the main source of life giving water both celestial and terrestrial. The sky is, however, a big hole, a gate, a yoni symbolizing the great mother goddess, Khagānanā. Unfortunately, the authors of the catalogue did not even try to understand the significance of Khagānanā, mentioned in the inscription of the laksacaitya painting, and simply identify her as Guhyeśvarī / Vajravārāhī. [30]

The catalogue is indeed a detail study. The introductory essays and the entries are written so elaborately that quite often they exhibit redundancies such as the valley is a mandala. Thus, the short articles, published in Orientations before the exhibition, are the better examples of their methodological approach than the catalogue itself. For instance, in his article Huntington writes:

In its present iconographic iteration, for which, as mentioned, there is direct physical evidence of having been in existence since the sixth century, the [Svayambhu] Mahachaitya is a polyvalent symbol of considerable complexity. [31]

According to this statement Svayambhū stūpa has multiple symbolism and the present iconography of the stūpa is the iteration of an early version that goes back at least to the sixth century A.D. Nowhere in this work Huntington suggests that earlier symbolism of the stūpa might be different from later interpretations. He believes that the deities of the shrines around the stūpa are arranged in accordance with the Pindīkrama sādhana of the Guhyasamāja Tantra. He writes:

… it can be seen that the Vairocana shrine is just to the South of Akshobhya’s. In this meditation, the practitioner visualizes himself at the east as Akshobhya in non-dual state with Sparshavajri/Mamaki. In the completion stage, the practitioner as Akshobhya/Sparshavajri is in the centre of the mandala and looks to the east to realize that Vairochana is present in no-dual state with Saptalochana, who is represented in the northeast shrine. Thus, for part of the meditation, Vairochana is actually on the east and his shrine is appropriately physically located there on the Mahachaitya. [32]

This is indeed a fascinating statement, but also a perfect example of his defective methodology. Admittedly one can study art historical materials as they are understood by a group of people in particular period of time and space. Such synchronic study may be valuable but it should not be accepted either as complete or accurate. In fact due to the lack of historiographic analysis such study is often dangerously misleading. This is the main reason that Huntington’s methodology is open to criticism.

His argument that the iconography of the stūpa is in perfect harmony with the Pindīkrama Sādhana of the Guhyasamājatantra is mainly based on the location of the Vairocana shrine at the eastern section of the Svayambhū Stūpa. But this shrine did not exist there before 1713. The Svayambhūpurāna , most of which was written around the time of King Yaksamalla's rule (ca. 1428-1482),  refers to only four Buddhas: Aksobhya , Ratnasambhava, Amitābha, and Amoghasiddhi at four directions of the stūpa. Since Vairocana was considered to be in the middle of the stūpa his shrine was not represented then outside the dome of the stūpa. According to contemporaneous diaries known to the Newars as thyasaphu the earliest image of Vairocana at eastern section of the stūpa is the contribution of the queen mother Bhuvanalaksmī, who was active during the first quarter of the eighteenth century. [33] Almost exactly a hundred years later, the Newar Buddhists of Kathmandu replaced the original contribution of the queen with much bigger statue of the god when they renovated the stūpa in 1816. However, the image of Vairocana that we see today is of even more recent date. According to the pedestal inscription, the image was erected only in 1918. Although most of these materials were at least partially published in Nepalese publications many decades ago, Huntington does not seem to have taken much time to study them. This was grievous omission. It led him on faulty grounds to adduce a startling new theory proclaiming that the iconography of the stūpa, presumed static since at least the sixth century, is based on the meditational practice prescribed in the Guhyasamāja. As the historical materials amply illustrate this cannot be substantiated. Similarly, his view that the image of Saptalocanā Tārā to the left of Aksobhya represents both Māmakī and Saptalocanā is unacceptable. [34] In 1918 when Svayambhū stūpa went through a massive renovation the Buddhist priests of Kathmandu rearranged the images of the Tārās, [35] clearly because there was a discrepancy in the scheme of the five Tārās’ association with the five Tahāgatas, apparently which was not settled even in the twentieth century. Without examining these historical documents one cannot jump to the conclusion that the image at the northeastern shrine of the caitya stands for both Māmakī and Saptalocanā alternately.

Trained in this inadequate methodology, Bangdel begins her own investigation and claims to explore the origin of Buddhist divinities of Kathmandu valley. Unfortunately, however, her study of Nepalese materials is also characterized by the weakness in understanding the difference between original and secondary development. For instance, explaining the significance of Vasudhārā or Vasundharā she writes:

For the Newar Buddhist lay practitioner, the great esoteric deity, Vajravarahi/Vajrayogini, is remote and aloof in the extreme, yet powerful beyond the wildest imagination. It is through her exoteric manifestations that the full force of her wisdom and power come to fruition for the laity. In Newar Buddhism, one of the preeminent exoteric manifestations of Vajravarahi is Vasundhara, one of the most beloved of the Buddhist goddesses. [36]

Fig. 12

According to this statement the significance of Vasundharā or Vasudhārā is based on the fact that she is the exoteric manifestation of Vajravārāhī. This reasoning does not differ much from the traditional explanation of the Newar Buddhist priests of Kathmandu. Bangdel simply accepts it without giving much attention to the fact that Vasudhārā’s pre-Buddhist association with agrarian prosperity can be detected easily with a careful observation of her iconographical features. The main attribute of the goddess is dhānyamanjarī, a Sanskrit compound word for “sheaf of rice paddy” (fig. 12). Although Bangdel translates it sometime as “sheaf of grain” and other time as “sheaf of wheat” we know for sure that in classical and Buddhist Sanskrit dhānya means rice paddy (Nepali dhān). Here it is important to note that in Nepal rice is harvested in the beginning of autumn, wheat in the winter. Thus Vasudhārā is worshiped annually on the first day of autumnal rice harvest known to the Newar farmers and Buddhists as Gātilā. During the Gātilā ritual, her image is placed on top of a heap of rice and an offering of rice paddy to her is considered mandatory. Clearly Vasudhārā is a deity presiding over fresh paddy. This is why she is conceived as yellow, and holds a sheaf of rice paddy as her main attribute. Illustrated Newari manuscripts in the National Archives in Nepal (A. 151; A. 153) dealing with auspicious and inauspicious sights often represent a heap of rice as very auspicious and identify it as Vasudhārā, so labeled immediately below the representation. Once we are familiar with such original significance of the goddess it becomes obvious that her affiliation with Vajravārāhī is secondary development. Such analysis is completely absent in the catalogue despite the fact that authors claim to investigate the origin of the valley’s Buddhist divinities “before the introduction of either Hinduism or Buddhism to the valley.” [37]

The lack of proper knowledge of Sanskrit is also a monumental problem for Bangdel. This is evident in many places e.g. in a following example. In her frequently quoted work as a main source for the catalogue, Bangdel describes Manjughosa "rejoicing [in sexual embrace] with Vajrasattva." [38] This embarrassingly erroneous translation, which is given twice in her work within few pages in the same chapter, resulted from the misspelling and misunderstanding of following Sanskrit sentence—iha bhagavān mahāvairocanātmā manjughosah suviśuddhadharmadhātujnānasvabhāvah svābhavajrasattvena mudritah, which actually means "Here Bhagavān Manjughosa stands for Mahāvairocana and symbolizes pure dharmdhātujnāna; [his crown] is marked by [the image] of Vajrasattva who bears similarity [in complexion] with (Manjughosa) himself." The correct reading of this text is found in Bhattacharya’s edition of the Nispannayogāvali. [39] But Bangdel did not understand the meaning of the technical iconographical terms svābha “bearing similarity (with his or her own color)” and mudritah “ marked, (a headgear marked by the image of a superior deity). Thus she changed svābha into svabhā and mudritah into muditah to obtain desired translation. Although space does not allow me to write in detail, this is not an isolated example of the desired translation but a trend throughout their works. The Newari inscription found in Jivarama’s sketchbook, for instance, does not contain any word or words to prove that “Jivarama personally wrote this [inscription].” [40]

Lack of the knowledge of Newari language and culture is yet another big problem of the authors. Huntington is a specialist of Tibetan art. His study of Newar Buddhist art is his new interest. In fact he began to investigate Newar art intensively only after Dina Bangdel, a daughter of Nepali art historian Lain Simh Bangdel, began to work with him. But Dina could not help him much because her knowledge of Newari language and culture was limited as well. Thus she argues that the appearance of Tibetan goddess Palden Lhamo in a Los Angeles painting is helpful to pinpoint the provenance of the painting as Tibetan. [41] But she is not familiar with the fact that according to Buddhist Newars Chvāsakāminī, or Ugracandī, the dreadful goddess who resides at chvāsa is Palden Lhamo. Chvāsa is a place originally distant from the Newar residential area where trash and debris are deposited and where are ritually thrown the clothes and other belonging of a dead person, as well as newly born children’s umbilical cords. She causes and cures pediatric diseases. An examination of Newari paintings reveals that she is popular in them as well as in Tibetan paintings. Likewise, Bangdel does not seem to be familiar with the fact that the Newari name Syeńgu or Sińgu is derived from sā-hmeńgu a compound word from classical Newari. As we know from the Newari translation of Amarakośa, a celebrated Sanskrit thesaurus, it literally means “a cow-tail.” Therefore in the Svayambhūpurāna this compound word is synonymously used for Gopuccha, the Sanskrit name for the Cow-tail Hill of Svayambhū. When the hilltops were visualized as horns it is also designated as Gosrńga “Cow’s Horn.” Unaware of the different meaning of these Newari and Sanskrit words, Bangdel combines both Sanskrit and Newari terms and creates a new word for the cow-tail hill Gosingu. And Huntington simply accepts it.

Before we conclude this review article, it will be prudent to examine briefly Huntington and Bangdel’s major study regarding the cult of Cakrasamvara and Vajravārāhī in Kathmandu. Prior to them, several scholars contributed on this subject. David Gellner, for instance, treated the subject in detail showing the relation between Cakrasamvara's mandala and the ritual of  pīthapūjā  "the worship of power places" in the valley which was visualized by the practitioners as the mandala of deities. Gellner correctly pointed out that the three pilgrimage circles were interpreted as standing for the circles (cakra) of the mandala, body, speech, and thought (kāya, vāk, citta). [42] Huntington and Bangdel incorporated this view in their work as their main theme. Apparently, well-known Tibetologist Robert Thurman is not familiar with the works on Newar Buddhism. Thus, in the foreword he highly praises the authors for “their discovery of a thriving practice still alive among the Newars of Nepal”.

It is true that the cult of Cakrasamvara plays an important role in modern Newar Buddhism. This is not however the reiteration of great antiquity. The early phase of Tantric Newar Buddhism of the valley was devoid of Cakrasamvara’s cult and remained so until it overlapped with the second phase, which was dominated by the cult of Cakrasamvara and Vajravārāhī. In order to study epigraphic and textual evidence related with this development we need to be familiar with an important architectural element of the Newar Buddhist monastery. This architectural element is locally known as āgamchem. The word āgam apparently derives from the Sanskrit word āgama "learning," and the second word chem is Newari and literally means an edifice. The significance of this edifice is based on the fact that esoteric rituals are performed in the second floor of this edifice. But its ground floor contains an exoteric shrine of Buddha Śakyamuni or one of the five Tathāgatas, who are known to the Newars as kvāpādya, a generic term used for all these Buddhist divinities when they are housed in that particular shrine. The etymology of "kvāpādya" is not clear but the well accepted Sanskritized version of this word is kosthapāla , which literally means "protector of a room". In this exoteric shrine the Buddhist priests and devotees perform regular devotional pūjā. My own observation indicates that most of the time the kvāpādya is Aksobhya and the main tantric deities of the esoteric shrine in the second floor of āgamchem often correlate with the iconographic identity of the kvāpādya enshrined in the first floor.

Earliest reference to such edifice with double shrines is found in a copper plate inscription dated 1388. This important inscription, which is in situ on the wall of the āgamchem of Hnaikam Bahi in Kathmandu, tells us that during the reign of Sthitimalla (ca. 1382-1395) a Buddhist devotee consecrated a multistoried shrine as well as an image of  kosthapāla  Buddha (kvāpādya) in a pre-existing Buddhist monastery known as Kirtipunya Mahāvihāra.  According to same inscription the upper story of this newly built shrine housed an image of Heruka Yāmalaka “Heruka in pair”. [43] Very likely this Heruka, is no other than Cakrasamvara because another Buddhist inscription (dated 1511) found in Viśvakarma Vihāra in Kathmandu refers to the establishment of new shrine of Heruka who is described there as the incarnation (avatāra) of Aksobhya. [44] Indeed the family head of Cakrasamvara is Aksobhya. For this reason the headgear of the early images of Cakrasamvara bears Aksobhya's image. It is true that the epithetic name Heruka is also used for Hevajra. But this god is not the member of Aksobhya's family. Note also the fact that the date of the earliest epigraphic reference to Heruka coincides with the growth of the popularity of artistic representations of Cakrasamvara and Vajravārāhī in Nepalese art.

Despite such popularity of these divinities, Newar Buddhist monasteries, which followed earlier convention, remained without a shrine for Cakrasamvara and Vajravārāhī even in later time. An inscription dated 1593 in situ near the entrance of Otubaha in Kathmandu informs us that the construction of Buddhist monasteries of the valley and the consecrations of the images in the monasteries closely followed the rules and rituals prescribed in Kuladatta's Kriyāsamgraha. [45] This important text is a manual of the Buddhist rituals for constructing and consecrating monasteries, stūpas, and images. According to Gustav Roth, this work “belongs to the category of of Buddhist Kriya-tantras of the eighth and ninth centuries AD” [46] The text begins with a salutation to Vajrasattva, who is treated there as the main tantric deity. This text, however, neither refers to Cakrasamvara nor suggests that it is compulsory to have a shrine of the deity in a Buddhist monastery. When we study this textual evidence and compare it with the information found in the Otubaha inscription (dated 1593) and the earliest reference to Heruka Yāmalaka’s shrine in Hnaikam Bahi inscription (dated 1388) it become abundantly clear that for a period of time the cult of Cakrasamvara developed in overlapping mode. Eventually, the Kriyāsamgraha became obsolete mainly because this early manual could not fulfill Newar Buddhists’ growing interest in the cult of Cakrasamvara and Vajravārāhī. As a result this text was gradually replaced by Jagaddarpana's Kriyāsamuccaya , another manual of Buddhist rituals of later time written after the twelfth century. The main tantric deity of this text is Cakrasamvara and Vajravārāhī. At the beginning of the work, the author salutes them as the most prominent divinities. The significance of this manual for our study, however, derives from the fact that much of the contemporary Newar tantric tradition including the rituals associated with Cakrasamvara and Vajravārāhī is mainly based on this manual instead of Kuladatta's Kriyāsamgraha. Although a comparative analysis of these two manuals awaits a longer study, even a brief observation like this clearly indicates that the cult of  Cakrasamvara  and Vajravārāhī is the new element that dissects early and later approaches of Tantric Buddhism of Nepal. The authors of the catalogue, once again, did not make any attempt to study such historical development of the cult in Nepal although all these epigraphic documents were published many years ago.

Finally it should be pointed out that there is plenty of evidence to show that in the milieu of the ongoing interaction between the Hinduism and Buddhism, Budddhist intellectuals including the author of the Svayambhūpurāna did not have any reason to show hesitation in assimilating Shaivite elements into Buddhism. Even śivalińgas are considered in the purāna as Vītarāga or Vaitarāga Bodhisattvas. [47] The list of Shaivite pīthas given in an unpublished manuscript called Nepāla-mandala-pītha-pūjā-vidhi (4. 214, reel no. b. 189.16) in the collection of National Archive of Nepal do not differ much from the three circles of Buddhist pīthas described in the Svayambhūpurāna. Furthermore, it is difficult to deny the similarity between Samvara and Natarāja as exemplified by various attributes. They have in common such as crescent moon, elephant’s hide, dance posture, khatvāńga, and Brahmaśiras. Unfortunately, however, in the methodology of the authors of the catalogue cross-religious analysis is almost completely missing. Moreover, any information suggesting Buddhist divinity’s association with Hindu deity is avoided sometime removing the name of a Shaivite deity such as Virūpāksa in translation although the god is clearly mentioned in the inscription. [48] Such unscholarly approach apparently stemmed from almost devotion-like over emphasis on the cult of Cakrasamvara and Vajravārāhī.

I would like to thank my younger sister Sumati Vajracharya for assisting me to collect important documents from the Vajracharya priests in Kathmandu. I am grateful to Prof. Alexander Rospatt for the information dealing with the Vairocana image commissioned by queen Bhuvanalaksmi. Furthermore, I would like to express my gratitude to Prof. Joanna Williams and Dr. Mary Slusser for providing me with excellent detail photographs of Nepali sculptures that I have used here for illustrations.