The spread of iconoclastic Zarathushtran and Abrahamistic cults over Eurasia have all but destroyed most traces of ancient iconic worship. In the midst of this destruction, which relentlessly continues under the crescent and the cross banners, one tradition alone preserves an ancient strain of the Eurasian systems of iconic worship. This is the system of the vaikhAnasa-s, a miniscule group of taittirIyaka-s who are officiants at several major viShNu temples of the drAviDa, Andhra and karnATa countries. The inner worship at richest shrine of modern bhArata, ve~NkaTAdri, represents one of those old unbroken vaikhAnasa lineages, so also several temples in the drAviDa country like kapisthala and chitrakUTa (of Chidambaram). We had earlier pointed out that the vaikhAnasa gR^ihya sUtra system of domestic viShNu worship, closely parallels vidhi-s for iconic worship of deities found in the bodhAyana gR^ihya sUtra and the atharvaveda parishiShTha-s and represent the earliest surviving textual layer of iconic worship among the Indo-Aryans. The rituals of iconic worship followed in these texts suggested that a distinctive form of the iconic worship emerged in the late vedic period with the following key features: 1) ritual actions relating to bathing, installation and empowerment of the icon with specific vedic formulae, modeled after the ritual actions using yajuSh formulae deployed during a shrauta rite 2)vedic offerings made into the aupAsana fire for invocation of a deity, resembling the regular non-iconic vedic household worship of deities. The vaikhAnasa-s additionally also possess a system related to their household worship of viShNu icons that is used specifically for worship of viShNu installed in public temples. The close parallelism of the temple and household worship of viShNu suggests that the earliest temple worship emerged within the vedic tradition as a mirror of the basic household system, even as the shrauta (~temple) and gR^ihya(~household) systems mirror each other in the older vedic layers.
The uniqueness of the vaikhAnasa temple worship is marked by the vaikhAnasa mantra-prashna, which contains several accented mantra-s of clearly vedic nature that are found in no other yajuSh saMhitA, but used exclusively in temple worship. This feature suggests that the vaikhAnasa system of temple worship is indeed ancient and that it predates the emergence of tantrika temple worship of the vaiShNava, shaiva and kaumAra variety. Indeed we find precursors for the tantrika worship in the vaikhAnasa system. The deployment of unique vedic mantra-s exclusively in the vaikhAnasa temple worship suggests that these actual belong to the phase of worship of images in the oldest temples, the chaitya-s, which were present before the nagna and the tathagata made their advent. While the extant vaikhAnasa mantra prashna might show some evidence for latter tampering, the core of it is archaic. The prashna-s 1-4 (vaikhAnasa saMhitA) are mantra-s for the gR^ihya rituals and parallel the bodhAyana and Apastamba mantra prashna-s of other taittirIyaka-s and the mantra brAhmaNa of the sAmavedin-s. The prashna-s 5-8 are devoted to the iconic worship of viShNu, his gaNa-s and other deva-s installed in the temple. The proper understanding of the temple worship using these mantra-s requires a knowledge of the saMhitA traditions of the vaikhAnasa-s attributed to atri, bhR^igu, kAshyapa and marIchi. I had the chance to study a ritual hand book compiled by a great medieval vaikhAnasa ritualist known of nR^isimha vAjapeyin. This lucidly provides the core of temple worship in a very succinct form. The author is reputed by tradition to have been a spectacular triple scholar who was simultaneously well-versed in vaikhAnasa temple ritual, vedic shrauta rituals (he performed the great vAjapeya himself) and tantrika mantra shAstra of various kinds. In the below descriptions we shall follow his description of ritual injunctions and accompanying mantra-s.
The vaikhAnasa is enjoined to recite vedic mantra-s, perform prANAyAma, japa, brahmayaj~na and vaishvadeva homa similar to a smArta before performing any temple action. But the vaikhAnasa is distinguished by the recitation of a unique very late vedic sUkta, Atma sUkta. In addition to his daily pArAyaNa with the Atma-sUkta the vaikhAnasa ritualist also deploys the same sUkta just before he prepares to invoke viShNu into the movable image.
AtmAtmA paramAntarAtmA mahy-antarAtmA
yash chAtirAtmA [drAviDa pATha; AndhrapATha= chAdirAtmA] sa tu [ some reciters= sa tano ] no .antarAtmA
vyAveShTi vishva(g)M sakalaM bibharti
yo vyakta-puNyas sa-tunaH pradhAnaH || 1 ||
prANaH praNItis sa udAna Adir
vara-do varAho vyAnash cha me syAt
tapasA~n cha mUrtiH kapilo munIndro
yash chApAno hayashIrSho naH || 2 ||
yat sarvam ashnAty ajaras samagra(g)M
shriyam Urja-yuktAM sa tu me samAnaH
balam AsuraM yat satataM nihantA
brahmA buddhir me gopa IshvaraH || 3 ||
savitA cha vIryam indush cha dhAtu
rasa-bhUtam-bhUtA bhUtAs sa-bhUtAH
dyaur me astu mUrdhodara-nAbho vA
bhUmir yathA~Nghrir vavR^idhe .aham IshaH || 4 ||
asthIni me syur atha parvatAkhyA
bhujagAsh cha keshA divi ye charantaH
dvau netra-rUpau vithu pR^ishni mukhyau
rudhiraM cha sAra(g)M sakalaM cha toyam || 5 ||
snAyavo me Asan nadIr bhR^igur me hR^idayam astu
sarve anye munayo .a~Nga-bhUtAH
vedA me AsyaM jihvA me sarasvatI
dantA maruta upajihvA upashrutiH || 6 ||
vR^iShaNau mitrA-varuNAv upasthaH prajA-patir
AntrA me vedAsh shruti-smR^itI medAdhAraNe
svedam me varShaM mUtra kosha(g)M samudraM
purIShaM kA~nchanam || 7 ||
sAvitrI gAyatrI maryAdA vedI [vede in certain manuscripts]
hR^it-puNDarIke vimale praviShTas
sakalas sa-lakShmIs sa-vibhUtikA~Ngo
yat sarvaM puNyaM mayy adhiShThAnam astu || 8 ||
sarveShAM devAnAm Atmakas sarveShAM
munInAm Atmakas tapo-mUrtir iha puNya-mUrtir Asan || 9 ||
An examination of the AtmasUkta shows that there is a mix of typical vedic and classical features suggesting its composition towards the end of the vedic productivity, like the nArAyaNa sUkta of the taittirIya-s: The accents show some differences between the reciters from the drAviDa and Andhra countries. Further, some accents are not consistent with the vedic accents just as in the case of late vaiShNava material in the taittirIya AraNyaka. There are words typical of the classical language such as satataM and shruti-smR^iti. Yet it shows vedic elements – the underlying Chandas in large part is a triShTubh (I have arranged it to illustrate that). Importantly, it seems to have no mention of the vibhava-s or avatAra-s of the sAtvata system such as vAsudeva and his vyUha-s. Instead it mentions only 3 avatAra-s explicitly: varAha – identified with vyAna; kapila with vyAna; hayashIrSha with apAna. Some modern vaikhAnasa AchArya-s state that the phrase “yat sarvam ashnAty ajaras samagra(g)M” actually refers to nR^isiMha. The only avatara-s that find mention in the floral worship of icons of various devata-s in the inner and surrounding shrines of a vaikhAnasa temple, according to the core ritual are: kapila, varAha, nR^isiMha, vAmana/trivikrama and hayAtmaka (hayashIrSha) (the penultimate one having two separate images). There is no mention of the dashAvatara-s in the core vaikhAnasa daily ritual manuals. They are only found in the bhR^igu saMhitA’s kriyAkhANDa. This suggests that the Atma-sUkta preceded the classic paurANic period and was possibly also before the sAtvata system. It also suggests that original avatAra system as recognized by the vaikhAnasa-s only had the above 5 avatAra-s. Of these it is not clear if kapila was originally considered an avatAra or merely a respected muni of the early vaikhAnasa tradition.. It is also possible nR^isiMha, vAmana and trivikrama are considered among the 8 vidyeshvara-s of the vaikhAnasa system – indicated in the invocatory formula in the VMP: 188.8.131.52. These vidyeshvara-s might be an early attestation of a class of deities whose cognates were also adopted by the shaiva saiddhAntika tantra-s most probably from their pAshupata precursors.
The main function of the AtmasUkta appears to be the identification of the officiant with viShNu in his cosmic form. This cosmic form of viShNu is implied by the identification of viShNu with the puruSha – the puruSha sUkta is recited after the AtmasUkta in vaikhAnasa practice. In addition to the AtmasUkta the vaikhAnasa-s have the brahmanyAsa rite which serves an identical purpose and is performed before bathing the image of viShNu. The brahmanyAsa is not found in the vaikhAnasa gR^ihya sUtra. It is mentioned in a rather abbreviated form relative to the modern performance in the original saMhitA-s of atri, bhR^igu, kAshyapa and marIchi. In modern performance is shows several tAntrika elements that are found nowhere in the core saMhitA-s which instead use distinctive vaikhAnasa-specific mantra-s. This suggests that brahma-nyAsa was perhaps specifically derived with the emergence of the vaikhAnasa temple ritual. It might have inspired the early evolution of tantrika mantra-mArga and has subsequently been overrun by laterally transferred elements of the tAntrika mantra mArga.
The core brahma-nyAsa comprises of a~Nga-nyAsa, bIja-dhyAna, kara-nyAsa and brahmaikvatvaM. While the modern performances have the usual formulae of hR^idayAya namaH etc. there is no evidence they were used in the old vaikhAnasa ritual. In fact they do not seem to appear in any of the old saMhitA-s, which only specify the basic mantra-s for each nyAsa.
hR^idaya (touching of heart): brahma brahmAntarAtmA brahma-pUtAntarAtmA brahmaNi brahma-niShTho brahma-gupto gupto .aham asmi ||
shiras: dyaur dyaur asi sarva ime prANA sthANava Asan ||
shikhA: shikhe udvartayAmi sa-vedAs sa-mantrAs sArShAs sa-somA devAH parivartayantAm ||
kavacha: devAnAm ayudhaiH paribAdhayAmi brahmaNo mantrair IshasyojasA bhR^igUNAm a~NgirasAM tapasA sarvam Atanomi ||
[This kavacha formula might be translated approximately as: By the deva-s weapons I block [the attackers], by the brahman power of mantra-s, by the manly strength of Isha(i.e. rudra) and by the heat of the bhR^igu-s and a~Ngirasa-s I pervade every thing. The last phrase is reminiscent of the yajurvedic incantation during the laying of the kapAla-s: bhR^igUNAM a~NgirasAM tapasA tapyadhvaM.]
astra: nArAyaNAya vidmahe vAsudevAya dhImahi | tan no viSNuH prachodayAt ||
sudarshanam abhigR^ihNAmi | ravipAm abhigR^ihNAmi ||
dakShiNa netra: sUryo .asi sUryAntarAtmA chakShur asi sarvam asi sarvaM dhehi ||
vAma netra: chandro .asi yaj~no .asi yaj~nAyAdhAnam asi yaj~nasya ghoShad asi ||
Then the officiant carries out bIja-dhyAna by meditating upon the Adi-bIja: a
Some vaikhAnasa-s hold that the akAra bIja should be surrounded by OM-s – this is found in the atri saMhitA and bhR^igu kriyAdhikAra. Then the ritualist performs karanyAsa by invoking the devatA-s: AbhuraNya, vidhi, yaj~naM, brahmA, indra on the five fingers. Then he performs the brahmaikvatva dhyAna with the formula:
antar asminn ime lokAH antar vishvaM idaM jagat | brahmaiva bhUtAnAM jyeSThaM tena ko .arhati spardhitum ||
This formula resembles the atharva vedic kANDa tarpaNa formula (AV-vulgate 19.22.21 and 19.23.30)
Now these vaikhAnasa practices represent an early version of the eponymous rites performed in the tantrika system. But they can also be compared to two other rites: One earlier than all these rituals — the vedic formulae for the identification of the body with the immortal devatA-s: the mantra-s beginning with “agnir me vAchiH shritaH …” (kAThaka 1.8.4-9). The second one is the shaiva ritual of nyAsa, which in turns occurs as the short laghunyAsa and the long mahAnyAsa. Such identifications of parts of the body with devatA-s are also seen in the medical saMhitA-s, like that of charaka. The laghunyAsa includes within it the vedic mantra-s beginning with agnir me vAchiH shritaH. The modern laghunyAsa and mahAnyAsa rites might be used in temples but there is no evidence to suggest that they were originally used thus. In fact they appear to be purely household rites for the worship of rudra by a yajamAna and his family. We have no clear way of dating when these texts were composed relative to other texts either. However, tradition holds that there are two forms of mahAnyAsa – the tAntrika version attributed to the rAkShasa rAvaNa and the vaidika version attributed to bodhAyana. The core of the mahAnyAsa is entirely vaidika and so also the laghunyAsa and these nyAsa-s are mainly performed by taittirIyaka-s. We propose based on the parallels to the vaikhAnasa ritual that these nyAsa-s represent the proto-shaiva versions of their own nyAsa-s. It is quite likely that this system of nyAsa-s are also linked to the period of the emergence of the medical saMhitA-s when they identified parts of the body with deities and are the precursors of the classical tAntrika nyAsa-s. The taittirIya tradition may be seen as the nidus of their emergence. These early vaidika nyAsa-s in turn emerged from the early shrauta ritual of saMbandha with the agnir me vAchiH. But the early development of the temple ritual amongst the vaikhAnasa-s resulted in these nyAsa-s being incorporated into their larger system of public iconic worship.