The provenance of the aShTA~NgasaMgraha
Before we go on to a survey of the kaumAra material incorporated into the aShTA~Nga saMgraha (AS) we shall consider some issues related to the provenance of this text. The aShTA~Nga saMgraha is a text in mixed prose and metrical form. There is a closely related text the aShTA~Nga hR^idaya saMhitA (AH) which is in metrical form and an abbreviation of the former. This latter text is the one typically listed as the last of the vR^iddha-trayi. The medical practice along the lines of the AH is a living tradition in India and is maintained to this date in the chera country by certain families known as the aShTAvaidya lineages. These families have a tradition of daily pArAyaNa of sections of the AH. The AH was also transmitted to the Moslems and Tibetans and is the basis of one of the foundational texts of the medical tradition of the latter. Despite this fame we have little knowledge of the authors of the AH and AS or the time of its composition. Much useful work on this particular topic has been done by two renowned scholars Priyavrat Sharma and G.J. Meulenbeld, which might be looked into for a more scholarly exposition on the topic. I am merely providing here a summary of my own inferences, which are limited by my more cursory study of the entire medical corpus relative to the more detailed study of the kaumAra material.
Clearly the AS is the ancestor of the AH because of the presence of two other texts cited by the famed medieval eastern physicians nishchala and shivadAsasena known as madhyavAgbhaTa and svalpavAgbhaTa. The surviving quotations of the madhya text show that it was also a mixture of prose and verse, but some of the prose text from the AS has become verse in it. This suggests that the AH evolved from the AS through a process of serial versification via madhya and svalpa intermediates. Tradition ascribes the authorship of both AS and AH to vagbhaTa. However, we have internal evidence for the author only in the case of the AS, where the author says he is vAgbhaTa, the son of siMhagupta, and the grandson of another vAgbhaTa a great physician. He says he was a native of the saindhava region (modern Sindh, currently occupied by the Mohammedan terrorist state) and compiled the saMgraha, based on his own study of numerous prior works, and what he learned from his teacher avalokita and more so his own father. Beyond this we have little concrete information about this famous medical author. We can infer that despite the many archaisms he was well after the charaka saMhitA and sushruta saMhitA were compiled, given that he cites portions including dR^iDhabala’s redaction of the former and appears to follow the latter even in the uttara tantra (corresponds to uttara sthAna in AH). He was definitely before 1190 CE, the beginning of the era of the extraordinary Hindu polymath hemAdri the prime minister of the yAdava kingdom who wrote a commentary on the AH. Given that either the AS or the AH (we simply cannot be sure which one) was probably alluded to by the Chinese traveler I-ching (690s CE) and the Moslem translator ibn Duhn al-Hindi (780-800 CE) some form of the text was present before this period. Priyavrat Sharma correctly notices certain similarities to the bR^ihat saMhitA of varAhamihira and the AV parishiShTa-s and suggests a temporal link. But this is not entirely convincing, especially given that Hindu authors are well-known to use material from much older texts quite freely (e.g. charaka itself in the case of the AS). Texts resembling AH or AS are found in the garuDa purANa, viShNu dharmottara and agni purANa, which, taken together with the evidence for presence of madhya and svalpa versions between the AS and AH, suggest a history of floating texts within the aShTA~Nga tradition, undergoing duplication and divergence, with little certainty as to when the original author AS operated.
A more decisive feature that might provide a hint regarding the date of the AS is the presence of multiple bauddha mantra prayoga-s like AryAvalokiteshvara, AryAparNashabarI, AryatArA, AryAparAjitA, bhaiShajyaguru, mahAmAyUrI and tathAgatoShnISha. While some of these deities are seen emerging relatively early in the Common Era, others are attested only a little later
[some approximate dates of attestation: avalokiteshvara in sukhAvatIvyUha with amitAbha and mahAsthamaprapta -200 CE; amitAyur-dhyAna sUtra 423 CE in chIna; 500s of CE mantra-s of ekAdashAnana avalokiteshvara translated into chIna; saddharma-puNDarIka sUtra 550 CE; bhaiShajyaguru: Dun-huang 624 CE; bhaiShajyaguru-sutra 457 CE in chIna; AryatAra~500s CE; mahAmAyUrI ~350 CE; parNashabarI ~500s of CE]. Based on this we can provide an approximate window of 300 to 600 CE for the composition of the AS, which ironically matches the time frame of varAhamihira suggested by Priyavrat Sharma. Thus, we can say with some certainty that the kaumAra material in the AS represents the development of the mantra mArga of kaumAra shAsana before or in this period. It probably represents the condition of the kaumAra sect in the yaudheya-gupta-vAkATaka era during which the worship of skanda was very prevalent. Though given the overlap with the mantra-s from sushruta it may also include some material from an earlier phase.
A possible link of AS to atharvavedic tradition
The AS and AH show an interesting mélange of AstIka and bauddha elements occurring side-by-side. This has led to the major debate as to whether vAgbhaTa was an AstIka or a nAstIka. We have vagbhaTa’s own testimony that his teacher was avalokita and this name makes it almost certain that his teacher was a bauddha. We also observed from varAhamihira and viShNudharmottara that the AstIka-s in that period were pretty eclectic and incorporated bauddha elements within their framework. This could explain vAgbhaTa’s bauddha elements, though given his repeated reference need for reverence towards the veda-s, brAhmaNa-s and vaidIka devata-s it is certain that he was dvija himself. As Priyavrat Sharma notices, he seems to have a certain affinity to AV-parishiShTa-s. This is reinforced by the fact that he declares Ayurveda to be an upaveda of the AV. In the first chapter of the uttarasthAna he mentions that a brAhmaNa practicing the atharvaveda should perform shantikarman for 10 days upon child birth. The features circumstantial support vAgbhaTa’s alliance with the AV tradition. The kaumAra rituals of the AS are associated with certain archaic mantra-s that appear to be of late vedic provenance. In light of the above of we suspect that AS preserves and draws in part from mantra-s found in lost parishiShTa-s of alternative AV traditions (perhaps paippalAda as against the predominantly AV shaunaka tradition that survives in large part in the extant parishiShTha-s). Given the presence of the skanda yAga in the extant parishiShTa-s, it is not unreasonable to suspect that the beginnings of the kaumAra tantra were in such AV parishiShTa-s. This is especially so given that they also contain elements of other proto-tAntric mantra-shAstra systems like Asuri kalpa and the uchChuShma kalpa. In this context it should be noted that even the kAshyapa tantra, which incorporates considerable kaumAra material is derived from the original work of a bhArgava of the clan of r^ichIka. One of the chapters containing kaumAra material in the kAshyapa tantra is the famed revatI kalpa, which also contains archaic language reminiscent of the late vedic layer rather than the classical period. An important point to note is that large parts of the AS or the kAshyapa tantra do not have the archaisms of mantra-s or language associated with the sections enriched in kaumAra material [though the kAshyapa tantra in general appears to be a bit more archaic than AS]. Thus, it is unlikely that this is an intentional imitation of an archaic style but actual preservation of an old layer of kaumAra material [In the case of the kAshyapa saMhitA it might even be a late lost brAhmaNa].
An outline of the kaumAra material in the AS
The kaumAra material in the AS that of our interest is found in the uttarasthAna, wherein chapters 1-6 cover pediatrics and chapters 7-8 bhUtavidyA. Chapter 1 mentions a rite performed to agni and skanda in the 4th month after birth, when the child leaves the birthing chamber; this is not mentioned in any of the other texts. Beyond this the main material of interest is found in chapters 3-6. Chapter 3 (bAla-graha-vij~nAnIya) provides the story of how the graha-s emerged. In the long past days, rudra had generated 5 male and 6 female graha-s as the guardians of skanda. The male graha-s are: 1) skanda himself, 2) vishAkha, 3) naigameSha, 4) shvan, 5) pitR^I. The female graha-s are: 1) shakunI, 2) pUtanA, 3) shItapUtanA, 4) andhapUtanA, 5) mukhamaNDitikA, 6) revatI, and 7) suSkarevatI. After skanda had attained maturity and became the leader of the forces of the deva-s, rudra as the graha-s to infect children in the house of parents who refuse to observe rituals to the deva-s and pitR^i-s. There are described as assuming the form of bears, owls or cats and can only be seen with the knowledge of the mantra-shAstra as they infect children:
Avishantashcha lakShyante kevalaM shAstra-chakShuSha shuddhena dehaM bAlAnAM gandharva iva yoShitAM ||
Their entry into children, even as the gandharva seizes a maiden, can only be seen by means of the pure eye of the shAstra [remember the vedic allusion of the gandharva seizing the daughter and wife of pata~njala kApya].
Chapter 4 describes the creation of a yantra on birch bark and tying it with the pratisara mantra which invokes indra, yama, kubera, varuNa, rudra, skanda, brahmA and the transfunctional goddess under many names. This is followed by a description of an associated baliharaNa rite. Then the description of a distinctive maNDala for invoking kumAra as bhUtapati accompanied by the graha-s as his parivAra-s along with sapta mAtR^i-s is provided. Then the satya-vachas mantra for pacification of kumAra graha-s is performed along with a bali offering. Then the celebrated kula-vidyA mantra for the worship and pacification of the 13 skanda-dUtI-s in conjunction with ShaNmukha is described. Then the agnidaNDa mantra along with prognostication using it is described.
Chapter 5 Has the famous kIla and snapana rite in the kaumara context similar to that in the revatI kalpa of the kAshyapa tradition. While the rite in general resembles the rite of bathing in the AV parishiShTa 42, it includes a distinctive fire offering to skanda, agni and the kR^ittikA-s. It also describes the preparation of a medicinal bath for children during this ritual.
Chapter 6 has a description of the rituals, mantra-s, drugs to be prepared and their deployment for each of the 12 graha-s described in chapter 3.
Survey of particular prayoga-s
The maNDala rite AS uttara-sthAna 4.41-49
atha.a.apatita govarcaH pralipte darbha saMsR^ite |
vR^itte vA chaturasre vA maNDale kusumojvale ||
nAnA-graha-parivAraM bhiShag bhUtapatiM likhet |
taM prati prA~Nmukho vidyAM paThan-nupaharet balim ||
namo vimArakasya [perhaps corrupt; manuscripts need to be examined] |
namaH kumArA~NgAya |
namaH saptAnAM mAtR^INAM |
namaH skandAya |
eSha parigraha mantraH |
mAtR^INAM AvAhanaM cha|
hili hili nimApaTalini svAhA |
anena sapta-kR^itvaH parijapitena pAnIya chulukenAtmAbhyukShayitavyaH | evaM AtmarakShA kR^itA bhavati |
The place is plastered with cow dung and strewn over with darbha straw. Then the maNDala is drawn either of a square or a circular form and decorated with flowers. The physician then writes bhUtapati in the middle with the many surrounding graha retinues. Then facing east the bali is offered with the provided mantra-s to kumAra and his hordes and the seven mothers. The seven goddess are invoked with a formula which is reminiscent of the formula in the offering to the goddess mAta~NgI in the kAshyapa tantra’s revatI kalpa. The mantra is muttered seven times and water is taken in cupped palms and sprinkled on oneself/the child. With that one protects oneself/the child. After the offering of the bali mantra is recited known as satya sAdhana.
indrANI, one of the 7 mAtR^i-s with a bAla (kindly provided by shrI Sarvesh Tiwari)
An important feature of this rite is the connection of kumAra with seven mothers. This connection is seen in the early narratives of the kaumAra tale, both in the vana-parvan and shalya-parvan of the mahAbhArata (e.g. the reference to seven mAtR^i-s: kAkI, halimA, rudrA, bR^ihalI, AryA, palAlA and mitrA in the former and sapta mAtR^igaNash chaiva … in the latter). Iconographically this depiction appears prominently in the gupta era cave temples in Madhya Pradesh where the seven goddesses come with a large image of kumAra. In the earlier Mathura temple ruins we find remnants of the 7 devI-s, like indrANI holding bAla-s, in the vicinity of the old kaumAra shrines. Interestingly, a mention of this association also appears in the inscription of Talagunda inscription of king mayUrasharman the kadamba, where mentions kumAra and the sapta mAtR^i-s bearing him aid to defeat his pallava enemies. It is also seen in the damaged Bihar Stone Pillar inscription of skandagupta. Several early chAlukya inscriptions mention how kumAra and the sapta mAtR^i-s protects their clan. Thus, it appears that the famous seven mothers who later as the 8 mothers become a central feature of the shaiva and shAkta tantric systems had their beginnings in this early layer of the phase 2 of kaumAra mantramArga (see essay on bAlagraha-s for the phases). Eventually, vinAyaka was to displace kumAra as the companion of the seven mothers. The seven mothers have been inherited by kumAra from his father agni, who right from the RV is described as a child nursed by 7 mothers (RV 1.141.2; RV 3.1.6; RV 9.102.4, RV 10.5.5). Even the RV, agni is sometimes termed guha in the sUkta-s mentioning these mothers – an epithet that was later inherited by kumAra. Interestingly, right from RV 1.191.14 these seven sister goddess are invoked along with 7 peacocks to counter toxins.
The skanda kulavidyA mantra AS uttara-sthAna 4.50-60
After performance of the above maNDala ritual and recitation of the satyasAdhana mantra, the kula-vidyA mantra is prescribed:
tatashchAnu paThedenAM kulavidyAM samAhitaH |
aulA nIlA khalA kAlA nakulA grahakArikA ||
senA susenA mAtA cha mahAsenA yashasvinI |
kR^itamAlAkSatashirA etAstu graha-mAtaraH ||
kumAra-vachanAch-ChIghraM pramu~nchantu shishuntvimam |
kAla-kalpA jvalach-ChaktiH ki~NkiNI-jAla-bhUShitA ||
kumAra-vachanAd-dUtI samprAptA muchyatAM shishuH |
patAkA kukukTaH ChatraM ghaNTA-barhadharaH shikhI ||
sharastambhash-cha te mudrA muchyatAM mAtaraH shishuH |
gandhaM dhUpaM cha mAlyaM cha kumAraste prayachChati ||
na sthAtavyaM tadAgrAya svAM gatiM gachCha mA chiraM |
dhAtryaH sarvAH kumArasya sa~NgatA gandhamAdane ||
rakta-dhvaja-patAkasya raktAMshuka-nivAsinaH |
kumArasyAlayaM tatra yAta bAlaM vimu~nchata ||
ShaNmukho dvAdashAkSho vai tasmin kAle vR^ito varaM |
vIraH kumArAdhipatistava putro.abhiShichyatAM ||
brAhmaNo brAhmaNAnAntu sarva-veda-vidAM varaH |
yakSha-rAkShasa-nAgAnAM gandharvAsurayor api ||
piShAcha-bhUta-sa~NghAnAM vyadInAM cha sa IshvaraH |
Aj~nApayati vaH skando gaMyatAM gandhamAdanaM ||
graho yaH kopayedAj~nAM kumArasya yashasvinaH |
sthAnAt chyutaH sa vai kShipraM bahya pariShado bhavet ||
A bali is offered with this mantra followed by the kindling of agni and a homa with various items (flowers, food and blood?) with 108 samidhs of khadira or chandana. This mantra known as the kula vidyA invokes the 13 dUti-s of skanda to release the child and proceed to the regions of gandhamadana under the orders of kumAra. It is perhaps one of the earliest uses of the term kula in the context of a mantra of a family or cluster of goddesses: here they are 13 graha-mAtR^i-s namely 1) aula; 2) nIlA; 3) khalA; 4) kAlA; 5) nakulA; 6) grahakArikA; 7) senA 8) susenA 9) mAtar 10) mahAsenA 11) yashasvinI 12) kR^itamAlA 13) kSatashirA. We believe that the use of the term kula for a family of devI-s is was directly transmitted from such systems of the early kaumAra mantra mArga to the early classical tantric kaula system. Evidence in this regard comes from the kaula-j~nAna-nirNaya (KJN) of the illustrious siddha of the kaliyuga, matsyendranAtha. The KJN appears precede the other great kula systems that were to emerge like those of kubjikA, kAlI, tripurA and trika. In the KJN, the kula is the family of the 64 yogini-s, who are arranged 1 per petal, on the 8 petals of the 8 chakra-s of this system. Importantly this kula of yogini-s is supposed to rove in the world in various avimorphic and theriomorphic states (KJN23.2-5ab; emended partially by me from PC Bagchi’s edition):
martye.asmin devatAnAntu saMchAraM shR^iNu bhAmini |
kapotikA tathA gR^idhrI haMsI chaiva nakhI tathA ||
kha~njaHkheTA bhASI tu kokAbhASI tu sundari |
ulukI pechakI vA tu sArasI valgulI tathA ||
shR^igAlI ajA mahiShI uShTrI mArjArI nakulI tathA |
vyAghrI hastI mayUrI cha kukuTI na …[break] ||
anyAni yAni rUpANi saMsthitAni mahItale |
These forms like that of a dove, vulture, goose, eagle, wagtail, pheasant, bustard, owl, nightjar, crane, bat, jackal, goat, buffalo, camel, cat, mongoose, tiger elephant, peacock and rooster are typical of the zoomorphic mAtR^i-s of skanda named in large numbers in the kumAropAkhyAna of the shalya parvan of the mahAbhArata. Further, it is precisely with some of these names that the forms of revatI are described in the archaic kaumAra material in the magico-medical kAshyapa tantra. Finally we find the skanda himself the center of the kulavidyA of the AS is subsumed under the name vaTuka (the boy) in the KJN. This is one of the earliest texts where the vaTuka is explicitly described as 6-headed (albeit in an ungrammatical sandhi, “aisha language”, marked below) KJN 16.52:
ahaM tvaM cha vishAlAkShi chandra-dvIpa-samAgatau |
[*ShaT-mukho] vaTuko jAtAH kShetrapAla-kulAgame ||
This is further clarified in bali mantra of vaTuka given in the KJN:
hrIM vaTukAya kapila-jaTAya pi~Ngala-netrAya devI-putrAya mAtR^i-putrAya
imAM baliM mamopanItAM gR^ihNa gR^ihNa churu muru hrIM ||
Here vaTuka is explicitly called devIputra and matR^iputra. This is again confirmed by another statement in the same chapter of the KJN where vaTuka is said to be kArttikeya (KJN 16.27):
ahaM chaiva tvayA sArddhaM chandra-dvipaM gato yadA |
tadA vaTukarUpeNa kArtikeyaH samAgataH ||
Thus, in addition to the sapta mAtR^i-s, we can state that the kaula system’s family or cluster of yogini-s also emerged from the matrix of the early kaumAra shAsana.
The rite of lustration
The bath rite performed on the child and his mother/nurse in the AS is comparable to the rohinI snAna described in the revatI kalpa of the kAshyapa tantra (more generally it follows the pattern of the bath in the AV tradition, parishiShTha 42 and the puShya bath of varAhamihira, bR^ihatsaMhitA 47). However, here several parvan days under different stars are offered: 1) revatI 2) shravaNa 3)svAtI 4) rohiNI 5)mR^igashIrSha 6) puShya 7) mUla. In the AS the authority is described as Atreya. It may be done in a cowshed, elephant shed, confluence of rivers, cross-roads, or a cemetery. Plants of different types including dUrvA grass are used. A maNDala with 3 colors and 4 doors in the form of a square is constructed and decorated with banners and spread with dUrva and barley shoots. In the four cardinal directions of the maNDala (E-S-W-N) the deities indra, yama, varuNa and kubera are inscribed along with their weapons and vAhana-s. Beyond yama’s post, the place of the mAtR^i is installed and nandikeshvara is stationed along with a bull. A mud image of vinAyaka is also installed. Beyond the station of varuNa, the 1000 spoked (sahasrAra) chakra, the weapon of viShNu is installed. Between the station of kubera and indra, shrI is installed with a lotus. The remaining aShTa-ma~Ngala symbols are then installed in order: the nandyAvarta, a~NkuSha, fish, svastika, sha~Nkha and shrIvatsa. In the interstitial directions the parivAra of indra (NE), yama (SE), varuNa (SW) and kubera (NW) are stationed. Then the lambini group of rAkShasI-s are installed in the maNDala thus: lambodarI, lambabhujA and lambakarNI are installed at the gate of indra; pralambini at the gate of yama; lambakeshI at the door of varuNa; lambanAsikA at the door of kubera. Then 6 kIla-s (stakes) of iron, 7 made of 3 other metals, and two of khadira wood are made and sprinkled with sandal water and transfixed into the ground all around the maNDala. 8, 16 or 32 pitchers of different colors marked with svastika-s are filled with pure water and various medicinal plants, and placed symmetrically at the 4 doors of the maNDala. Also added to the pots depending on the direction are gold pieces, pearls, seeds of herbs, and mustard and rice. Pure soil from various places such as temples, elephant stables, horse stables, mountain peaks, crossroads, termite mounds, river banks, courtesan’s dwellings, king’s dwelling, secret passages, sea shores should be brought along with kindled fire, samidh-s of latex-bearing plants, garlands, perfumes. Honey, ghee, rice, fried rice, five colored rice dishes and sesame readied for oblations.
The child and the mother/nurse are made to sit down in the central lotus figure of the maNDala on an audumbara or pAlasha pITha strewn with green darbha grass. Then the fire is installed in the northern quarter of the maNDala by the physician and he begins making oblations to skanda, agni and the kR^ittika-s with the mantra-s:
agnaye svAhA | kR^ittikAbhyash-cha svAhA | svAhA | namaH skandAya devAya grahAdhipataye namaH |
shirasA tvAbhivandehaM pratigR^ihNIShva me baliM |
nIrujo nirvikArashcha shishur-bhavatu sarvadA | svAhA ||
After these oblations are completed with the different oblation materials the physician clothed in white, having observed the requisite fast of 3 nights and having controlled his senses begins the lustration with the soil, herbs and the water from the pitchers. The pitchers from each door are consecrated with mantra-s to the respective deities (indra, yama, varuNa and kubera) and then oblations are made to these gods and the child is bathed. Then a series of oblations are made to prajApati, rudra, kumAra and the yoginI-s along with the calling of various kumAra graha-s, bhUta-s and other agents. The oblations to kumAra are offered with the formula:
namo bhagavate kumArAya pili pili khilli khilli khiNi khiNi svAhA ||
The yogini-s are offered oblations with the formula:
namo bhagavatIbhyo mahAyogIshvarIbhyo nimi-nimi menu-menu turu-turu svAhA ||
After the rite the child should not eat meat for 5 days and should avoid excesses for the same period. While the rite confers freedom from disease, its performance without adherence to the vidhI kills the physician and the child.
The rite of lustration as seen in the AS appears to have been adapted from the AV-parishiShTha tradition for a kaumara ritual. The similarly adapted bath, i.e. the rohiNI bath, in the revatI kalpa of the kAshyapa tantra also includes a rite for the installation of images of kumAra, ShaShThI and vishAkha made of gold, silver, root or dhUrva grass in front of the sacrificial fire. Then they are offered oblations of ghee in the fire. Given that the revatI kalpa is a linguistically ancient text it is likely that this bath ritual was adapted early in the development of the kumAra shAsana. The AV-rite was also adapted by vArahamihira for his rite of the puShya bath, but here the context is an auspicious bath for the king. In the case of the puShya bath of the bR^ihat saMhitA too skanda and vishAkha are deities among the deities stationed in the maNDala:
grahAMsh cha sarva-nakshatrai rudrAMsh ca saha mAtR^ibhiH |
skandaM viShNuM vishAkhaM cha lokapAlAn surastriyaH ||
However, beyond that there is not specific rite involving kumAra, and several other deities receive oblations down the line. Hence, the version of varAhamihira most probably followed a trajectory independent of the kaumAra system.
The mention of the yoginI-s in the AS version and mAta~NgI in the kAshyapa saMhitA do add additional support to the old link between the kaumAra system and the goddess who were to become elements of the kaula systems. But, another interesting point seen in the snapana ritual of the AS appears to have been linked to the development of kumAra in the nAstIka tradition as vajrakumAra. In the AS rite 15 kIla-s or stakes are transfixed into the ground. In the nAstIka tradition vajrakumAra is often known by an alternative name as vajrakIla and is depicted as a stake or a vajra combined with a stake. The reason for this connection between vajrakumAra, who is otherwise a rather straight forward nAstIka adaptation of kumAra, and the stake was previously largely mysterious to most students of these matters. However, in light of the distinctive use of kIla-s in the above maNDala it is quite likely that vajrakumAra’s alias vajrakIla is a remnant of this ancient feature of a kaumAra ritual.