The Amrtasiddhi: Hathayoga’s tantric Buddhist source text
The Amrtasiddhi: Hathayoga’s tantric Buddhist source text
James Mallinson SOAS,University of London jm@soas.ac.uk Like many of the contributors to this volume,I had the great fortune to have Professor Sanderson as the supervisor of my doctoral thesis, which was a critical edition of an early text on hathayoga called the Khecarīvidyā. At the outset of my work on the text, and for several subsequent years, I expected that Professor Sanderson's encyclopedic knowledge of the Śaiva corpus would enable us to find within it forerunners of khecarī mudrā, the hathayogic practice central to the Khecarīvidyā. However, not withstanding a handful of instances of teachings on similar techniques, the fully-fledged practice does not appear to be taught in earlier Śaiva works. In subsequent years,as I read more broadly in the corpus of early texts on hathayoga (which, in comparison to the vast Śaiva corpus, is very small and thus may easily be read by one individual), I came to the realisation that almost all of
thepracticeswhichdistinguishha.thayogafromothermethodsofyogawereuniquetoitat the time of their codification and are not to be found in the corpus of earlier Śaiva texts, despite repeated assertions in secondary literature that ha.thayoga was a development from Śaivism(or“tantra”more broadly conceived). The texts of the hathayoga corpus do,however,couch their teachings in tantric language.
The name of the hathayogic khecarī mudrā,for example,is also that ofa nearlier but different Śaiva practice. When I was invited to speak at the symposium in Professor Sanderson's honour held in Toronto in ,I decided to try to articulate my rather in choatet houghts on this subject by presenting a paper entitled“ Hathayoga’s Śaiva Idiom”. The inadequacy of myt heories wasbrought home to me some months after the symposium when Istarted to ∗This is a draft of an article to be published in a fest schrift for Professor Alexis Sanderson containing contributions from
participants at a symposium held in his honour in Toronto in March The festchrift is being edited by Dominic Goodall, Shaman Hatley and Harunaga Isaacson, and will be published by Brill. I thank Alexis Sanderson, Péter-Dániel Szántó and Jason Birchfor reading various drafted itions of the Am.rtasiddhi with me between and and Kurtis Schaeffer and Leonard van der Kuijp for sharing with me photographs of printouts
from a microfilm copy of witness C. Professor Schaeffer also kindly shared his draft edition of the Tibetan translation of the Amrtasiddhi given in witness C. I thank Viswanath Gupta for his help transcribing the Grantha manuscript M. My recent intensive study of the Amrtasiddhi has been made possible by a Consolidator Grant from the European Research Council to fund the five-year Hatha Yoga Project, on which see http://hyp.soas.ac.uk.
read, together with two other ex-students of Professor Sanderson, Dr Péter-Dániel Szántó and Dr Jason Birch, a th-century manuscript of the earliest text to teach many of the key principles and practices of hathayoga,the Amrtasiddhi. I had already read much of the text with Professor Sanderson and others, but only from later manuscript sources. As we read the older manuscript it gradually became clear that the Amrtasiddhi was composed in a Vajrayāna Buddhist milieu. Thus my notion of hathayoga having a Śaiva idiom needed readdressing. One might perhaps talk instead of its “tantric idiom”. But I shall leave reflections on that topic for a later date and in this short paper focus on the Amrtasiddhi and,in particular,the features of it which make it clear that it was composed ina Vajrayāna milieu. I am currently preparing a critical edition and annotated translation of the text with Dr Szántó; what follows here results from our work in progress. Despite our edition being incomplete, I am confident that the conclusion drawn here about the origins of the text is sound (and that further work on the text will provide additional and complementary evidence) and I think it important enough to warrant preliminary publication. Subsequent publications will address this unique text’s many other remarkable features.
The Amrtasiddhi The importance of the Am.rtasiddhi was first brought to scholarly attention by Professor Kurtis Schaeffer in an article published in Here I shall reprise as little of his rich and dense article as is necessary to provide the background to what follows. Schaeffer focuses on the twelfth-century manuscript of the text, photographs of printouts from a
microfilm of which he and Professor Leonard vander Kuijp have kindly shared with me. At the time that the microfilm was made, the manuscript was in Beijing, although Professor Schaeffer believes that it has since been returned to Tibet. The manuscript is unique in that it is bilingual, with three registers: the Sanskrit text in a Nepali or east Indian script, a transliteration of the Sanskrit in Tibetan hand-printing script and a translation into Tibetan in theT ibetan cursives cript. This manuscript is referred to in what follows by the siglum C. The other witnesses
of the text which have been collated are considerably later than C (the oldest is perhaps thec.th-century K). They present versions of the text in which redaction has removed or obscured some of the Buddhist features evident in C. These witnesses may be divided into two groups. The first is a single Grantha manuscript from the Mysore Government We were joined at our reading sessions by Sam Grimes, Diwakar Acharya, Camillo Formigatti, Anand Venkatkrishnan and Paul Gerstmayr, whom I thank for their valuable comments. Prior to Schaeffer‘s article, the only mention of the text of which I am aware (other than in manuscript catalogues)is Gode ,in which its citations in the Yoga cintāmani are noted. Schaeffer says that the manuscript’s colophon gives a date which“may read ”There ading is clear:
ekāśītijute[juteisNewarscribaldialectforSanskrit°yute]śākesahāsraiketuphālgune|k. r. s. nā. s. tamyā. m samāpto 'ya.m krtvāmrtasiddhir mayā || (f.v). The eighth day of the dark fortnight of the lunar month of Phālguna in Śāka corresponds to March nd CE (according to the calculator at http://www.cc.kyotosu.ac.jp/ yanom/pancanga/). The mansucript’s Tibetan colophon says that the Tibetan translation is that of the “monk of the Bya [[[Wikipedia:clan|clan]]]” (Bya ban de) Pad ma ’od zer, who worked towards the end of the eleventh century, which provides us with a nearlier terminusantequem for the text than the date of the manuscript itself.
As note in the manuscript’s Tibetan colophon,the translation is of a different recension of the Sanskrit text from that given in the manuscript. At some places, e.g. and the translation corresponds to the text as found in the other witnesses,but not that inC.
Oriental Library (M), the second seven north Indian and Nepali manuscripts, two from Jodhpur’s Maharaja Man Singh Pustak Prakash (J and J = J) and five from the Nepal German Manuscript Preservation Project The text of the Amrtasiddhi consists of verses divided into short vivekas. The first tenvivekas teach the constituents of the yogic body. Vivekas teach three methods of manipulating those constituents (mahāmudrā, mahābandhaandmahāvedha)andviveka teaches the practice (abhyāsa),i.e.how the three methods are to be used together. Vivekas
teach the four grades of aspirant, the four states (avasthās) of yoga,and the final transformation of the body leading up to nirvāna. The Amrtasiddhi in the Hathayoga tradition Citations and Borrowings
The Amrtasiddhi is a seminal work in the hathayoga textual tradition. chaeffer mentions its citations in the Yogacintāmani and Hathapradīpikājyotsnā In addition, several hathayoga texts borrow directly from the Amrtasiddhi without attribution. The century Goraksaśataka shares three half-verses with the Amrtasiddhi. The Vivekamārta. n.da,which is also likely to date to the thcenturyredacts four of the Amrtasiddhi’s verses into three. The c. th-century Amaraughaprabodha shares six verses with the Amrtasiddhi and paraphrases it extensively elsewhere. The Goraksayogaśāstra ( the century or earlier)borrows two and a half verses and extensively paraphrases other parts of the text. The Śivasamhitā is much the biggest borrower from the Amrtasiddhi,sharingverses with it. The Hathapradīpikāshares five half-verses with the Amrtasiddhi, but these may be borrowed from the Amaraughaprabodha since all the shared passages are also in that text.
Full details of the witnesses are given at the end of this article. There are vivekas in the Beijing ms and in the others. All verse numbering given here corresponds to the order of verses inC(which does not itself give verse numbers). Vivekas are interspersed with very short chapters on a variety of topics. In the first viveka there is a list of the topics to be taught in the text. The list corresponds exactly to the vivekasuptoviveka,but then goes awry. More analysis is needed to be sure, but it seems likely that at least some of the vivekas after are later additions to the text. Despite the compound hathayoga being found in earlier Vajrayāna works (Birch and its teachings being central to later hathayogic texts, the Amrtasiddhi does not call its yoga method hatha. This paradox will be addressed in subsequent publications. Yoga cintāmani (with differences),Ha.thapradīpikājyotsnā ad This verse is also found at Gherandasamhitā
Doctrinal Innovations SeveraloftheAm.rtasiddhi’steachingshavenopriorattestationandarecentraltoteachings on ha.thayoga in later texts, where they are either reproduced verbatim, as noted above, or incorporated in to new compositions. These maybe summarised as follows. The Yogic Body (a) The Amrtasiddhi is the first text to relocate to the body the ancient triad of sun, moon and fire. The idea of a moon in the skull dripping amrta is found in many earlier tantric works,but that of the sun in the stomach consuming it is new,as is the conflation of the sun and fire. i. The Moon meruś. rngesthitaścandrodvira. s. takalayāyuta. h| aharniśa.mtu.sārābhā.msudhā.mvar.satyadhomukha. “ThemoonisonthepeakofMeruandhassixteendigits. Facingdownwards,itrainsdewynectardayandnight.” ii. TheSun madhyamāmūlasa.msthāneti. s. thatisūryama. n.dala. h| kalādvādaśasa.mpūr.nodīpyamāna.hsvaraśmibhi. ūrdhva.mvahatidak.se.natīk. s. namūrti.hprajāpati. h| vyāpnotisakala.mdeha.mnā.dyākāśapathāśrita. grasaticandraniryāsa.mbhramativāyuma. n.dale | dahatisarvadhātū.mścasūrya.hsarvaśarīrake °pathāśrita.h] ;yathāśrita . m K,yathāśrita. hcett. b °ma. n.dale]conj.Szántó;°ma. n.dalai. h C,°ma. n.dala . mcett. “ The sphere of the sun is at the base of the Central Channel, complete with twelve digits, shining with its rays. () The lord of creatures (Prajāpati), of intense appearance, travels upwards on the right. Staying in the pathways in the spaces (ākāśapatha) in the channels it pervades the entire body. The sun consumes the lunar secretion,wanders in the sphere of the wind and burns up all the bodily constituents in all bodies.”
kalābhirdaśabhiryukta.hsūryama. n.dalamadhyastha. h| vasativastideśecavahnirannavipācaka. yovaivahni.hsavaisūryoya.hsūrya.hsahutāśana. h| etāvekātmakaud. r. s. tausūk.smabhedenabheditau|| madhyastha.h]conj.;°madhyata. madhyaga. hY,°madhyata cvasativastideśe]conj.;vasativatideśeC,vasatevastideśeY,vasatirasthideśecett. ekātmakau]conj.Sanderson;ekataraucodd. bheditau] C;bhedinaucett. Thisisaśle.sa: ākāśapatha canal some an the suns orbit int he sky.
“(Endowed with ten digits, in the middle of the sphere of the sun in the region of the stomach dwells fire, which digests food. () Fire is the sun;the sun is fire. The two look the same [but] differ subtly.”
(b) The use of the word bindu for semen,its identification withth eam.rtadripping from the moon,its preservation being essential for life and its division in to male and female are alli nnovations of the Amrtasiddhi which are widelya dopted in later hathayoga texts. i. adhaścandrām.rta.myātitadām.rtyurn. r. nā.mbhavet | yāti] K;yati C,°m.rta .myasya J “The nectar of immortality in the moon goes downwards;as a result men die.” ii. bindu pātenav.rddhatva.mm.rtyurbhavatidehinām| “The fall of bindu makes men grow old[and]die.” iii. sabindurdvividhojñeya.hpauru.sovanitābhava. h| bīja.mcapauru.sa.mprokta.mrajaścastrīsamudbhavam|| anayorbāhyayogenas. r. s. ti.hsa.mjāyaten. r. nām| yadābhyantaratoyogastadāyogītigīyate | kāmarūpevasedbindu.hkū.tāgārasyako.tare | pūr.nagiri.mmudāsparśādvrajatimadhyamāpathe || yonimadhyemahāk.setrejavāsindūrasannibham| rajovasatijantūnā.mdevītattvasamāśritam| binduścandramayojñeyoraja.hsūryamayastathā | anayo.hsa.mgama.hsādhya.hkū.tāgāre’tidurgha.te | cdyadābhyantaratoyogastadāyogītigīyate]CHPJ;yadātvabhyantare yogastadāyogohibha.nyatecett. a kāmarūpe] C;kāmarūpo Jbkū.tāgārasya°]em.;kutāgārasyaC;kūtādhāra.nyaJ kū.tādhārasya dpūr.nagiri .m]C(Tibetantranscription);pūr.nagiricett. • mudā]C; sadā J vrajati] C;rājanticett. samāśritam]conj.Sanderson;samādh.rtamC,samāv.rtaK samāv.rta. h cett. kū.tāgāre]ku.tāgāre C “Know bindu to be of two kinds,male and female. Semen(bīja)is said to be themale[ bindu]andrajas(female generative fluid) is female. ()As a result of their external union people are created. When they are united internally, then one is declared a yogi. () Bindu resides in Kāmarūpa in the hollow of the multi-storied palace (kū.tāgārasya).As a result of touch, with delight it goes to Pūr.nagiri along the Central Channel. Rajas resides in th great sacred field in the perineal region(yonimadhye). On the kū.tāgāra,seebelow,p..
It is as red as China rose and takes shelter in the Goddess element (devītattvasamāśritam). Know bindu to be made of the moon and rajas to be made of the sun. Their union is to be brought about in the very in accessible multi-storeyedp alace.” (c) A connection between the mind and breath is taught as early as the Chāndogya Upanisad . The Amrtasiddhi is the first text to teach that mind, breath and bindu are connected,a notion found in many subsequent hathayoga texts. calatyaya.myadāvāyustadābinduścala.hsm.rta. h| binduścalatiyasyāya.mcitta.mtasyaivacañcalam||calaty aya .m yadā ] C; yadāyan calate M, yadā ca .mcalate JK, calaty e.sa yadā YHJ b cala.hsm.rta.h] JKYHJ;cala.hsm. r⌈ta⌉. h C,cacañcala. h M cbinduścalatiyasyāya .m]C;yasyāyancalatebindu. hM,yasyāya .mcalate binduś JK,binduścalatiyasyā˙nge YHJ d tasyaiva] CKYHJ;tasthyai∗M,tathaiva J “It is taught that when the breath moves bindu moves; the mind of he whose bindu is moving is restless.”
(d) The three gran this. The Amrtasiddhi’s system of three granth is,brahma°,vi. s. nu° and rudra°,which are situated along the central channel of the body and are to be pierced by the mahāvedha (),is very common insubsequent hathayoga texts.. The three practices,mahāmudrā,mahābandha,mahāvedha (vivekas). These practices,which involve bodily postures and breath control,are used to make the breath enter the central channel and rise upwards. They are an innovation of the Amrtasiddhi and are taught in all subsequent hathayoga texts,albeit sometimes with different names. . The four avasthās The four avasthās, “states” or “stages” of yoga practice (ārambha, ghata, paricaya, nispanna/nispatti) taught in the Amrtasiddhi (vivekas ), are taught in many Sanskrit hathayoga texts;they are also mentioned in the old Hindi Gorakhbā.nī(śabds ). In addition to these innovations, in viveka (abhyāsa, “practice”) the Amrtasiddhi describes, at a level of detail unparalleled in other texts, the internal processes brought about by its methods,in particular the movement of the breaths. Granthis are mentioned in many earlier Śaiva texts, some of whose lists include brahma, vi. s. nu and rudra granthis but not in the Amrtasiddhi’s configuration. See e.g. Kubjikāmatatantra , in which there are sixteeng ranth is nd Netratantra ,in which the reare twelve.
Buddhist features of the Amrtasiddhi In Schaeffer’s analys is of the Amrtasiddhi ,he notes how it is unique amongst Tibetan Buddhist works because its teachings are said to bestow jīvanmukti, “liberation whilel iving”,and make the yogi identical with Śiva. Despite the se Śaiva features,however, close reading of manuscript C, the th-century bilingual witness of the text, shows that the text was composed with in a Vajrayāna milieu. Furthermore,it pitsits teachings against those of other Vajrayāna schools,not Śaiva ones. As can be seen in the examples given below, manuscript C generally has the best readings of the text and presents its Buddhist teachings intact. In the other manuscripts the specifically Buddhist doctrines found in C are either unwittingly included,misunderstood (and sometimes presented in corrupt forms as a result)orde liberately changed or omitted.
Someofthetext’sBuddhistfeaturesareambiguousorobscureenoughforthemtohave beenpreservedbytheredactorsofthetextaspresentedinthelaterwitnesses. Thus we find multiple examples of Vajrayāna (or more broadly Buddhist) terminology such as mahāmudrā (viveka vajrapañjara jñānasambhāra śūnya nispanna (and abhiseka Similarly ,Amrtasiddhi mentions the very specifically Vajrayāna notion of thefour blisses: ānandāyeprakathyanteviramāntā.hśarīrata. h| te'pibindūdbhavā.hsarvejyotsnācandrabhavāyathā |āntā.h]°ā .mtā JK “The[four bodily blisses whose last is[the bliss of] cessation all arise from bindu,just as moonlight arises from the moon.”
Other Buddhist features of the text as found in C are deliberately omitted or altered in the later witnesses. Examples of these are listed below. This list is not exhaustive; furt her closer eading of the text is likely to reveal more examples.
Chinnamastā Manuscript C opens with a sragdharā ma˙ngala verse in praise of the goddess Chinnamastā: nābhauśubhrāravinda.mtaduparivimala.mma. n.dala.mca. n.daraśme. h sa.msārasyaikasārātribhuvanajananīdharmavartmodayāyā | tasminmadhyetrimārgetritayatanudharāchinnamastāpraśastā tā.mvandejñānarūpā.mmara.nabhayaharā.myoginī.myogamudrām|| a śubhrā°] C;candrā vimala .m] C;vivara . tasmin tasyā . m • tri° ] ; tre° C • chinnamastā praśastā ] C;cittahasthā . mpraśastā . m M d tā .mvandejñānarūpā .m] C;vandejñānasvarūpā . m M OnthefourblissesseeIsaacson&Sferra passim.
“At the navel is a white lotus. On top of that is the spotless orb of the sun. In the middle of that, at the triple pathway, is she who is the sole essence of samsara [and] the creator of the three worlds, who arises on the path of dharma, who has three bodies [and] who is lauded as Chinnamastā, ‘she whose head is cut’. I worship her, she who has the form of knowledge, who removes the danger of death, the yoginī, the seal of yoga.”
Until the t century , Chinnamastā is not mentioned in non-Buddhist texts (Bühnemann ). Her Vajrayāna origins have been demonstrated by Sanderson , who notes how the epithet dharmodayā, found in the Am.rtasiddhi as dharmavartmodayā, is “strictly Buddhist”. One might argue that this mangala verse could be an addition to the text when it was redacted by a Vajrayāna tradition, but the verse is also found in the Granth a manuscriptM in a corrupt form. Chinnamastā'sname is given there in as Cittahasthā,but the epithetsd harmavartmodayā and tritayatanudharā are preserved. The Rajasthani and Nepali manuscripts omit this verse. . chandoha AtAm.rtasiddhi manuscript C uses the specifically Buddhist term chandoha: sāgarā.hsaritastatrak.setrā.nik.setrapālakā. h| chandohā.hpu.nyatīrthānipī.thānipī.thadevatā. h|chandohā.h]em.;chandohā C,sa mbhedā. h MJK “There are oceans, rivers,regions[and] guardians of the regions; gathering places (chandohā.h), sacred sites, seats [of deities and] he deities of the seats” In Śaiva texts chandoha is found as sa.mdoha. That the manuscripts other than C read sa.mbhedā. h, which makes no sense, suggests that they may derive from an archetype that had sa.mdohā. h,which subsequent copy ist did not understand. The foure lements Amrtasiddhi refers to four physical elements: p.rthivyādīnicatvārividh.rtānip.rthakp.rthak catvāri] C;tattvānicett. “The four [[[elements]] ]earth etc. are kept separate [by the breath].”
In Śaiva and other Hindu traditions there are five primary physical elements. The later manuscripts therefore change catvāri,“four”,totattvāni,“elements”. Sanderson Sandersonloc.cit.: “Thiss ubstitution of initialch-fors-/ś-is probably an east-Indianism”.
kū.tāgāra This is a very common termin Vajrayāna works,denoting a“multi-storeyed palace” in the middle of a mandala (Reigle It is not found in Śaiva texts and is not recognised by the later north Indian and Nepali witnesses of the Amrtasiddhi. abkāmarūpevasedbindu.hkū.tāgārasyako.tare | rūpe] CM;°rūpo kū.tāgārasya] C,∗ū.tāgārasya kū.tādhāra.nya° J,kū.tādhārasya K “Bindu resides at Kāmarūpa, in the hollow of the multi-storeyed palace.” trivajra in C mentions the three vajras, i.e. the common Vajrayāna triad of kāya, vāk and citta. In the other witnesses trivajrā.nā. misfound as trivargā.nā. m. trivajrā.nā.msamāveśastadāvaijāyatedhruvam|| trivajrā.nā .m] C;trivargā.nā . “The absorption in to the three vajras is sure to arise.’
trikāya A reference to the Buddhist notion of the triple body is expunged in the later witnesses: sarvajñatva.mtrikāyasyasarvajñānāvabodhakam| lak.sa.na.msiddhacittasyajñātavya.mjñānaśālibhi. kāyasya] C;°kālasya °kāryasya Jbodhakam] bodhanam JK siddhacittasya] C;siddhivittasya JK “Omniscience, which brings about complete understanding of the triple body,should be known by the knowledgablet obethemarkofhe whose mind has been mastered.”
buddha Verses in which C reads(or its archetype isl ikely to have read) buddha are reworked in the later witnesses. bindur buddha.hśivobindurbindurvi. s. nu.hprajāpati. h| bindu.hsarvagatodevobindustrailokyadarpa.na. h|a buddha.h]em.;v.rddha. h C,ūrdhva. hcett. “Bindu is Buddha,bindu is Śiva,bindu is Vishnu,the lord of creatures, bindu is the omnipresent god,bindu is the mirror of the three worlds.” The Mahāmudrātilaka (draft edition of Staatsbibliothek zu Berlin Preuss. Kulturbesitz Orientabteilung locatesthebodilyKāmarūpabetweentheeyebrows.
abtāvadbuddho'pyasiddho'saunara.hsā.msārikomata. h| buddho] C;∗ddho dvo ,siddho vaddho K “Even a Buddha, as long as [he remains] unperfected [by means of the practice taught in the Am.rtasiddhi],is considered a worldly man.” svādhi. s. thānayoga In two places the Am.rtasiddhi mentions svādhi. s. thāna yoga. This is a method of visualising oneself as a deity which is central to the teachings of a wide variety of Vajrayāna texts (e.g. Guhyasamāja where it is called svādhidaivatayoga, and the Pañcakrama, whose third krama is called the svādhi. s. thānakrama). In the two verses from the Am.rtasiddhi given below, the methods of svādhi. s. thāna yoga are said to be ineffective;to achieve the goals of yoga one must use the practice taught in the Am.rtasiddhi. The later witnesses of the text do not understand the phrase svādhi. s. thānena yogenaand,presumably surmisingsvādhi. s. thāna to refer to the second of the six cakras in the system taught in many hathayoga texts (but not in the Amrtasiddhi, which makes no mention of cakras),they change yogenatomārge.nainaa attempt to make the phrase refer to a path way in the yogic body. svādhi. s. thānenayogenayasyacitta.mprasādhyate | śilā.mcarvatimohenat. r.sita.hkha.mpibatyapi || yogena ] C; mārge.na yasya ] JK; yastuś C, yatna prasādhyate] MJK;prasādhyati C
“He who tries to master his mind by means of self-established yoga deludedlychews arockand,thirsty,drinks the s ky.” svādhi. s. thānenayogenanak.sīyetegu.naun. r. nām| astimudrāviśe.se.nagurumukhābjasa.mbhavā | ayogena]C;mārge.naM bnak.sīyete]em.;nak.sīyateC,prak.sīyante nāk.sipeti JK • gu.nā ] gu.nau C, gu.no viśe.se.na ] viśe.sādvā guru°] CJK;guror °mukhābja°] C;°vaktrābja °mukhāttu JK • °sa .mbhavā] JK;°sa .mbhavā . m C,°sa .mbhavāt M “Men’stwo[unwanted]gu.nas[,rajasandtamas,]arenotdestroyedbyself-established yoga. Thereisamudrāespecially[forthat],bornfromthelotus-mouthoftheguru.”
Conclusion The Am.rtasiddhi was composed in a Vajrayāna Buddhist milieu and its intended audience was other Vajrayāna Buddhists. Its teachings are subsequently found in ha.thayoga texts from a wide range of non-Buddhist traditions. This does not mean, however, that hathayoga itself was product of Vajrayāna Buddhists. I have argued else where (e.g. Mallinson that the practices of hathayoga were current among ascetics long before their codification. The Am.rtasiddhi was the first text to codifym any of ha.thayoga’s distinctive techniques and was thus the first to assign names to them. As a result the Dattātreyayogaśāstra,the first text to teach the practices of hathayoga under the name hatha, includes among its techniques the Am.rtasiddhi’s mahāmudrā, mahābandha and mahāvedha (with slight variations in their methods). In addition to these physical techniques, the Dattātreyayogaśāstra also adoptsfromtheAm.rtasiddhithemoretheoreticaldoctrineofthefouravasthāsorstagesof yoga,showing that the Am.rtasiddhi’s influence was more than simply terminological Because they share traditions of siddhas, several scholars have posited connections between Vajrayāna Buddhists and the Nāth yogis,with whom the practice of hathayoga has long been associated. The Am.rtasiddhi’s Vajrayāna origins and its borrowings in subsequent hathayoga texts,some of which are products of Nāth traditions,provide the first known doctrinal basis for this connection and a stimulus for its further investigation.
Witnesses of the Am.rtasiddhi Manuscripts collated • (C) China Nationalities Library of the Cultural Palace of Nationalities MS No. Paper. Nepali and Tibetan(hand-print and cursive). • Maharaja Man Singh Pustak Prakash,Jodhpur Paper. Devanāgarī. Paper. Devanāgarī. • Nepal-German Manuscript Preservation Project. All entitled Am.rtasiddhi. Paper. Devanāgarī. Paper. Devanāgarī. Paper. Newari. Paper. Devanāgarī. Paper. Newari. The Dattātreyayogaśāstra isaVai. s. navawork but attributes the teaching of the mahāmudrā to Bhairava(verse Among the other ha.thayoga practices taught in the Dattātreyayogaśāstra is vajrolīmudrā, a technique of urethral suction said to bring about mastery of bindu (Mallinson forthcoming). Its name is clearly redolent of Vajrayāna Buddhism; the example of the Amrtasiddhi suggests that it was in a Vajrayāna text that the practice of vajrolī was first codified too. Although such usage is not found in pre-modern texts, to avoid confusion I use the word “Nāth” to refer to ascetics usually called yogīs or jogīs in texts and travellers’ reports and whose traditions, with some exceptions such as those which trace their lineages to KānhapaorK. s. nācārya,came,by the sixteenth century at the latest,to be grouped together in twelve panthsor lineages. On the NāthSa .mpradāya,see Mallinson An early and influential linking of the Buddhist and Nāth siddha traditions was made by Sastri (, first edition Material evidence dating from the thto th centuries at two Konkan sites,Kadri in Karnataka (Bouillier and Panhale Kajiin Maharashtra (Deshpande also points to a connection between Vajrayāna Buddhists and Nāths, as well as to a possible location of the composition of the Amrtasiddhi. Virūpa or Virūpāk.sa, to whom the text’s teachings are ascribed in its mansucript colophons, is named in all Tibetan lists of siddhas and included in two early groupings of nine Nāths from the Deccan and Konkan,Gaura.na’s th century Telugu Navanāthacaritraandthec. century Kadalīmañjunāthamāhātmya(but not in any subsequent groupings of Nāths), and, writing in the early century, the Tibetan Tāranātha says that Virūpa lived in Mahrata,i.e. Maharashtra(Templeman
(M)My sore Government Oriental Manuscripts LibraryD . Palm leaf. Grantha.
Other collated witnesses • (Y) Yogacintāma.ni ed. HaridāsŚarmā,Calcutta Oriental Press,n.d. • (HJ)Ha.thapradīpikājyotsnāof Brahmānanda,ālocanātmaksa.mskara. n(Hindī),ed.Svāmī Maheśānand, Dr Bāburām Śarmā, Jñānaśa .mkar Sahāy, Ravindranāth Bodhe. Lonavla: KaivalyadhāmS.M.Y.M.Samiti.
Manuscripts not yet collated My sore Government Oriental Library D Paper. Grantha. Mys ore Government Oriental Library R- Palm leaf. Grantha. Incomplete. Adyar Library Palmleaf. Grantha. Baroda Oriental Institute (b). Palmleaf. Grantha.
Works Cited Primary Sources Amaraughaprabodha,of Gorak.sanātha,ed.K.Mallikin The Siddha Siddhānta Paddhati and Other Works of Nath Yogis. Poona: Poona Oriental Book House. Kadalīmañjunāthamāhātmyam, ed. Śambhu Śarmā Ka.dava. Kāśī: Gorak.sa .Tilla Yoga Pracāri.nī. Kubjikāmatatantra, Kulālikāmnāya version, ed. T. Goudriaan and J.A. Schoterman. Leiden: E.J.Brill. Guhyasamāja tantra , ed.YukeiMatsunaga. Osaka: Toho Shuppan. Gorak.sayogaśāstra. National Archives of Kathmandu (= Nepal-German Manuscript Preservation Project . Gorak.saśataka. Government Oriental Manuscripts Library,Madras,MSNo.R Gorakhbānī,ed.P.D.Ba.dathvāl. Prayāg: Hindī Sāhity Sammelan. Ghera. n.dasa.mhitā,ed.andtr.J.Mallinson. NewYork: YogaVidya.com. Chāndogya Upanisad in The early Upani.ads: annotated text and translation,ed.andtr. Patrick Olivelle. New York,Oxford: Oxford University Press. Dattātreyayogaśāstra Dattātreyayogaśāstra. Unpublished edition by James Mallinson. Navanāthacharitra of Gaura.a, ed. K.Ramakrishnaiya. Madras University Telegu Series Madras. This edition was read with Professor AlexisS anderson, Jaso nBirch,Péter-Dániel Szántó and Andrea Acriin Oxford in early ,all of whom thank for their valuable emendations and suggestions.
Netratantra with commentary (Uddyota)byK.semarāja,ed.Madhusūdan Kaul Śāstrī. KSTS Srinagar. . Pañcakrama in Pañcakrama: Sanskrit and Tibetan Texts Critically Edited with Verse Index and Facsimile Edition of the Sanskrit Manuscripts, ed. Katsumi Mimaki and Toru Tomabechi. Tokyo: The Centre for East Asian Cultural Studies for Unesco. Yogacintāmani of Śivānandasarasvatī, ed. Haridās Śarmā. Calcutta: Calcutta Oriental Press. No date of publication. Vivekamārta. n.da of Gorak.sadeva. Oriental Institute of Baroda Library. Acc.No.. Ha.thapradīpikā of Svātmārāma, ed. Svāmī Digambarjī and Dr Pītambar Jhā. Lonavla: Kaivalyadhām S.M.Y.M.Samiti. . Ha.thapradīpikājyotsnā of Brahmānanda, ālocanātmak sa.mskara. n (Hindī), ed. Svāmī Maheśānand,DrBāburāmŚarmā,Jñānaśa .mkarSahāy,RavindranāthBodhe. Lonavla: KaivalyadhāmS.M.Y.M.Samiti.
Secondary Literature Birch, Jason. . “The Meaning of ha.tha in early Ha.thayoga”, pp. in Journal of the American Oriental Society Bühnemann,Gudrun. The Iconography of Hindu Tantric Deities volume I. The Pantheon of the Mantramahodadhi. Groningen: EgbertForsten. Gode, P.K. Studies in Indian Literary History Vol. II. Bombay: Bharatiya Vidya Bhavan. Isaacson, Harunaga & Sferra, Francesco. The Sekanirdeśa of Maitreyanātha (Advayavajra) with the Sekanirdeśapañjikā of Rāmapāla. Critical Edition of the Sanskrit and Tibetan Texts with English Translation and Reproductions of the MSS. (With contributions byKlaus-Dieter Mathesand Marco Passavanti). Serie Orientale Roma fondatada Giuseppe TucciVol.CVII.Napoli:
Universitadegli Studidi Napoli“L’Orientale”. Mallinson,James. “NāthSa .mpradāya,”entry in the Brill Encyclopedia of Hinduism, Vol.,ed.KnutA.Jacobsen,pp. Leiden: Brill. “Śāktism and Ha.thayoga,” pp.in Goddess Traditions in Tantric Hinduism,ed. Bjarne Wernicke Olesen. London: Routledge. Forthcoming. “Yoga and Sex: What is the Purpose of Vajrolīmudrā?” in Yoga in Transformation,ed. Philipp Maas. Vienna: V&Runipress. Reigle,David. “The Kālacakra Tantra on the Sādhana and Mandala”,pp.in Journal of the Royal Asiatic Society Series Sanderson, Alexis. . “The Śaiva Age”, pp. in Genesis and Development of Tantrism,ed.Shingo Einoo. Tokyo: Institute of Oriental Culture, University of Tokyo. Sastri, V. V. Ramana. . “The Doctrinal Culture and Tradition of the Siddhas”, pp. in Haridas Bhattacharyya (ed.). The Cultural Heritage of India. Vol. : The Religions. Calcutta: The Ramakrishna Mission Institute of Culture.
Schaeffer, Kurtis “The Attainment of Immortality: from Nāthas in India to Buddhists inT ibet”,Journal of Indian Philosophy vol.No. Netherlands: Springer.
