Amṛtasiddhi A Posteriori: An Exploratory Study on the Possible Impact of the Amṛtasiddhi on the Subsequent Sanskritic Vajrayāna Tradition

Abstract

Recent research into source materials for haṭhayoga (Birch, Mallinson, Szántó) has revealed that the physical techniques and esoteric anatomy traditionally associated with Śaiva practitioners likely found a genesis within Vajrayāna Buddhist communities. The physiology and practices for longevity described in the 11th-or-12th-century Amṛtasiddhi are easily traced in the development of subsequent physical yoga, but prior to the discovery of the text’s Buddhist origin, analogues to a haṭhayoga esoteric anatomy found in Vajrayāna sources have been regarded as coincidental. This paper considers both the possibility that the Amṛtasiddhi, or a tradition related to it, had a lasting impact on practices detailed in subsequent tantric Buddhist texts and that this haṭhayoga source text can aid in interpreting unclear passages in these texts.

1. Introduction

This paper will outline an investigation of the possible impact of the haṭhayoga “source text” upon two late Vajrayāna mūla tantras, namely, the 12th-century Saṃvarodayatantra (SUT) and the 13th-century Caṇḍamahāroṣaṇatantra (CMT). The haṭhayoga source text is the 11th-century Amṛtasiddhi. This paper will also note some precedents of and parallels with the physiological model found in the Amṛtasiddhi in earlier and contemporaneous Vajrayāna texts.

I was first exposed to the Amṛtasiddhi in 2016 when James Mallinson regularly came to Oxford to work through it with Péter-Dániel Szántó and I joined in the readings. It was almost immediately evident that the text had Buddhist influences (for example, its opening maṅgala verse is an invocation to the Buddhist goddess Chinnamastā), and Mallinson subsequently released an article1 detailing some of these Buddhist elements. Mallinson and Szántó will publish a full edition of the text and translation in 2020.

The text first received treatment in contemporary scholarship in Kurtis Schaeffer’s pioneering article2 in which he pointed out that it is difficult to identify the text as Buddhist, Nātha, or anything else. Schaeffer also highlighted the importance of the text in Tibet. Mallinson especially,3 but Jason Birch and others in the last two decades have illustrated the impact of the text upon the development of haṭhayoga. The goal of this paper is to begin to investigate the impact the Amṛtasiddhi may have had on subsequent developments in Sanskritic Buddhism.

For the purposes of this paper there are two key ideas in the Amṛtasiddhi I will discuss, which, when combined, would eventually become the fundamental physiological theory of haṭhayoga. The first idea is that if one is able to preserve the store of bindu, or amṛta, that is, semen, in the cranial vault, that one may extend one’s own life. This bindu drips, drop by drop, over a lifetime, and once it is exhausted, the individual dies. In females it is present as well, although as female sexual fluid or menstrual blood, not seminal fluid. The second idea is that through prāṇāyāma and physical manipulation of the body, a yogin can force wind through the nāḍīs, the subtle channels of the body, into the central channel and upwards, and reverse the downward course of the ever-dripping bindu for the specific purpose of prolonging one’s own life. This is the amṛtasiddhi, the accomplishment of immortality.

2. Precedents in Vajrayāna

There is a long, well-documented history of alchemy and practices for longevity in Vajrayāna prior to the Amṛtasiddhi,4 but the text’s particular model does not appear to exist in tantric Buddhism, at least not in any of the materials that we are aware known of. That is, the model of a store of life-sustaining substance in the cranium, and the ability to move subtle wind through the channels of the body to propel this substance upwards for the purpose of life-extension or death-prevention. There are, however, similar physiological systems within Vajrayāna prior to the Amṛtasiddhi. The examples presented in this section are not intended to be exhaustive (and indeed they are not), but to showcase the existence of such systems in Vajrayāna prior to or within the 11th century.

The concept of something called bindu is found in the 8th-century Guhyasamājatantra as a point upon which a yogin places a seed-syllable visualized in meditation.5 There is also a mention of bindu as a drop of seminal fluid produced through sexual union.6 In the Guhyasamājottara, the 18th chapter of the Guhyasamājatantra, and likely a late 8th-century addition to the original tantra, we see bindu as an object of the mind during prāṇāyāma, a drop or point of prāṇa called “prāṇabindu”.7 Here bindu is an object in meditation, the point one places one’s own seed mantra upon, not a physical substance; however, it is noteworthy that centuries later, at the time of the Amṛtasiddhi’s production, the famed Nāropā locates this bindu in the forehead.8

The Laghukālacakratantra, or simply Kālacakra, produced circa 1030 CE (so more or less contemporaneous with the Amṛtasiddhi), includes four, individual bindus located in respective cakras, which lead to wrong speech and behavior but, when purified, result in a perfected body. This purification happens through binduyoga, where a drop of semen is sent upwards and then diffuses throughout the body, leading to bliss.9

As regards the nāḍīs, a fully developed system of 32 channels, with sun and moon corresponding to the primary right and left channels, respectively, exists in Vajrayāna as early as the late 9th-century Hevajratantra.10 With the Kālacakra we see the number of channels suddenly explode to number into the hundreds. The Kālacakra also employs a six-limbed yoga (ṣaḍaṅgayoga), detailed in a sort-of commentary on the tantra, the 11th-century Ṣaḍaṅgayoga, which comments on a number of texts to explicate its system. Here, in this six-limbed yoga, the goal is upasiddhi, a partial awakening, leading to mahāmudrā, the perfected state11 As mentioned before, the deeply complicated subtle body detailed in the Kālacakra mentions bindu, although not quite in the same way as the Amṛtasiddhi. The tantra prescribes methods of prāṇāyāma, but to put wind into the central channel for the goals of perfection and awakening, not to prolong the yogin’s life.12 There is an equating of bindu with seminal fluid and a connection with the moon, but this connection is explained by both the Kālacakra and the Ṣaḍaṅgayoga as having to do with a co-mingling of fluids in the uterus leading to the development of a fetus, which is part of a larger journey of perfection in which a yogin is influencing his rebirth, or so it seems.13

Bindu can be used by the yogin in his own mental transformation towards perfection in the Kālacakra, but what the bindu actually does is not made clear.14 Somehow an arresting of the semen at the glans of the penis will bring about awakening if, for whatever reason, a perfected yogin does not reach the ultimate goal at the moment he should—“If he does not get this desired perfection of mantrins through pratyāhāra and so on (that is, through the ṣaḍaṅgayoga), he can accomplish it forcefully (haṭhena) by a repetition of sound by arresting the bindu in the vajra-jewel in the lotus.”15 Haṭha and the involvement of the seminal fluid are a last resort, employed when all else fails.

It is noteworthy that in texts prior to the timing of the Amṛtasiddhi, as well as contemporaneous with it, the concept of using force to bring about physical effects is frowned upon in Vajrayāna. Jason Birch and James Mallinson have catalogued over a dozen important Vajrayāna texts that censure using haṭhayoga. Nearly all these texts are earlier than the 11th-century Amṛtasiddhi, although a few are contemporaneous with it; for example, the previously-mentioned Ṣaḍaṅgayoga.16

So we have a subtle physiology, that is to say a “yogic body”, and there is manipulation of vāyus around the body, and, in the case of the Kālacakra and its subsequent tradition, a stoppage of the flow of seminal amṛta/bindu through the body. But even with all this, we do not have the models of life-extending bindu, or techniques to send the substance back up into the cranial vault. That is, we have models where bindu functions as a means to an end, but there are no models where bindu is the end itself, so to speak. However, variations of this model exist in Vajrayāna after the Amṛtasiddhi, in both the SUT and the CMT. That is to say, this model of an ever-dripping, life-extending seminal bindu that can lengthen life and empower the yogin if he manages to get it back into his cranium, through some means.

Yet, there is no smoking gun, an “ity aṃrtasiddhiḥ” or obvious direct quotation. But I believe the accumulation of the uncanny connections is enough to make the case that, even if the texts were not drawing directly from the Amṛtasiddhi itself, they were likely drawing from the same socio-religious paradigm from which the Amṛtasiddhi was produced. Francesco Sferra tells us that the Ṣadaṅgayoga is actually a “collage of quotations”17 drawn from a number of works, giving us the impression that this text, and by extension possibly the Amṛtasiddhi (as well as the Kālacakra), are products of a larger movement within late, esoteric, Sanskritic Buddhism.18

3. The Saṃvarodaya and Caṇḍamahāroṣaṇa Tantras

Circumstantial evidence links the Saṃvarodaya and Caṇḍamahāroṣaṇa tantras to the overall movement towards physical yoga in Vajrayāna, as reflected in the Kālacakra tradition, but none of it ties either tantra specifically to the Amṛtasiddhi.

3.1. Subtle Physiology Similarities with Amṛtasiddhi

In both the SUT and CMT, we see discussions detailing moving vāyu around the body, with the fifth chapter of the SUT and the 22nd chapter of the CMT devoted to the subject, but as mentioned before, prāṇāyāma is integral to Kālacakra literature as well. There is also an association in the SUT and CMT of the sun and moon with various nāḍīs,19 but this idea is also found in Kālacakra literature, as is the concept of these internal celestial bodies eclipsing one another.

But there are similarities in the two tantras that do not seem to be found anywhere in Vajrayāna prior to the period of the Amṛtasiddhi’s creation. We find a development of a kuṇḍalinī analog in the Amṛtasiddhi’s 13th chapter. Here the wind itself ascends the body of the yogin through the female central channel (madhyamā) and pierces (vedha) its way to the top through a series of knots (granthi), culminating in breaking through the door of Brahmā (brahmādvāram). This makes yoga successful (yogaḥ prasidhyati), along with the completion of the Great Seal (mahāmudrā) and the Great Lock (mahābandha).20 In the SUT’s 31st chapter there is a similar scenario, where a dormant female energy that lies at a knot around the base of the spine, is moved to action and ultimately ascends the body of the yogin. However, this female energy in the SUT is identified as “yoginī”,21 which is neither the name given in the Amṛtasiddhi for the central channel itself, which is madhyamā, nor the kuṇḍalinī of haṭhayoga.

A unique similarity with the Amṛtasiddhi is found in the 22nd chapter of the CMT, the vāyuyogapaṭala. Here we see a placement of the five winds that mirrors the Amṛtasiddhi, with prāṇa at the heart, apāna in the anus, samāna around the navel, udāna around the throat, and vyāna pervading the body.22 The only other place we see this particular organization of the five vāyus predating the Amṛtasiddhi is in the 6th-century Śaiva Niśvāsatattvasaṃhitā Nayasūtra;23 however, it is possible that such an organization has been overlooked elsewhere. The CMT could have taken this organization from the earliest mantramārga text, although it seems that a tradition within Vajrayāna, one which the Amṛtasiddhi may be connected to, is a more probable source. The Padmāvatī commentary of the CMT links a breathing yoga to obtainment of five abhijñas, or superknowledges of a Buddha, but there is nothing regarding life-extension.

3.2. Storage of Bindu in the Cranial Vault

We also find mention in both the SUT and CMT of a store of amṛta24 above the ghaṇṭikārandhra, that is, the hole just behind and above the uvula. The model in the SUT is clear. “In the Saṃbhogacakra [which is situated] in the throat, there is a red lotus of sixteen petals; in the middle of the lotus is the character OṂ; above the character, amṛta flows incessantly through the way of the “hole in the uvula”;25 that is, the ghaṇṭikārandhra. This model, of an endless flow of amṛta is certainly prevalent later on in haṭhayoga, such as in the khecarīmudrā where the yogin bends his tongue back to catch the dripping liquid.26

The CMT presents this model of a store of amṛta in the cranium above the ghaṇṭikārandhra in a peculiar way. Following coitus, the yogin is told he “should inhale [his ejaculate] through a pipe in his nose to increase his power.”27 The Padmāvatī commentary of the CMT tells us “sometimes he should draw [male and female sexual fluids] from the Lotus (i.e., a female sexual organ) with his mouth, place them in a vessel, insert a straw (nālikāṃ), take note of his breath, and ingest it through his nostril, that is to say the aperture [beyond] the uvula”; the ghaṇṭikārandhra.28 After giving an alternate way to nasally ingest these substances, the commentary tells us why one should do this—“by constantly performing these procedures, there will be a great increase in the yogī’s strength, inasmuch as he will stop wrinkling, greying, and [even] death.”29 When considering the model we find in the Amṛtasiddhi, this procedure suddenly seems less strange. Bearing in mind the model of a store of life-extending amṛta above the aperture behind the uvula, I suggest the technique the CMT prescribes is alchemical, performed, as it says, to stop aging and even death. The Kālacakratantra itself details a number of elixirs to eliminate wrinkles and gray hair; however, there is no mention anywhere of sending seminal fluid into the cranial vault by nasal means for the purpose of death-prevention.

3.3. Amṛtasiddhi as a Textual Interpretation Aid

It looks as if the Amṛtasiddhi may potentially help clear up textual puzzles in subsequent Vajrayāna works. In the SUT, the description of bindu ascending the body through a progression of corporeal locations corresponding to vowel characters reaches its climax by saying that on the day of the full moon, the yoginī (the apparent kuṇḍalinī analog who/which has presumably been travelling upwards with the bindu) is to the left and the right of mada (intoxication, intoxicating, or even semen, among other meanings).30 Now this makes no sense. What is mada doing here? The Tibetan translator changed the meaning to mūrdhan (forehead or head),31 which thematically makes sense, since the bindu is in the cranium at this point, but the Sanskrit manuscript witnesses all read mada. The SUT has a chapter on mada, devoted to spirituous liquors and their production, so there is precedent with the term in the text, but not in a way that makes sense here. We can of course guess that the meaning of mada is essence or seminal fluid, but the text does not give us precedence to do so.

The seventh chapter of Amṛtasiddhi can strengthen the case of this interpretation, however. It glosses bindu by equating it with several apparent synonyms. “This is bindu, this is candra, this is bīja, this is mada, this is tattva, this is jīva, this is the essence of everything.”32 So here we have mada used in the SUT as the thing in the midst of the ascended yoginī at the aperture of the head following a description of bindu’s ascent, and we have the Amṛtasiddhi telling us that mada can mean bindu. I do not think it is a stretch to interpret the SUT’s mystery mada as bindu. In fact, mada is also mentioned at the beginning of this overall section in the SUT when we are told that a subtle, four-petalled lotus is located in the mahāsukhacakra in the head and that it is the place of mada. Again, the place of intoxication makes no sense in this context. We can see that mada must refer to bindu, and the Amṛtasiddhi gives us support. Just below this, mahāsukhacakra is said to be a 32-petalled lotus from which the syllable HAṂ flows downwards—“haṃkāro’adhomukhaṃ sravati.”33 We are told that this 32-petalled lotus is bījabhūta, made of seed. Here, again, the Amṛtasiddhi can potentially clear things up if we take take bījabhūta to mean this 32-petalled lotus is made of bindu; however, it is also possible bīja refers to the downward-flowing HAṂ, since bīja, or more frequently svabīja, has the esoteric sense of referring to a seed syllable.

4. Final Remarks

Based strictly on the materials reviewed here, namely the Saṃvarodaya and Caṇḍamahāroṣaṇa tantras, it does not appear that the Amṛtasiddhi was important in subsequent Sanskritic34 Vajrayāna textual traditions. While the physiology the yogic text employs does include innovations which seem to be found in these two tantras, the lack of any quotations or references to the text itself makes the possibility of drawing a direct connection very unlikely.

Techniques for placing bindu back into the cranium, whether through the circulation of wind detailed in the Amṛtasiddhi, or snorting it through a straw as in the Caṇḍamahāroṣaṇatantra illustrate common physiological models, but not a textual connection. The shared organization of the five winds in this tantra and in the Amṛtasiddhi is striking,35 but again, this may be attributed to something wider than a single–possibly originally fringe–yogic text. The ability of the Amṛtasiddhi to clarify the otherwise-confusing usage of mada in the Saṃvarodayatantra also lends support for the existence of a shared intellectual landscape regarding subtle physiologies and how the body might be manipulated for longevity. As mentioned in the opening, Mallinson has made clear the importance of the Amṛtasiddhi on subsequent haṭhayoga traditions. The present inquiry could be expanded to ask, not only why the text did not gain traction within Sanskritic Vajrayāna traditions, but also why it was nevertheless so successful in the so-called “Hindu” haṭhayoga ones.

Far more study needs to be done on the milieu from which all these ideas were generated. For example, the Saṃvarodayatantra has a chapter on “rasāyana” that is as yet unedited. There is a project underway by a group of Japanese scholars to slowly edit the Padminī commentary of the Saṃvarodayatantra and I am myself in the process of editing and translating the Padmāvatī commentary of the Caṇḍamahāroṣaṇatantra. The Ḍākarṇavatantra, which is contemporaneous with the Saṃvarodaya and Caṇḍamahāroṣaṇa tantras, apparently has lengthy discussions of physical manipulation of the subtle body, however, with the exception of its apabhraṃśa songs, it remains unedited. David Grey and others are in the process of editing and translating the Sanskrit text. There is also a need for a comprehensive collation of materials preserved in Tibetan, Javanese, Khmer, and other translations, as the Buddhist yogic tradition was never confined to India and it continues to this day.