ISSUE #004 - Aug 27, 2017

The Yogīs’ Latest Trick

James Mallinson

Please visit for a downloadable version of this article and other writings by Dr. James Mallinson.

On Sinister Yogīs1

David Gordon White’s wide-ranging scholarship on tantra, yoga and alchemy has inspired many students and scholars to undertake research in those fields. White worked as an assistant to Mircea Eliade and his doctorate from the University of Chicago was in History of Religions. His research methodology, true to this scholastic heritage, is not as deeply rooted in textual criticism as that of the current vanguard of scholars working on tantra and yoga, whose philological studies rarely reference his work. The accessibility of his books and articles, however, together with his engaging writing style and the excitement that imbues his scholarship, mean that indologists specialising in other fields and authors addressing nonscholarly audiences frequently draw on his publications.2 White’s prominence in the study of yoga and tantra requires all scholars working on those subjects to address his work.

In the preface to his latest monograph, Sinister Yogis, White writes (pp. xi-xii) that the book is the third part of an unplanned trilogy. The Alchemical Body (1995) sought to show that haṭhayoga owes its origins to alchemy. Kiss of the Yoginī (2003) tried to find textual evidence for the “power substances” which White believes underlie both haṭhayoga and alchemy. This raised the question of “why the Tantras used the term ‘yogi’ for practitioners whose goals were supernatural powers, rather than liberation or salvation”. Sinister Yogis is White’s answer to this question; it is an attempt to identify “the (unexpected) origins of yoga in South Asia”.

As suggested by the book’s title, in Sinister Yogis White expounds the bold and provocative thesis that the primary referent of yoga3 in Indic discourse has not been the quietist, meditation-based “classical” yoga practices of Patañjali’s Yogasūtra and later haṭhayogic works, but more occult and extrovert techniques of effecting union by projecting the self outwards in order to overcome death, enter other bodies and effect various kinds of wizardry. This thesis was first advanced by White in a recent article4 and Sinister Yogis is an expansion of that article. Parts of chapters two and four are taken directly from it; the rest of the book fleshes out his argument and, in its first and last chapters, contextualises it with, respectively, fictional stories of yogis in Indic works and reports by travellers in India from the medieval period onwards.

The subject matter of each of the book’s six chapters is as follows. Chapter one, “Tales of Sinister Yogis”, sets the scene with stories of fiendish black magicians found in texts dating from the seventh to twentieth centuries CE and asks at its end “If these be yogis, then what is yoga?”. In chapter two, “Ceci n’est pas un Yogi”, White highlights usages of the word yoga in the Brāhmaṇas, Upaniṣads and Mahābhārata that are far removed from its best known understanding as a meditation-based soteriology, focussing in particular on the vedic chariot warrior’s journey to the sun “hitched to his [chariot] rig” (yogayukta). In chapter three, “Embodied Ascent, Meditation and Yogic Suicide”, White traces descriptions of soteriological ascent, both embodied and visualised, through early Upaniṣads, the Maitrāyaṇīyopaniṣad, the Bhāgavatapurāṇa and Śaiva scriptures, before examining yogic suicide (utkrānti) in tantric works. Chapter four, “The Science of Entering Another Body”, examines both the mechanics of parakāyapraveśa, drawing on theories of perception based on rays of light, and specific instances of it in the Maitrāyaṇīyopaniṣad and Mahābhārata. Chapter five, “Yogi Gods”, explores how gods came to be portrayed as yogis and yogis understood as gods, relying extensively on the Mokṣadharmaparvan of the Mahābhārata as well as Buddhist works, the Bhagavadgītā and various Purāṇas. Chapter six, “Mughal, Modern and Postmodern Yogis”, begins with a comprehensive survey of references to yogis and other ascetics in foreign travellers’ accounts, then turns to yogis as alchemists, soldiers and traders, before looking at interactions between yogis and the British, which, White suggests, precipitated the downfall of the traditional yogi, whose practice has been replaced by the reinvented meditation- and āsana-based yoga of today.

In a relatively short book, White thus covers a wide array of material, ranging in time and space from the Mohenjo-Daro seal identified as Śiva by Sir John Marshall to present-day yogis in America. Throughout White argues his thesis like the most insistent of pūrvapakṣins. There is, perhaps, a need to inform those interested in yoga that there is more to it than sitting (or stretching) quietly and waiting for liberation, and that its textual foundation goes beyond Patañjali’s Yogasūtra. White, however, leaves no room for nuance, ignoring almost everything that argues against his position, in particular the elephant in his room—the huge body of Indic texts written over the last two thousand years which teach a meditation-based yoga. Where contradictions to his thesis are noted, they are dismissed with hubris. The verse cited in translation in the title of the article on which Sinister Yogis is based, “‘Never Have I Seen Such Yogis, Brother’: Yogīs, Warriors and Sorcerers in Ancient and Medieval India”, is attributed to Kabīr: the poet is railing against those soldiers and traders who call themselves yogīs. In the article, White declares that he “will argue against the implicit model of the yogin in this poetic verse,” suggesting that the sixteenth-century poet who wrote the verse did not know what a yogī was but he does.5 In Sinister Yogis, in a note to explain the Yogasūtra’s saṃyama,6 White says that, in his opinion, the aṣṭāṅgayoga taught in the Yogasūtra was not yoga, but “meditative practice”.7

It is White’s wish to give his book an overarching thesis, a grand unifying theory, that is at fault here, hindering reflective scholarship. This is a shame, because there are some interesting observations to be found in Sinister Yogis. See, for example, the argument in chapter two that the archetypal yogic posture padmāsana was originally emblematic of sovereignty; or the idea presented in chapter three of the soteriological journey to the world of brahman being relocated to the body; or the way that the theory of perception being caused by rays of light emitted from the perceiver is used to explain the mechanics of parakāyapraveśa in chapter four; or, in the same chapter, the idea, developed from the work of Johannes Bronkhorst, that meditational yoga originated in Greater Magadha; or the assertion in chapter five that to talk of “microcosm” in the context of yogic meditation is inappropriate — the yogi is to see himself as the universe itself. Chapter six contains a useful survey of travellers’ descriptions of yogis and fakirs which builds on and complements those of Pinch (2006) and Singleton (2009). And the book has an admirable methodological aim, espoused in the preface (p.xii): to investigate the history of yoga through the history of yogis. Despite this promise, however, there is little focus on the yogi other than in the first and last chapters and even in these narrative and historical accounts the yogi is rarely contextualised. White does not address the question of who actually practised yoga, however that yoga might be understood.

The overriding problem with White’s thesis that yoga, or at least yogis, were “sinister” is caused by his conflating the practice of yoga with the siddhis it produces. The Yogasūtra itself lists various supernatural powers which the yogi can attain through his yoga practice. They are numerous and include the ability to enter another body (parakāyapraveśa), which is the subject of White’s fourth chapter. As noted above, White’s explanation of the mechanics of this siddhi are novel and interesting, but such niceties are not mentioned in yoga texts, wherein certain siddhis are simply said to result from certain practices. Thus, in the Yogasūtra’s vibhūtipāda, parakāyapraveśa is said to result from loosening the causes of bondage and understanding the workings of the mind;8 in the Śivasaṃhitā it is one of various siddhis achieved through prāṇāyāma when practised in the second stage (ghaṭāvasthā) of yoga.9 So parakāyapraveśa is not yoga practice itself; it is one of its results.10 This conflation of yoga with its siddhis is evident from the first chapter of the book, a survey of literary evidence of yogis getting up to no good. The yogis described have achieved magical powers as a result of their practice of yoga and other techniques, which they then put to evil ends. The powers are not the practice.

In the Pātañjala and haṭha yoga traditions (which coalesced in texts composed from the sixteenth century onwards, such as Śivānanda Sarasvatī’s Yogacintāmaṇi), siddhis have been said to be impediments to the ultimate aim of yoga, liberation, since the composition of the Yogasūtra in the fourth or fifth centuries CE.11 In the tantric traditions, on the other hand, siddhis are the main aim of the initiate. Various means of attaining them are taught, but the most common is mantra-repetition. When the haṭha variety of yoga practices started to be codified in Sanskrit texts in the eleventh or twelfth centuries, it was soon appropriated by Kaula tantric traditions who brought their bubhukṣu emphasis on siddhis with them. Before long, however, these traditions were sidelined, and haṭhayoga was again predominantly for the mumukṣu.12 In many tantric texts a distinction is made between the sādhaka, who has undertaken the third highest of the four tantric dīkṣās and practises mantra-repetition for siddhis, and the yogin, whose practice is separate from this system, and for whom liberation is usually the main aim. In texts of the haṭhayogic corpus this distinction is reproduced and the word sādhaka is rarely used to refer to the practitioner of haṭhayogic techniques; in the Dattātreyayogaśāstra it is used once, disparagingly, to refer to the practitioner of mantra-repetition.13 In Sinister Yogis, White makes no reference to this distinction and reproduces the conflation of the sādhaka and the yogin found in some Śākta works, such as the Brahmayāmala,14 thereby compounding his conflation of siddhis and yoga.15

Changes in the referents of the words yoga and yogin (and the latter’s vernacular derivatives) lie at the heart of a methodological problem which afflicts Sinister Yogis: it is often unclear whether White is exploring the history of the word yoga or of yoga itself. He says (pp.42-43) that there was a “pure” yoga called yoga which can be discovered from texts as old as the Brāhmaṇas and that anything not called yoga at the time of their composition had nothing to do with yoga, however yoga might now be understood. Thus yoga as the yoking of a warrior’s chariot in the Brāhmaṇas is the original yoga, but the posture assumed by the figure in seal 420 from Mohenjo-Daro has nothing to do with yoga because, according to White, in yoga’s earliest textual formulations āsana was not among its aṅgas.16 Similarly, because early Buddhists did not call any of their practices yoga, those practices were not yogic (p.55). Meanwhile, when the word yoga is used in Āyurveda to denote the interaction between people and their environment, it is, White suggests (p.136), associated with his “pure” yoga; the same is said of the astrological understanding of yoga as a conjunction of heavenly bodies (p.191). Elsewhere, practices which were not called yoga but fit with White’s understanding of what constituted yoga are apparently yogic, such as the cosmic displays of the Buddha described in chapter five. Meanwhile, as noted above with reference to the yogis scorned in the verse attributed to Kabīr and the yoga of Patañjali’s Yogasūtra, people and practices that were called yogīs and yoga were apparently nothing of the sort; others called Yogī (or Jogī), whether snake-charmers or bhajan-singing Nāth householders, are yogis.

White only occasionally admits to the changing referents of the word yoga (but not to its multivalence),17 preferring to imply that there was a constant substrate underpinning its various manifestations and that this substrate had as its bedrock the chariot-yoking of the warrior of the Brāhmaṇas. White translates the latter’s yoga as “rig”; when yogayukta, he is “hitched to his rig”. But White persists in applying this interpretation when it has become inappropriate. Thus evaṃ hy āha | ṣaḍbhir māsais tu yuktasya nityamuktasya dehinaḥ | anantaḥ paramo guhyaḥ samyag yogaḥ pravartate in Maitrāyaṇīyopaniṣad 6.28 is translated as “It has been said of the embodied [individual] who has constantly [remained] “hitched up” for six months [and] who is [thereby] released, [that] his eternal, transcendent, mysterious properly aligned rig rolls forward”. Leaving aside the other mistranslations, yukta and yoga would clearly make better sense here as either “having practised yoga” and “yoga” or “having meditated” and “meditation”.

A common, and old, meaning of the word yoga not remarked upon by White is “magic”. It is hard to see how yoga° in yogacūrṇamiśram auṣadham (Mudrārākṣasa 2.17) or yogarocanā (Mṛcchakaṭikā 3.15) might have anything to do with soteriological yoga, either the chariot-yoking variety espoused by White or the more usual meditational sort. Similarly, yogins are sometimes better understood as “magicians” or “wizards” rather than practitioners of yoga (cf. the sādhaka/yogin confusion outlined above).18

As well as varying the criteria for what constitutes yoga to suit his thesis, White cherry-picks his evidence to do the same, citing passages that support his argument while ignoring those in the very same texts that would argue against it. Thus, on page 196, he mentions a passage in the Jayākhyasaṃhitā in which yogins are described as evil beings. But no mention is made of the same text’s 33rd paṭala, which teaches a six-limbed meditative yoga, by means of which the yogin may achieve mukti. Rastelli, in an article cited in n.140 on p.295 of Sinister Yogis, writes (2000:357): “The yogin is, according to the [Jayākhyasaṃhitā], an ambiguous figure. On the one hand, his refuge (gati) is God; he always thinks of God as being present in his heart; he stays in a temple practising samādhi; he does not even think of something that is harmful to others; and when he attains emancipation, he achieves unity and identity with God. On the other hand, yogins are described as cruel beings…”.

Elsewhere (p.32), White cites a passage from the Bhagavadajjukīya in which by means of yoga (yogena) a yogi enters a young woman’s corpse. But, again, his entering the corpse is not yoga, it is the result of yoga. In the same text (p. 45) yoga is said to happen as a result of vijñāna, saṃyama and tapas, and (p. 48) to be the root of knowledge, the essence of asceticism, abiding in goodness, the destruction of dualities and free from hatred and passion.19

In spite of Sinister Yogi’s dependence on texts, there is much in it to frustrate the philologist. As is to be expected in a book of such broad scope, White relies on the work of several other scholars in his analyses of texts, notably James Fitzgerald, Angelika Malinar, Johannes Bronkhorst, Marion Rastelli, Somdev Vasudeva and Alexis Sanderson. There is little in the way of new textual research and, where drawn on directly, editions are used uncritically: no variant readings are noted nor manuscripts consulted. One of the few works about which White makes text-critical assertions is the Maitryopaniṣad/Maitrāyaṇīyopaniṣad (MU), whose teachings on yoga, and in particular the early date that he assigns them, are pivotal to several of his arguments. White’s dating of the parts of the text which teach yoga might surprise philologists. He says (p.89) that they “belong to one of the latest strata of the text, which I would date, on the basis of its language and content, to about the third century of the common era”. White does not identify what it is about its language and content which leads him to this conclusion but does add that he agrees with those scholars who consider parts of the text “to be late ‘tantric’ additions”.20 In a note to this remark (n.24 p.272), he identifies MU 6.19-22 as being among these later portions. On the next page he says that “MU 6.18 contains the earliest upanishadic account of ‘six-limbed yoga’” and that “MU 6.21 contains the sole mention (in the classical Upanishads) of the suṣumnā as the subtle channel that leads to immortality”. Both these assertions are made notwithstanding the fact that neither ṣaḍaṅgayoga nor the suṣumnā is mentioned in any other text until the seventh century.21 A similarly cavalier approach is applied to the dating of the Siddhasiddhāntapaddhati and Gorakhbāṇī, both of which are said to be older than available evidence suggests. The Siddhasiddhāntapaddhati is assigned to the 12th- 16th centuries (p.175) but there is no evidence for its existence prior to the late eighteenth century, when its oldest dated manuscript was written.22 The Gorakhbāṇī, to which recent scholarship ascribes a date of the 16th-17th centuries,23 is said to date “from no later than the fourteenth century” (p.198).

White’s interpretations of the contents of texts are often erroneous too. In his discussion of early formulations of rājayoga, which he rightly seeks to distance as far as possible from Swami Vivekananda’s identification of rājayoga with Patañjali’s aṣṭāṅgayoga, White suggests (p.46) that the practice of vajrolīmudrā, in which the yogin (or yoginī) draws liquids up through the urethra, is part of rājayoga, adducing Amanaska 2.32,24 the import of which is in fact that those who practise various esoteric techniques including vajrolīmudrā will not achieve physical perfection without rājayoga. One might dismiss this as a simple oversight, but White made the same mistake in his 2003 Kiss of the Yoginī (p.81 and n.85 thereon), citing a translation of Amanaska 2.32 which I had sent him and which he had misunderstood. I wrote to him in 2003 to point this out and he accepted my correction. In the meantime Jason Birch started work on a critical edition of the Amanaska.25 In an email communication with White in 2006, after White had again suggested that Amanaska 2.32 states that vajrolīmudrā is a part of rājayoga, Birch also corrected him.26 Ironically, White would not have had to look far within the corpus of early texts on haṭhayoga for an assertion that vajrolīmudrā, if not being part of rājayoga, at least leads to it: this is said at Dattātreyayogaśāstra 160.27

Page 171 of Sinister Yogis introduces a section entitled “Yogi Practice in Mahābhārata 12.289”. White says that “[a]ccording to its colophon, the title of this chapter is Yogavidaḥ, a compound that would normally translate as ‘An Understanding of Yoga’.” Surprised by this apparent mistranslation, I checked the Pune critical edition. In 25 of 26 witnesses the chapter is called yogavidhiḥ; one calls it yogakathanam. Nowhere in any manuscript colophon, nor in the chapter itself, is the compound yogavidaḥ to be found. Where it does occur in the Mahābharata, it is as the nominative plural of yoga-vid, “knower of yoga”.28

Doubts about the linguistic rigour of Sinister Yogis are raised from its outset. The Note on Transliteration (p. xvii) mentions a “Bhairāvanand Yogī”. The star of the preface is one “Bhandarināth” (surely “Bhaṇḍārīnāth”, as in Śiva’s Hindi epithet “Bhole Bhaṇḍārī”). Certain key words are consistently misspelt (both in Sinister Yogis and in White’s 2009 article): jina is found everywhere, including in the index, as jīna; pāda (as in a textual division of a tantric work or the Yogasūtra) as pada; nikāya as nikaya. The same is true of important Hindi terms: akhāḍā is everywhere ākhāḍa; Kumbh(a) Melā is Kumbhā Melā; and the long final -ī of sect names such as Rāmānandī, Udāsī and Daśanāmī is shortened. The English in the book contains a higher than average, but passable, number of typographic errors. The Sanskrit and Hindi citations, however, are full of them and other infelicities.29 There is no consistency in the forms of the words or compounds cited: sometimes they are presandhi, sometimes post-sandhi, sometimes stems. Text names are sometimes compounded, sometimes not, and there is no consistency in their abbreviated forms, which, since every name is abbreviated from its second instance onwards, can lead to confusion.30

This linguistic sloppiness has little effect, however, on White’s main thesis: an argument devoid of nuance does not hinge on the length of a vowel. One sometimes suspects that White himself is aware that his argument is one-sided. Rather than let the reader decide its merits for him- or herself, he coerces the reader to accept his opinion. A trivial but telling example of this is found on p.213 when White cites a passage from the memoirs of John Fryer, a British doctor who travelled in India in the late seventeenth century. The passage describes a yogi showing off the haṭhayogic technique known as nauli, which White says “could only be uḍḍīyāna bandha”, a related but quite different practice. One also detects a subtle coercion in the unexplained change in the name White gives members of the Gorakhnāthī ascetic order. In The Alchemical Body they were Nāth Siddhas; in Sinister Yogis they are Nāth Yogīs.31 White’s not letting the reader make his or her own inferences is a corollary of his approach. One suspects, and one’s suspicions are compounded by White’s explanation of the origins of the second and third books of his trilogy, that he goes looking for particulars when undertaking his textual research, rather than reading broadly and making his own inferences. This scholarly pitfall is all the more likely to catch us out in the age of etexts, when in a few seconds we can find multiple references to any word we search for and seize on them out of context.

The last chapter of Sinister Yogis is predicated on White’s theory that an original siddhi-based yoga and its yogis declined progressively to the point where it has now almost disappeared, having been eclipsed by a sanitised, rational “rājayoga” propounded most famously by Vivekananda. White rightly associates these siddhi-oriented yogis with the order which from perhaps the early eighteenth century came to be known as the Nāths, but whose members, heirs to the Paścimāmnāya sādhaka tradition that originated in the Deccan, were before then known simply as Yogīs.

That these “Nāth” Yogīs were the originators and foremost exponents of haṭhayoga is a given of all historical studies of yoga. But these Yogīs were in fact the willing and complicit beneficiaries of the semantic confusion which has caught out White and many other scholars. White rightly notes (p. 205) that reports of these Yogis actually practising yoga are almost non-existent. This is because they did not practise much yoga (or at least the clearly visible methods of haṭhayoga). As the heirs to the Paścimāmnāya sādhaka tradition they pursued siddhis through means such as tantric ritual, mantra-repetition, alchemy and visualisation-based layayoga. These Yogīs were yogis as magicians. Meanwhile ascetic orders which did practise the physical techniques of haṭhayoga but were quite separate from the Yogīs were flourishing from at least the early medieval period.32 This is evinced in texts such as the thirteenth-century Dattātreyayogaśāstra and the 15th-century Śivasaṃhitā. The Dattātreyayogaśāstra is the first text to teach a systematised haṭhayoga and call it as such. Its haṭhayoga involves the cultivation and preservation of bindu, semen. The yoga of one of the earliest texts of the Gorakṣa Yogī tradition, the thirteenth- to fourteenth-century Vivekamārtaṇḍa, is a visualisation-based Kuṇḍalinī yoga (which originated in the Paścimāmnāya Kaula tradition) overlaid onto the bindu-oriented physical haṭhayoga of the Dattātreyayogaśāstra. Neither the Vivekamārtaṇḍa nor a contemporaneous work on yoga also attributed to Gorakṣa, the Gorakṣaśataka,33 calls its yoga haṭha.34

Although, like most yoga texts, they teach that the yogi’s sectarian affiliation or lifestage is of no importance, certain features of the the Dattātreyayogaśāstra and the Śivasaṃhitā show that they were affiliated to two ascetic traditions which in the sixteenth or seventeenth centuries formally coalesced, with other lineages, to form the Daśanāmī Saṃnyāsī saṃpradāya, and from subgroups of which it is likely that the Rāmānandīs seceded, probably at some point in the seventeenth century.35

Members of both these ascetic orders, and their forerunners, have practised a liberation-oriented haṭhayoga, in contrast to the siddhi-oriented yoga of the Gorakhnāthīs, since at least the time of composition of the first haṭhayoga texts and probably for much longer. They have been almost completely ignored in scholarship on haṭhayoga, even though all the many Sanskrit texts on haṭhayoga and commentaries thereon composed or compiled since the fifteenth century have been written by scholars of these traditions. I do not know of a single Sanskrit text on haṭhayoga or commentary written by a member of the Gorakṣa Yogī tradition since the Haṭhapradīpikā.36 The one doctrinal (as opposed to hagiographical or devotional) Sanskrit text composed by Nāths after the formalisation of their order, the c. 18thcentury Siddhasiddhāntapaddhati, does not teach haṭhayoga; on the contrary, in its more avadhūta moments it scorns it.37 To this day the Nāth Yogīs’ association with yoga is little more than in name: I have tried in vain to find any adepts of haṭhayoga among today’s Nāths.38 None of the important schools of haṭhayoga in India today has a Nāth guru; the majority are affiliated with the Daśanāmī Saṃnyāsīs.

The reasons for the Daśanāmī Saṃnyāsīs’ and Rāmānandīs’ being ignored in scholarship on yoga are many and various, and I will mention only one here, since it is germane to the subject in hand: the name “Yogī”. While Daśanāmīs and Rāmānandīs will happily refer to themselves in conversation as yogīs and the word is used frequently in their liturgies, as a sect name it has long been the preserve of the Gorakhnāthīs. Its ‘sinister’ connotations and its use as a caste name by a variety of low-status groups including snake-charmers and Muslim devotional singers have led to its being eschewed as a title by even the Nāths themselves (as noted by White on p.223), as well as by higher-status Yogī castes in Rajasthan who now also call themselve Nāths.39 These undesirable connotations have long since led the Daśanāmīs and Rāmānandīs, whose members have in the main been drawn from higher-status social groups than those of the Nāth Yogīs, to distance themselves formally from the appellation Yogī (although members of those orders who have mastered haṭhayoga do regularly append the honorific title “Yogirāj” to their names).40 But such semantic nuances are, understandably, not immediately apparent to scholars in search of practitioners of yoga, so Nāth Yogīs are usually their first port of call.

The mumukṣu yoga tradition of haṭhayoga, which is espoused by the Rāmānandīs and Daśanāmīs, has been the predominant variety of yoga practice for at least five hundred years and is the basis of much of the yoga practised in the West, but it is completely ignored or overlooked by White in Sinister Yogis. This is well illustrated by the legend he relates on page 242 (which is perhaps based on a historical incident from the 16th century) in which he sees the victory in a magical contest of the Rāmānandī Kṛṣṇadās Payohārī over the Yogī Tāranāth as that of a “nonyogi” bhakta over a yogi Yogī. Kṛṣṇadās Payohārī is remembered in the Vaiṣṇava tradition as a yogi par excellence. As his soubriquet “Payohārī” implies, he was said to live off milk, and in the Bhaktamāl of Nābhadās, which was composed in about 1600, he is described as a mahāmuni whose seed was turned upwards (ūrdhretā – the condition of the yogi who has mastered his bindu) and who was waited on by kings.41 In one of the maṅgala verses at the beginning of his Braj Bhāṣā manual of [haṭha]yoga, the Jogpradīpakā, which was composed in 1737 CE, the Rāmānandī Jayatarāma invokes “Payhārī”. Rāmānandīs today tell the story of Kṛṣṇadās Payohārī visiting Kullu with his troop of 400 celibate ascetics. The king asked if there was anything he could do for them and Kṛṣṇadās replied that he needed lakḍī, firewood, for his men. The king returned the next day with 400 laḍkīs, girls. Kṛṣṇadās, to preserve the girls’ honour, made his troop marry them (thus founding the Vairāgī castes of the Kullu valley), but could not bring himself to do so and went into samādhi in a cave near Naggar, from which the locals are still waiting for him to emerge.

Payohārī Jī’s yogic mastery was not exceptional in the Rāmānandī tradition. His disciple Kīlha and grand-disciple Dvārkādās are remembered as great yogis who conquered death (just as Gorakhnāth does in a variety of Nāth legends), with Kīlha being said in Nābhādās’s Bhaktamāl to have used the power of his yoga practice to exit his body through the aperture of Brahmā at the top of the skull.42 Members of another bhakta saṃpradāya, the Dādūpanth, also combined the practice of yoga with a devotional attitude. The 16th-century sant Dādū and his troop of yogis are portrayed in the late 18th-century Kaṛakhā or “Battlesong” of Santdās Mārū Galtānī as defeating Yama, the god of death.43 Dādū composed a text on bodily yoga, the Kāyābelī, and the 17th-century Dādūpanthī scholar-poet Sundardās wrote extensively on yoga in his Sarvāṅgayogapradīpikā and Jñānasamudra.

In spite of these well-known bhakta yogis, bhakti and yoga are often thought to be incompatible,44 which is perhaps why White assumes Payohārī Jī to have been a “nonyogi”. Bhakta yogis have in fact long been a common feature of the Vaiṣṇava devotional tradition as evinced by the 9th- to 10th-century Bhāgavatapurāṇa.45

The final chapter of Sinister Yogis charts the supposed demise of the yogi (i.e. the Nāth Yogī). This is predicated upon a historiographical model not dissimilar to that employed in early analyses of Buddhism in which Gorakhnāth established a panIndian Yogī order that has been in slow decay ever since. But this is not borne out by our sources. The earliest references to an organised Yogī order date to the late sixteenth century.46 Even after that time, notwithstanding occasional localised Nāth influence, there is little evidence of a truly unified pan-Indian Yogī order until the establishment of the Yogī Mahāsabhā at the beginning of the twentieth century.47 While many nijī or private Nāth establishments have, over the years, met their end—as is the nature of monasteries dependent on their residents’ charisma to attract patronage—others, especially the pañcāyatī establishments controlled by the Mahāsabhā, are flourishing. The impressive campus around the Gorakhnāth mandir at the Nāth headquarters in Gorakhpur is a fine example of a flourishing nijī sthān. Its current mahant, Yogī Ādityanāth, is the BJP MP for Gorakhpur, in the light of which White’s assertion that after independence “virtually all of South Asia’s yogis were reduced to beggar status” (p.243) is questionable at best.

Before drawing attention to the demise of the yogis, White surveys descriptions of them in the accounts of travellers from the early fifteenth century onwards. On p.200 he notes how Europeans used the terms yogi and fakir “indiscriminately”. Yet White does the same, conflating Muslim fakirs with Hindu yogis throughout chapter six and adducing reports of bizarre practices undertaken by holy men of any sort to support his argument that yogis were sinister. Thus he includes (pp. 211-212) Tavernier’s description of Muslim fakirs in their trademark patchwork cloaks and dragging chains, and (pp. 213-214) Fryer’s description of a “Fakier” (who is explicitly said to be “a holy man among the Moors”). This insistence on highlighting the yogis’ (and fakirs’) exoticism, together with Sinister Yogis’ thesis, which similarly highlights the bizarre but ignores the more exoteric facets of yoga and yogis, opens White to accusations of an Orientalist bias. But of course sinister yogis, black magicians using their siddhis for personal gain, make for better stories than benevolent yogis quietly meditating on the fringes of society or ministering to the needs of their devotees.48

In the last pages of Sinister Yogis, White muses upon the ability of yogis to take on others’ identities, whether through parakāyapraveśa or deceit, and summarises two cases from the nineteenth century of princes reappearing as yogis some years after having died. The Yogīs’ appropriation of haṭhayoga and adoption of the guise of haṭhayogins has, as noted above, happened for various reasons, in some of which the Yogīs have themselves been complicit. The Yogisampradāyāviṣkṛti (Yogī 1924), which assimilates a tradition of nine Nārāyaṇas headed by the divine yogin Dattātreya with that of the nine Nāths and which is claimed in its preface and by others, including White (1996: 408 n.183), to be a translation into Hindi of a Bengali translation of a Marathi work written in the thirteenth century by Jñāndev, is in fact a twentieth-century account of the Yogī order, drawing in places on older texts but written in the first person by one Candranāth Yogī. The Yogīs’ appropriation of haṭhayoga began long ago: the c. 1450 CE Haṭhapradīpikā compilation includes twenty verses from the Dattātreyayogaśāstra, the first text to teach a system of haṭhayoga and call it as such, but credits Matsyendra, Gorakṣa and other mahāsiddhas with haṭhayoga’s invention and makes no mention of Dattātreya. The appropriation continues to this day. Recent editions of the Yogabīja and Amanaska(yoga) from the Gorakhpur Nāth mandir attribute both texts to Gorakṣanātha although there is nothing in the manuscripts of either to connect it with the Nāth tradition. White regularly repeats these attributions in his works.49

The Nāth Yogīs’ usurpation of the yogic tradition of the Daśanāmī Saṃnyāsīs is enshrined on the front of the dustjacket of Sinister Yogis, where we find a reproduction of a painting of an ascetic from the wall of a haveli in Mukundgarh, Rajasthan, dated to 1880 CE.50 The picture is said in its caption to be of a Nāth Yogī. It is in fact almost certainly of a Daśanāmī Nāgā Saṃnyāsī. The tilak, the moustache with an absent or perhaps tied-back beard, the earrings through the earlobes and the sword are of a piece with contemporaneous images of Saṃnyāsīs of the Giri order.51 Particularly pertinent are the sword—contrary to received opinion, there are no records of Nāth Yogīs ever having been militarised in the manner of the Daśanāmī and Rāmānandī nāgās52—and the earrings. By the beginning of the nineteenth century, as part of the consolidation of their order under the tutelage of Gorakhnāth, the Nāth Yogīs had begun to distinguish themselves from other ascetic orders by wearing their earrings through holes cut in the cartilages of their ears, not in the lobes (hence their somewhat pejorative appelation kānphaṭā). In misidentifying the Daśanāmī Saṃnyāsī in the painting thus, White has unwittingly brought about the latest instance of the Nāth Yogīs’ siddhi of parakāyapraveśa.

Works consulted

Primary Sources

Amanaska, see Birch 2013.

Amṛtasiddhi of Virūpākṣanātha. Maharaja Mansingh Pustak Prakash, Jodhpur, Acc. Nos. 1242 and 1243.

Kanhāvat of Malik Muhammad Jāyasī, ed. Parameśvarī Lāl Gupta. Vārāṇasī: Annapūrṇā Prakāśana. 1981

(Śrī) Guru Granth Sāhib, with complete index prepared by Winand M. Callewaert. 2 parts. Delhi: Motilal Banarsidass. 1996.

Jayākhyasaṃhitā, ed. Embar Krishnamacharya. Baroda: Oriental Institute. 1967.

Jñān Samudra of Sundardās, in Sundar-Granthāvalī, ed. Rāmeś Candra Miśra. Naī Dillī: Kitāb Ghar. 1992.

Dattātreyayogaśāstra, unpublished critical edition by James Mallinson.53

Padmāvatī of Jayasī = The Padumāwati of Malik Muḥammad Jaisi, ed. G.A.Grierson and S.Dvivedi. Calcutta: Asiatic Society. 1911.

Bodhasāra of Narahari, ed. Jennifer Cover; Grahame Cover; Kanchan V.Mande. Charleston, S.C.: CreateSpace. 2010.

(Śrī)bhaktmāl, ed. Śrīvrajvallabhśaraṇ. Vṛndāvan: Śrīsarveśvar Pres. 1960.

Bhagavadajjukīyam of Bodhāyana Kavi, ed. P. Anjuan Achan. Jayantamangalam: Paliyam Mss. Library. 1925.

Mudrārākṣasa of Viśākhadatta, ed. M.R.Kale. Delhi: Motilal Banarsidass. 1976.

Mṛcchakaṭika of Śūdraka, ed. M.R.Kale. Delhi: Motilal Banarsidass. 1999.

Mahābhārata, ed. V.Sukthankar, with the cooperation of S.K.Belvalkar, A.B.Gajendragadkar, V.Kane, R.D.Karmarkar, P.L.Vaidya, S.Winternitz, R. Zimmerman and other scholars and illustrated by Shrimant Balasaheb Pant Pratinidhi. (Since 1943 ed. S.K.Belvalkar). 19 Vols. Poona: Bhandarkar Oriental Research Institute, 1927–1959.

Maitrāyaṇīyopaniṣad, ed. J.A.B. van Buitenen. The Hague: Mouton and Co. 1962.

Yogabīja, ed. Rām Lāl Śrīvāstav. Gorakhpur: Śrī Gorakhnāth Mandir. 1982.

Yogaśāstra of Hemacandra, ed. Olle Quarnström. Cambridge: Harvard University Press. 2002.

Yogasūtra of Patañjali with the commentaries (Bhāṣya, Tattvavaiśāradī, and Yogavārttikā) of Vyāsa, Vācaspatimiśra, and Vijñānabhikṣu, ed. Nārāyaṇa Miśra. Benares: Bhāratīya Vidyā Prakāśan. 1971.

Vārāṅ Bhāī Gurdās, ed. Jodh Singh. Patiala: Vision \& Venture. 1998.

Śivasaṃhitā, ed. and tr. J. Mallinson. New York:

Śūnyasaṃpādane, Vols. I and V, eds. S.C. Nandimath, L.M.A. Menezes, R.C. Hiremath, M.S. Sunkapur. Dharwar: Karnatak University. 1965-1972. 2006.

Sarvāṅgayogapradīpikā of Sundardās, in Sundar-granthāvalī, ed. Purohit Hari N. Śarmā. Calcutta. 1936.

Sidh Goṣṭh, pp. 938-946 in Śrī Guru Granth Sāhib.

Siddhasiddhāntapaddhati of Gorakṣanātha, ed. M.L. Gharote and G.K. Pai. Lonavla: Lonavla Yoga Institute. 2005.

Haṭhapradīpikā of Svātmārāma, ed. Svāmī Digambarjī and Dr Pītambar Jhā. Lonavla: Kaivalyadhām S.M.Y.M. Samiti. 1970.

Secondary Sources

Alter, Joseph S. 2011. Moral Masculinity: Sex and Maculinity in Modern India. Delhi: Penguin.

Badger, George Percy. 1863. The Travels of Ludovico di Varthema in Egypt, Syria, Arabia Deserta and Arabia Felix, in Persia, India and Ethiopia, A.D. 1503 to 1508. Translated from the original Italian Edition of 1510, with a Preface, by John Winter Jones, Esq., F.S.A., and edited, with notes and an introduction, by George Percy Badger. London: The Hakluyt Society.

Birch, Jason. 2013. The Amanaska. King of All Yogas. A Critical Edition and Annotated Translation with a Monographic Introduction. Thesis submitted for degree of Doctor of Philosophy, University of Oxford.

Bouillier, Véronique. 2008. Itinérance et vie monastique. Les ascètes Nāth Yogīs en Inde contemporaine. Paris: Éditions de la Maison des sciences de l’homme.

Bronkhorst, Johannes. 1998. The Two Sources of Indian Asceticism. Second edition. Delhi: Motilal Banarsidass.

Burchett, Patton. 2012. Bhakti Religion and Tantric Magic in Mughal India: Kacchvāhās, Rāmānandīs, and Nāths, circa 1500-1750. PhD thesis, Columbia University.

Clark, Matthew. 2006. The Daśanāmī-Saṃnyāsīs: The Integration of Ascetic Lineages into an Order. Leiden: Brill.

Dalrymple, William. 2009. Nine Lives: In Search of the Sacred in Modern India. London: Bloomsbury.

Doniger, Wendy. 2009. The Hindus: An Alternative History. London: Penguin.

Gharote, M.L. and Bedekar, V.A. 2005. Descriptive Catalogue of Yoga Manuscripts (Updated). Lonavla: Kaivalyadhama S.M.Y.M. Samiti.

Gold, Anne Grodzins. 1992. A Carnival of Parting: The Tales of King Bharthari and King Gopi Chand as Sung and Told by Madhu Natisar Nath of Ghatiyali, Rajasthan. Berkeley: University of California Press.

Hopkins, E.Washburn. 1901. Yoga-Technique in the Great Epic, pp. 333-379 in Journal of the American Oriental Society, Vol. 22.

Horstmann, Monika. 1991. “On the Dual Identity of Nagas”, pp. 255-271 in Devotion Divine: Studies in Honour of Charlotte Vaudeville, eds. D.L. Eck and F. Mallison. Groningen and Paris: E. Forsten/École Française d’Extrême-Orient.

Locke, John K. 1980. Karunamaya. The Cult of Avalokitesvara-Matsyendranath In the Valley of Nepal. Kathmandu: Sahayogi Prakashan.

Mallinson, James. 2011a. “Siddhi and Mahāsiddhi in Early Haṭhayoga,” pp. 327-344 in Yoga Powers, ed. Knut Jacobsen.

——————. 2011b. “The Original Gorakṣaśataka, pp. 257-272 in Yoga in Practice, ed. David Gordon White. Princeton: Princeton University Press.

—————–. 2011c. “Nāth Saṃpradāya,” entry in the Brill Encyclopedia of Hinduism, Vol. 3, ed. Knut A. Jacobsen, pp. 407-428. Leiden: Brill. —-

————-. 2013. “Yogic Identities: Tradition and Transformation”. Smithsonian Institute Research Online:

—————–. Forthcoming. “Śāktism and Haṭhayoga”, in The Śākta Traditions. London:Routledge Curzon. A draft of this article is available on my page.

Offredi, Mariola. 1999. “Some Concepts of Gorakh Yoga through the Analysis of Three Nāthpanthī Manuscripts,” pp. 267-286 in Studies in Early Modern Indo-Aryan Languages, Literature and Culture, ed. Alan W. Entwistle and Carol Salomon with Heidi Pauwels and Michael C. Shapiro.

Ondračka, Lubomír. 2007. “Perfected Body, Divine Body and Other Bodies in the Nātha-Siddha Sanskrit Texts,” pre-print of the paper presented at the 13th Sanskrit Conference (Edinburgh 2006), accessed from

Paramasivan, Vasudha. 2010. Between Text and Sect: Early Nineteenth Century Shifts in the Theology of Ram. Dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in South and Southeast Asian Studies in the Graduate Division of the University of California, Berkeley.

Pinch, William R. 2006. Warrior Ascetics and Indian Empires. Cambridge: Cambridge University Press.

Rastelli, Marion. 2000. “The Religious Practice of the Sādhaka According to the Jayākhyasaṃhitā,” pp.319-395 in Indo-Iranian Journal Vol. 43, No. 4.

Sastri, V. V. Ramana. 1956. “The Doctrinal Culture and Tradition of the Siddhas”, pp. 300-308 in Haridas Bhattacharyya (ed.). The Cultural Heritage of India. Vol. 4: The Religions. Calcutta: The Ramakrishna Mission Institute of Culture.

Shea, David and Troyer, Anthony. 1843. The Dabistān: or School of Manners. London: Oriental Translation Fund of Great Britain and Ireland.

Singleton, Mark. 2010. The Yoga Body. New York: Oxford University Press.

Unbescheid, Günter. 1980. Kānphatā. Untersuchungen zu kult, mythologie und geschichte sivaistischer tantriker in Nepal. Wiesbaden: Franz Steiner Verlag.

Vasudeva, Somdev. 2004. The Yoga of the Mālinīvijayottaratantra. Pondicherry: Publications de l’Institut français d’Indologie No. 97.

White, David Gordon. 1996. The Alchemical Body. Chicago: University of Chicago Press.

——————–. 2003. Kiss of the Yoginī: “Tantric Sex” in its South Asian Contexts. Chicago: University of Chicago Press.

——————–. 2009. “‘Never Have I Seen Such Yogis, Brother’: Yogīs, Warriors and Sorcerers in Ancient and Medieval India,” pp. 86-113 in Ancient to Modern: Religion, Power, and Community in India, ed. Ishita Banerjee-Dube and Saurabh Dube. New York: Oxford University Press.

Yogī, Candranāth. 1924. Yogisampradāyāviṣkṛti. Ahmadābād: Śivanāth Yogī


  1. This is a review article of Sinister Yogis by David Gordon White. Chicago: Chicago University Press. 2009. ISBN: 978-0-226-89513-0. I am grateful to Alexis Sanderson, Harunaga Isaacson, Shaman Hatley, Dominic Goodall, Mark Singleton, Jason Birch, Patton Burchett, Matthew Clark, Alex Watson, James Fitzgerald and Andrew Nicholson for their comments on earlier drafts.
  2. As examples of the former, see Pinch 2006:200-210, Doniger 2009: ch. 15 or Alter 2011: ch. 5; for the latter see e.g. Dalrymple 2009: ch. 9.
  3. A note about terminology is necessary here. When I write of “yoga”, it is in the general English dictionary sense of a spiritual discipline; its italicised form, “yoga”, indicates that I am referring to a specific usage in an Indic language; “yogi” is a practitioner of yoga; “yogī/yogin”, like “yoga”, refers to a specific usage in an Indic language; “Yogī” or “Jogī” refer to the religious orders or castes which go by those names. Thus on page 17 it is not a typographical error when I mention a “yogi Yogī”.
  4. White 2009.
  5. On the referents of the word yogī in this verse, see footnote 40.
  6. Note 215 on pp. 288-289.
  7. Cf. the dismissal in n. 140 (p. 277) of Alexis Sanderson’s understanding of the word yoga in Śaiva sources as referring to techniques of meditation.
  8. Yogasūtra 3.38: bandhakāraṇaśaithilyāt pracārasaṃvedanāc ca cittasya
  9. Śivasaṃhitā 3.58-63.
  10. An exception to this is found in the instructions on how to practise parakāyapraveśa in Hemacandra’s Yogaśāstra (5.264-273), in which the yogi is to insert his breath into inanimate objects of increasing size. This technique, which develops similar practices known as vedhas taught in earlier Śākta works, suggests a greater rôle for the breath in parakāyapraveśa than it is given in White’s analysis.
  11. Yogasūtra 3.37: te samādhāv upasargā vyutthāne siddhayaḥ ||
  12. See Mallinson 2011a.
  13. In the fourfold classification of yoga first taught in the Dattātreyayogaśāstra, mantrayoga is the lowest variety, suitable for the lowest category of aspirant, he of puny intellect:
    alpabuddhir imaṃ yogaṃ sevate sādhakādhamaḥ |
    mantrayogo hy ayaṃ prokto yogānām adhamas tathā ||14||
    Amongst other works of the early haṭha corpus, only the strongly Śākta Śivasaṃhitā, whose yoga includes repetition of the Śrīvidyā mantrarāja, repeatedly refers to the practitioner of haṭhayoga as a sādhaka.
  14. I am grateful to Shaman Hatley for drawing my attention to the conflation of sādhaka and yogin in the Brahmayāmala (personal communication 7th December 2010).
  15. See especially p.195 and below, p. 13.
  16. This assertion is predicated upon White’s claim that the earliest systematisation of the limbs of yoga can be found in the Maitrāyaṇīyopaniṣad, his dating of which does not stand up to scrutiny: see below, p. 9. The earliest extant division of yoga into aṅgas is found in the Yogasūtra, in which āsana is the third aṅga. Whether or not the Mohenjo-Daro seal has anything to do with yoga is, I would argue, a moot point because it is so far removed in time from our earliest evidence of yoga and we know so little of its context. But White’s logic for dismissing it as not being yogic is flawed. A parallel can be drawn with the mentions of ascetics standing on their heads (or at least hanging upside-down) in the Mahābhārata (see e.g. 1.26.2, 3.185.5, 12.126.18, 13.3.9, 13.7.11). This practice is nowhere therein explicitly associated with yoga, but more than a thousand years later the headstand became an iconic yogic āsana. It seems to me to be plausible to suggest that some of the yogic features of the practice were already present at the time of the Mahābhārata before it was taught in later yoga texts.
  17. On pp.108-109 White concedes that the word yoga went through a semantic shift in the first half of the first millennium CE, one of a series of blows apparently suffered by the “pure” yoga throughout its history (this one was administered by the Bhagavadgītā, Abhinavagupta and other Śaiva jñānins then did it down further, the British had a go during the Rāj and Swami Vivekananda finished it off). Despite his elsewhere not allowing for the possibility of yoga having multiple meanings, on p. 41 White refers to the two usages of the root yuj noted in Pāṇini’s Dhātupāṭha: yoge, i.e. “in [reference to] union” and samādhau, whose specific referent is unclear— it may not yet have acquired its yogic meaning. White’s essentialist adherence to the idea of a “pure” yoga requires him to understand as orthogenetic subsequent developments in practice which he believes are in keeping with it, thereby forcing him to make some unlikely connections. On p. 140 we learn that the yoga of the chariot warrior branched into utkrānti and visionary ascent, and (loc. cit.) that “Śākāyanya’s sojourn in Bṛhadratha-Marut’s heart… becomes the model for the dynamic of ‘co-penetration’ (samāveśa) in tantric initiation…”.

    That yoga did not acquire the meaning by which it is now best understood until the later classical Upaniṣads, despite some of its techniques being taught in earlier ones, was pointed out more than a hundred years ago by Hopkins (1901:334; see ibid.:338 on other meanings of yoga in the Mahābhārata).

  18. It is easier to distinguish between these two types of yogin in vernacular languages. Yogī can still have ‘sinister’ connotations in Hindi, but that role is better played by jogī, the latter also carrying with it associations of low social status. Once, in a typical tourist-shopkeeper interaction in Jodhpur, I facetiously responded to the usual questions by saying, in Hindi, that I was a jogī. The shopkeeper was most upset. When I tried to explain that I meant that I practised yoga, he told me that I was then a yogī and that I must never call myself a jogī.
  19. Bhagavadajjukīya p. 48:
    jñānamūlaṃ tapaḥsāraṃ sattvasthaṃ dvandvanāśanam |
    muktaṃ dveṣāc ca rāgāc ca yoga ity abhidhīyate ||
  20. See for example Vasudeva 2004:375-376 n.18.
  21. The suṣumnā is mentioned in the Niśvāsatattvasaṃhitā’s Uttarasūtra (5.37), where it is paired with the iḍā channel and not located in the centre of the body as in later texts. The Niśvāsa’s nayasūtra (7.42) mentions ṣaḍaṅga yoga in a context which suggests that the concept was already well established (Dominic Goodall, draft introduction to a critical edition of the Niśvāsatattvasaṃhitā).

    As noted above, on pp. 55-56 White had already implied that the MU’s passage on ṣaḍaṅgayoga dates to the third century CE and adduced its omission of āsana from its six limbs to his argument for the lateness of āsana’s inclusion among the limbs of yoga, further implying that the MU’s ṣaḍaṅgayoga predates the aṣṭāṅgayoga of the c. 4th century CE Yogasūtra (which does include āsana).

  22. On the date of this manuscript, see Gharote & Bedekar 2005:460-461.
  23. See e.g. Offredi 1999:270.
  24. Amanaska 2.32 (as Jason Birch argues in his introduction to the critical edition of this text, its name is likely to have been Amanaska rather than the commonly found Amanaskayoga):
    ke cin mūtraṃ pibanti svamalam atha tanoḥ ke cid ujjhanti lālām
    ke cit koṣṭhaṃ praviṣṭā yuvatibhagapatadbindum ūrdhvaṃ nayanti
    ke cit khādanti dhātūn akhilatanuśirāvāyusaṃcāradakṣāḥ
    naiteṣāṃ dehasiddhir vigatanijamanorājayogād ṛte syāt ||2.32||
  25. Birch has critically edited the Amanaska under the supervision of Professor Alexis Sanderson as part of a thesis for the degree of Doctor of Philosophy at Oxford University, which he successfully submitted for examination in May 2013.
  26. Birch is cited (p.264 n.32) in Sinister Yogis in the context of the Amanaska’s rājayoga being “at variance with the teachings of the [Yogasūtra]”.
  27. In most texts of the early haṭhayoga corpus, rājayoga is a synonym of samādhi, rather than a particular system of practices. For a detailed analysis of the meaning of rājayoga, see Birch 2013:65-69. In two Braj Bhasha works, the seventeenth-century Sarvāṅgayogapradīpikā of Sundardās (ch. 3) and the eighteenth-century Jogpradīpakā of Jayatarāma (v. 552), vajrolīmudrā is equated with rājayoga. One reason for this identification is that through vajrolīmudrā a king may achieve yoga without becoming a celibate ascetic. Cf. Divākara’s commentary on Bodhasāra 14.1: … rājayogo rājñāṃ nṛpāṇāṃ svasthāne sthitvāpi sādhayituṃ śakyatvāt
  28. I am grateful to Professor James Fitzgerald for providing me with this information.
  29. Chapter four, for example, contains the following:
    p. 129 utkrāmanta for utkrāmantam, parākranti for parākrānti;
    p. 130 puruṣa is described as kūṭasthā;
    p. 282 n. 60 ātmanātmānam is said to be a compound;
    p. 141 Kauśītaki for Kauṣītaki;
    p. 143 praviṣṭhā for praviṣṭā;
    p. 144 nāna for nānā;
    p. 149 yogātmakenośanasā is translated as “by the yogic self of Uśanas”;
    p. 153 aṇima for aṇimā;
    p. 154 vyāptitva for vyāpitva;
    p. 156 indrīya for indriya;
    p. 157 pratibhaṃ for prātibhaṃ; gulika for gulikā;
    p. 158 pratibhāt for prātibhāt; jāmbu-dvīpa for jambu-dvīpa; (p. 158 et passim) pada for pāda;
    p. 160 svabhāvikam for svābhāvikam;
    p. 163 Spandakārikas for Spandakārikās;
    p. 164 (et passim) Yoga Vasiṣṭha for Yogavāsiṣṭha;
    p. 165 (Hindi) dhāranā for dhāraṇā.
  30. Thus we find YS for “Yoga Sūtras” but PSū for “Pāśupata Sūtra” etc. Rām for Rāmāyaṇa is particularly infelicitous: see e.g. p.65 “… in the Rām’s account…”.
  31. White’s use of the hybrid Hindi-Sanskrit name Nāth Siddha in his other works points to his being in the lineage of scholarship on the Nāth tradition started by V.V. Ramana Sastri in The Cultural Heritage of India (1937). Sastri’s entry on “The doctrinal culture and tradition of the Siddhas” (Sastri 1956 in the bibliography below—the reference is to the second, enlarged edition) has influenced much subsequent scholarship on the broader Nāth and Siddha traditions, including that of Shashibhushan Dasgupta, Mircea Eliade and Kalyani Mallik, as well as White. Sastri based much of his entry on the Tamil Siddha tradition and this southern slant, together with the Bengali focus of Dasgupta and Mallik, has led to an undue emphasis being placed on notions such as the siddhadeha, alchemy (on which see note 32) and sinisterness in subsequent analyses of yoga. On this academic lineage and its rôle in mistaken understandings of the yogic body, see Ondračka 2007. As far as I am aware, Sastri coined the term Nāth(a) Siddha, which is not found in the writings of the Nāth order itself until the second half of the twentieth century.
  32. White’s conflation of Paścimāmnāya sādhakas, who came to be known as Yogīs and who included alchemists among their number, with practitioners of haṭhayoga is partly responsible (see also note 31) for the thesis of his 1996 monograph The Alchemical Body, namely that “if they were not one and the same people, [haṭhayogins and alchemists] were at least closely linked in their practice” (ibid.:10). The alchemical and haṭhayogic traditions do share some esoteric vocabulary, but this shared terminology is simply drawn from a pool accessed by adepts of a broad range of traditions, from the Kashmiri exegetes of Śaivism to the Sants. Many works on haṭhayoga show it to be quite distinct from alchemy. In the Dattātreyayogaśāstra alchemy is said to be one of the obstacles to success in the practice of yoga (52). Elsewhere haṭhayogic practices are said (with tongue in cheek) to bestow the siddhis of alchemy: in the Śivasaṃhitā (3.61) and Dattātreyayogaśāstra (99) it is said that the yogi in the ghaṭāvasthā can turn objects into gold by smearing them with his faeces and urine. One might expect to find the closest links between haṭhayogic and alchemical practice in a Kaula work such as the Khecarīvidyā, but that text in fact trumps alchemical practice with the yogic technique of aṅgamardana, in which the body is subjected to techniques paralleled, in both name and manner, by those of alchemy (Khecarīvidyā 2.72-79, on which see Mallinson 2007:220 n.328).
  33. On which see Mallinson 2011b.
  34. The first text of the Gorakṣa tradition to teach a haṭhayoga called as such is the c. fourteenth-century Yogabīja. For more details on the development of early haṭhayoga see my “Śāktism and Haṭhayoga” paper, which is to be published in a volume on the Śākta Traditions by Routledge Curzon and a draft of which is available on my page.
  35. The Śivasaṃhitā is a product of a Śaiva sect in the Śrīvidyā tradition of the maṭha at Shringeri now said to have been established by Śaṅkara; the Dattātreyayogaśāstra is the product of a Vaiṣṇava school of yogis. For more details on the affiliations of these texts see Mallinson forthcoming. Both these orders came together in the originally loose-knit Daśanāmī Saṃnyāsī order. Shringeri’s Śaiva orientation came to dominate the order, but several traces of the original Vaiṣṇava orientation of many of its members can be seen in both historical sources and among the Daśanāmīs today. Thus, for example, many of the Giris and Puris (two of the Daśanāmīs’ ten names) portrayed fighting each other in front of Akbar in a late 16th-century miniature painted to illustrate an Akbar Nāma manuscript now housed in the Victoria and Albert Museum (IS.2:61-1896 and IS.2:62-1896, viewable online at and abattlebetween/) sport Vaiṣṇava ūrdhvapuṇḍra tilaks; the formula with which Daśanāmīs invariably greet one another, oṃ namo nārāyaṇ, is a contraction of the aṣṭākṣara mantra oṃ namo nārāyaṇāya used as a mantra and salutation by Śrīvaiṣṇavas and other Vaiṣṇava orders; three of the Daśanāmīs’ four pīṭhas, Badrinath, Dwarka and Puri are archetypal Vaiṣṇava dhāmans with no Śaiva connections. I suspect, but have no evidence to prove it, that Shringeri supplanted Rameshwaram, the fourth Vaiṣṇava dhāman, in the Daśanāmī scheme. For more on the development of the Daśanāmī and Rāmānandī saṃpradāyas, see Mallinson 2013.
  36. Even the ascription of the Haṭhapradīpikā to the Nāths, or at least their forerunners, is uncertain. It may be representative of a broader siddha tradition. In its list of siddhas (1.8) it includes Allāma Prabhu, who in contemporaneous hagiography was an opponent of Gorakhnāth (Śūnyasaṃpādane Vol. 5).
  37. Siddhasiddhāntapaddhati 6.79-91.
  38. The Nāths have very recently started to capitalise on their haṭhayogic heritage, real or not: at the 2013 Allahabad Kumbh Melā daily displays of complex āsanas were, for the first time, carried out by Nāth Yogīs on a stage at the front of their camp. But I remain unable to find Nāth practitioners of the defining techniques of haṭhayoga, its mudrās.
  39. Gold 1992:51.
  40. In the verse attributed to Kabīr used by White as the title of his 2009 article (see page 3), the poet is making a pun on two meanings of the word yogī: at the time of the verse’s composition, yogī was a generic term for the members of various ascetic orders (the verse is directed specifically at Daśanāmī Saṃnyāsīs) for whom the practice of yoga, as implied by the verse, was not essential; at the same time, yogī preserved its older meaning of “a practitioner of yoga”.
  41. Bhaktamāl 38 (p.265).
  42. Chappay 39 from Nābhā’s Bhaktamāl, which is discussed and translated by Burchett (2012:91; see also ibid.:93-95). On Dvārkādās see ibid.:98. Another of Payohārī Jī’s disciples, Agradās, is remembered as a rasik bhakta, but amongst his disciples were the ascetic yogis Bhagvān Jī and Puraṇ (ibid.:100).
  43. Horstmann 1991.
  44. A conflict between yoga and bhakti is drawn attention to by Tulsīdās, but it is specifically the yoga of the Nāths that he is referring to: Gorakh jagāyo jog, bhakti bhagāyo log, “Gorakh awakened yoga and drove bhakti away from the people” (Kavitāvalī Uttarakāṇḍa 7.84; see Burchett 2012:289). It may be that Tulsīdās was opposed to yoga in general, but the descriptions of yoga-practising bhaktas in the Bhaktamālā show that this was not the prevalent Vaiṣṇava attitude during his time. It was not until the early 19th century that Tulsīdās’s Rāmcaritmānas became a key text for the Rāmānandīs (Paramasivan 2010:12).
  45. Burchett 2012:102-107. The Daśanāmī Saṃnyāsīs’ adoption of a Śaiva orientation and advaita vedānta have led to their practice of bhakti being overlooked, but they too have long practised both devotion and yoga (Mallinson 2013).
  46. The first mentions of an order of yogis divided into twelve panths are found in Sikh sources dating to the beginning of the seventeenth century. At Guru Granth Sāhib 34.3 jogīs are said to be divided into twelve (and the saṃnyāsīs into ten—“six [and] four”—this is also our earliest reference to the ten-fold division of the Saṃnyāsīs, from which they became known as Daśanāmī; see Clark 2006:174). In the Sidh Goṣṭh, a dialogue between Nānak and some siddhas which is part of the Guru Granth Sāhib, the siddhas include Loharīpā, who is said to be of the lineage of Gorakh (Sidh Goṣṭh 7.4), and they are members of one of twelve schools of yogīs (Sidh Goṣṭh 9.2, 34.3). In a vār written by Bhāī Gurdās in 1604 it is said that saṃnyāsīs have ten names and the jogīs twelve panths (Vārāṅ Bhāī Gurdās 8.13). The Dabistān, which was written in the first half of the seventeenth century, is our first source to list the twelve panths (Shea & Troyer 1843: Vol. 2, p. 130). It is also in the seventeenth century that a twelve-fold Yogī panth is first mentioned in Nepal: Unbescheid 1980:175-177, 197; Locke 1980:436. On the history of the Nāth saṃpradāya, see Mallinson 2011c.
  47. Bouillier 2008:25-26. On whether there existed a supreme authority ruling over the Nāths prior to the Yogī Mahāsabhā, Bouillier (ibid.:50) writes “Je l’ignore”.
  48. Cf. Bronkhorst 1998:76 in The Two Sources of Indian Asceticism: “Non-Vedic asceticism, as we have come to know it in the preceding pages, has quite different aims. It aims primarily at inaction, with the ultimate goal of liberation from the effects of one’s actions. These are hardly ideals which easily give rise to stories, as do the aims of the Vedic ascetic.”
  49. On the Amanaska, see Sinister Yogis p.46 and p.264 n.31, White 1996:141, White 2003:81; on the Yogabīja, see White 1996:100, 141, 255.
  50. The picture is also reproduced in the book itself, on page 221.
  51. See for example Pinch 2006:24,227,228.
  52. Our only evidence of militarised Nāths, or more correctly of militarised Yogī forerunners of the Nāths, is from the 16th century, in Varthema’s stories of the exploits of the “King of the Ioghe” and his troop of warriors who were active along India’s western coast in the 16th century (Badger 1863:111-113, 273-274) and the references to armies of jogīs in the Padmāvati (Jogīkhaṇḍ) and Kanhāvat (p. 342), although the latter two literary references are perhaps less reliable in terms of sectarian identification than Varthema’s eyewitness account, in which certain features identify the ascetics as forerunners of the Nāths. After the 16th century and the confederation of the Yogīs into twelve panths there are no references to their being militarised. Unlike the Daśanāmīs and Rāmānandīs, the Nāths have never been organised into akhāṛās nor displayed military might in processions at Kumbh Melās. See also Mallinson 2013 n. 46 and Mallinson 2011c:418-419.
  53. This edition was read at All Souls College, Oxford in 2012 with Professor Alexis Sanderson, Dr Péter-Dániel Szántá, Dr Jason Birch and Dr Andrea Acri, whom I thank for their many valuable comments and emendations.

ISSUE #004

On Modern Yoga Research
Image: On Modern Yoga Research

Read and reference at your own pace.
Download this issue of Tarka as a PDF to access the full-length, unabridged articles.

Embodied Philosophy Forum

A Private Facebook Community