In FEATURES by Dr James MALLINSON0 Comments

James Mallinson is a scholar of the texts and practices of traditional yoga and yogis in India and an Associate at the Oriental Institute, Oxford University.
Published in Nāmarūpa with the kind permission of Brill, Leiden, Netherlands.
“Haţha Yoga,” in: K.A. Jacobsen et al., eds., Brill’s Encyclopedia of Hinduism, vol. III, Leiden, 2011, 770-781. (

The word haţha (lit. force) denotes a system of physical techniques supplementary to yoga more broadly conceived; Haţha Yoga is yoga that uses the techniques of haţha. Haţha Yoga is first referred to by name in Sanskrit texts dating to around the 11th century CE, but some of its techniques can be traced back at least a thousand years earlier, to the epics and the Pali canon. Why these techniques were called haţha is not stated in the texts that teach them, but it seems likely that, originally at least, they were called thus because, like tapas (asceticism), with which they were associated, they were difficult and forced their results to happen.
In this article, only those aspects of Haţha Yoga that set it apart from other techniques of yoga shall be discussed in detail. Important principles and practices that are shared with other methods of yoga, such as subtle physiology, dhāraņā (fixation [of the elements]), and nādānusandhāna (concentration on the [internal] sound), are not analyzed. Furthermore, it should also be noted that the modern “Haţha Yoga” taught by B.K.S. Iyengar is not the same as traditional Haţha Yoga.

In its eHatha Yoga.fig.1arliest formulations, Haţha was used to raise and conserve the physical essence of life, identified in men as bindu (semen), which is otherwise constantly dripping downward from a store in the head and being expended. (The female equivalent, mentioned only occasionally in our sources, is rajas, menstrual fluid.) The preservation and sublimation of semen was associated with tapas (asceticism) from at least the time of the epics, and some of the techniques of early Haţha Yoga are likely to have developed as part of ascetic practice. The techniques of early Haţha Yoga work in two ways: mechanically, in practices such as viparītakaraņī, “the reverser,” in which by standing on one’s head one uses gravity to keep bindu in the head; or by making the breath enter the central channel of the body, which runs from the base of the spine to the top of the head, thereby forcing bindu upward. In later formulations of Haţha Yoga, the Kaula system of the visualization of the serpent goddess Kuņďalinī rising as kuņďalinī energy through a system of cakras, usually six or seven, is overlaid onto the bindu- oriented system. The same techniques, together with some specifically kuņďalinī-oriented ones, are said to effect kuņďalinī’s rise up the central channel (which is called the sušumnā in these traditions) to a store of amŗta (the nectar of immortality) situated in the head, with which kuņďalinī then floods the body, rejuvenating it and rendering it immortal.
The aims and results of Haţha Yoga are the same as those of other varieties of yoga practice: siddhis (both mundane benefits and magical powers) and mukti (liberation), the latter often understood as being attained in a body immortalized by Haţha Yoga practices. In keeping with the physical orientation of Haţha Yoga practices, its siddhis are predominantly physical, ranging from the loss of wrinkles and grey hair to divine sight or the ability to levitate. In common with earlier formulations of yoga, in particular Kaula ones, the techniques of Haţha Yoga can be used to effect kālavañcana (cheating death), utkrānti (yogic suicide), or parakāyapraveśa (entering another’s body).

As in Patañjali’s Yogasūtra, siddhis are usually said to be a hindrance to or distraction from Haţha Yoga’s ultimate aim – liberation – but in some Kaula- influenced texts, the pursuit of specific siddhis through specific techniques is taught (Mallinson, 2011a).

Haţha Yoga is sometimes distinguished from other types of yoga, in particular mantrayoga, layayoga, and rājayoga. Swami Vivekananda (1863–1902) identified Rāja Yoga with the “mental” yoga taught in Patañjali’s Yogasūtra and said that other yogas, in particular Haţha, or “physical,” Yoga, were inferior to it (Michelis, 2004, 178–180). This understanding of Rāja Yoga has become widespread, but it is not what it means in Sanskrit texts, wherein it is simply the ultimate aim of yoga (which is usually samādhi) and not a means of attaining it. There is no opposition between Patañjali’s yoga and the techniques of Haţha Yoga in early Haţha Yoga texts; the practices of Haţha Yoga are supplementary to those of ašţāńgayoga (eightfold yoga, i.e. Pātañjala Yoga). (The Vivekamārtaņďa, in keeping with its Śaiva Mantramārga tradition, teaches a sixfold yoga without Patañjali’s yama and niyama [ethical and behavioral observances] but does not call it Haţha.) By the 17th century, Haţha Yoga had become an integral part of most formulations of yoga, including those based on Patañjali’s Yogasūtra, as evinced by the creation of a corpus of Yoga Upanišads, whose texts borrowed widely from works that teach Haţha Yoga (Bouy, 1994). The modern yoga widely practiced around the world today is derivative of Haţha Yoga, although it places a greater emphasis on āsana (physical postures) than is found in traditional Haţha Yoga and includes under the āsana rubric innovations from Indian and foreign sources (Singleton, 2010) that are not to be found in traditional teachings on Haţha Yoga.

TextHatha Yoga.fig.3
For the early period of Haţha Yoga prior to the composition of the Haţhapradīpikā (which is often called the Haţhayogapradīpikā in secondary literature; c. 1450 CE), Sanskrit texts are our only sources for its practice. (Two vernacular sources that are said to predate the Haţhapradīpikā, the Marathi Jñāneśvarī and the Tamil Tirumantiram, do describe Haţha Yoga techniques, but without further text-critical studies of these works, we cannot be sure of the age of the passages that include those teachings.) A handful of travelers’ descriptions of yoga practice from this period do survive, but they do not provide any details of specific Haţha Yoga techniques. The same is true of later travelers’ reports, which, while useful for determining the social history of yoga and yogīs, add little to our understanding of Haţha Yoga.
The earliest text to teach a systematized HaţhaYoga and call it such is the Dattātreyayogaśāstra, which was probably composed in the 13th century CE. In its section on Haţha Yoga, after teaching a traditional eightfold yoga that it attributes to Yājñavalkya and others, it describes ten Haţha Yoga practices that it says were undertaken by the ŗši Kapila and other ŗšis in addition to those of Yājñavalkya.

These practices, which will be examined in more detail below, are of the variety that came to be known collectively as mudrās (lit. seals, a variety of physical techniques for controlling vital energies, including kuņďalinī, breath, and bindu) in later Haţha Yoga texts and that constitute the techniques of early Haţha Yoga.

The Dattātreyayogaśāstra teaches the following such mudrās: mahāmudrā, mahābandha, khecarīmudrā, the three bandhas (lit. locks; jālandharabandha, uddiyāņabandha, and mūlabandha), viparītakaraņī, vajrolī, amarolī, and sahajolī. Other texts that predate the Haţhapradīpikā and describe Haţha Yoga mudrās (without teaching Haţha Yoga as such) include the Amŗtasiddhi, which dates to the 11th century CE and teaches mahābandha, mahāmudrā, and mahāvedha; the Vivekamārtaņďa, which is contemporaneous with the Dattātreyayogaśāstra and teaches mahāmudrā, nabhomudrā (i.e. khecarīmudrā), the three bandhas, and viparītakaraņī; the Gorakšaśataka, which is also contemporaneous with the Dattātreyayogaśāstra, teaches the three bandhas and śakticālanīmudrā; and the Khecarīvidyā, which teaches only khecarīmudrā. None of these texts calls its techniques Haţha Yoga. The practices of the Amŗtasiddhi and Dattātreyayogaśāstra are used to raise bindu or prevent it from falling; the mudrās of the Vivekamārtaņďa work on bindu, not kuņďalinī, even though raising it is an important part of the yoga it teaches; and those of the Gorakšaśataka and Khecarīvidyā are used to raise kuņďalinī (they mention bindu only in passing).

The only other texts older than the Haţhapradīpikā to teach Haţha Yoga mudrās are the Śivasaμhitā, Yogabīja, Amaraughaprabodha, and Śārńgadhara- paddhati. Each of these texts, which are likely to postdate all those described above, mentions Haţha Yoga by name. The Śārńgadharapaddhati is an anthology of verses on a wide range of subjects compiled in 1363 CE, which in its description of Haţha Yoga includes the Dattātreyayogaśāstra’s teachings on five mudrās. The Yogabija gives an esoteric definition of the word Haţha that is much repeated in later texts, commentaries, and secondary literature: ha means the sun, ţha means the moon, and Haţha Yoga is their union (yoga). In this context, the sun and moon can be variously interpreted as meaning the upper and lower breaths (prāņa and apāna; Amrţasiddhi 6.11–13), the pińgalā and iďā nādīs, Śakti and Śiva as menstrual blood and semen, or the tip of the tongue and the forehead. The Yogabīja teaches the raising of kuņďalinī by means of breath retention and mudrā.

The HaŢhapradīpikā and Classical HaŢha Yoga
The HaŢhapradīpikā
he haţhapradīpikā was composed by Svātmārāma in the 15th century CE (Bouy, 1994, 85). It is for the most part a compilation: it includes verses from all eight texts mentio
ned above and at least twelve more. The Haţhapradīpikā is the first text that explicitly sets out to teach Haţha Yoga above other methods of yoga. In addition to all the mudrās taught in earlier works, it names āsana (posture), kumbhaka (breath retention), and nādānusandhāna (concentration on the internal sound) as Haţha Yoga’s constituents. These four types of practice are found in most subsequent descriptions of Haţha Yoga. Together with the cleansing practices that also became emblematic of HHatha Yoga.fig.4aţha Yoga and that are taught in the Haţhapradīpikā without specifically being said to constitute part of Haţha Yoga, they constitute what is termed herein “classical Haţha Yoga.” The Haţhapradīpikā became the root text of Haţha Yoga: all subsequent Sanskrit Haţha Yoga anthologies and commentaries refer to it, and most take its definition of the practices of Haţha Yoga to be authoritative.
The Haţhapradīpikā is the first text on yoga to include āsana among the techniques of Haţha Yoga. It teaches 15 āsanas. Eight are varieties of sitting (or lying) positions suitable for meditation, and seven are nonseated positions. The verses describing seated āsanas are taken from a variety of earlier texts, including the Dattātreyayogaśāstra, Vivekamārt- aņďa, Vasišţhasaμhitā, Yogayājñavalkya, and Śivasaμhitā. The descriptions of three of the nonseated āsanas (mayūrāsana, kūrmāsana, and kukkuţāsana; see below) are taken (with metrical modifications) from the Vasišţhasaμhitā but can also be found in earlier Pāñcarātra and Vaikhānasa Samhitās, including the circa 10th century Vimānārcanākalpa, the Pādmasaμhitā, and the Ahirbudhnyāsaμhitā. The verses teaching paścimatānāsana (back stretch posture) are taken (again with metrical modifications) from the Śivasaμhitā. No source text has yet been identified for three of the Haţhapradīpikā’s nonseated āsanas: uttānakurmāsana (upside-down tortoise), dhanurāsana (bow), and matsyendrāsana (Matsyendra’s pose).
The Haţhapradīpikā teaches eight varieties of kumbhaka (breath retention; see below). The verses describing four of these (sūryā, śītalī, bhastrikā, and ujjāyī) are taken from the Gorakšaśataka; source texts have not been identified for the remaining four (sītkārī, bhrāmarī, mūrcchā, and plāvinī).
The Haţhapradīpikā teaches the ten mudrās found in the Dattātreya- yogaśāstra, supplemented by mahāvedha and śakticālanī (it also mentions yonimudrā in passing). Its verses on mudrā are taken from the Dattātreyayogaśāstra, Vivekamārtaņda, Gorakšaśataka, Khecarīvidyā, and Amaraughaprabodha.
No source t
ext has been identified for the Haţhapradīpikā’s verses on nādānusandhāna which are said to have been taught by Gorakša. This practice, which involves putting one’s fingers in one’s ears and listening to a succession of internal sounds (nādas), is said to be a technique of laya (dissolution). As noted above, in earlier texts laya was taught, along with mantra and Haţha, as a distinct method of achieving Rāja Yoga. Svātmārāma used verses from texts that made this distinction to compile the Haţhapradīpikā and emphasized the complementarity of Haţha and Rāja Yoga, but he ignored mantrayoga altogether (the Haţhapradīpikā makes no mention of any mantras) and subsumed within Haţha Yoga many of the techniques of layayoga, including, besides nādānusandhāna, the raising of kuņďalinī, śavāsana, śāmbhavīmudrā (using verses taken from the Amanaskayoga and Candrāvalokana), a nonphysical variety of khecarīmudrā, and meditation on the point between the eyebrows.

The cleansing practices known as šaţ karmāņi, “the six acts,” which became emblematic of Haţha Yoga, are taught in the Haţhapradīpikā in verses that have not been found in earlier works ; in fact, no earlier texts that teach these practices have yet been identified.

The vacuum in the abdomen created by one of the cleansing techniques, nauli, is used in basti and vajrolīmudrā to suck liquids through the anus and penis, respectively. We can thus infer that nauli was practiced at least as early as the 13th century, the time of writing of the Dattātreyayogaśāstra, the first text to teach vajrolīmudrā.
The Principles of HaŢha Yoga
As noted above, in the earliest formulations, the purpose of Haţha Yoga was to raise and preserve bindu, semen, by means of the Haţha Yoga mudrās. Onto its techniques those of layayoga, in particular the raising of kuņďalinī, were subsequently super- imposed. The Haţhapradīpikā says that the purpose of the Haţha Yoga mudrās is to raise kuņďalinī.
This resulted in some conflicts. In the visualizations taught in texts of the Paścimāmnāya lineage of Kaula Śaivism, kuņďalinī, on reaching the store of amŗta located in the head, returns to the ādhāra (base) at the bottom of the spine from which it came, flooding the body with amŗta as it goes. This is what it does as a result of the Haţha Yoga khecarīmudrā taught in the Khecarīvidyā. The purpose of bindu-oriented Haţha Yoga practices is to keep bindu in the head. Thus in the Vivekamārtaņďa, which is the earliest text to synthesize the two paradigms, khecarīmudrā is said to seal the uvula and prevent bindu from falling, but later in the text, the same technique (although not named khecarīmudrā) is said to result in the body being flooded with amŗta. In the Haţhapradīpikā, these verses are found together in the description of khecarīmudrā.
The Haţhapradīpikā’s synthesis of a broad range of practices results in some ingenious assimilations and reinterpretations of earlier practice, a process that continues to this day. Vajrolīmudrā is first taught in the Dattātreya-yogaśāstra as a method of achieving siddhi (success) while not observing the niyamas (restrictions) of yoga. He or she who knows the technique of sucking liquids up the urethra can resorb his bindu or her rajas after sexual intercourse and thereby not suffer from its loss. This technique was hard to assimilate with kuņďalinīyoga, but it was open to a Śākta reinterpretation: verses from the Dattātreyayogaśāstra are used in the Haţhapradīpikā to describe vajrolī, but in addition it is said that absolute success (sarvasiddhi) results from combining bindu and rajas in one’s own body. In contrast, a doggedly celibate Daśanāmī saμnyāsī practitioner of Haţha Yoga living in Gangotri in 2006 reported that vajrolī needs to be mastered in order to resorb semen, in case it is spontaneously ejaculated when kuņďalinī reaches the svādhišţhāna cakra (the cakra located in the genital region).
The techniques of Haţha Yoga, and their development, reflect the ongoing interplay of practice and theory, to which might be added exegesis. The śakticālanī mudrā, for example, originally involved wrapping the tongue in a cloth and tugging it in order to awaken kuņďalinī. Its method was forgotten in certain lineages, but its description was preserved in their texts. Textual corruption obscured the location in the body of where the cloth is to be applied, and now those who teach it, perhaps influenced by the physical location of its benefits (and, of course, their own practical research), say that it is to be done by using nauli, “churning the stomach” (Mallinson, 2011b).
The Haţhapradīpikā’s success ensured that the raising of kuņďalinī became the rationale for many of the practices of Haţha Yoga. With kuņďalinī came a variety of other practices and aims, and when trying to understand the sometimes contradictory notions of Haţha Yoga, it is useful to bear in mind other oppositions parallel to that of bindu and kuņďalinī: mukti (liberation) and siddhis (powers), tapas (asceticism) and bhoga (enjoyment), and haţha (force[d]) and sahaja (natural).

The Practices of Classical HaŢha Yoga

In the Haţhapradīpikā, these techniques are used for nothing more than cleansing the body and balancing its došas (humors) in order to prepare it for the practice of yoga; Svātmārāma adds that some teachers say that prāņāyāma alone will suffice for this purpose. In the Haţharatnāvalī, the cleansing practices are said also to cleanse the six cakras, and some later commentators, seeking to impute a directly soteriological value to all Haţha Yoga practices, say that they directly facilitate various methods of reaching samādhi.

Complicated physical postures are first included among the techniques of Haţha Yoga in the Haţhapradīpikā. The earliest textual reference to nonseated āsanas is in the circa 10th-century Vimānārcanakalpa, a Pāñcarātra work, and it seems likely that the practice of nonseated āsanas developed within a Pāñcarātrika milieu. The 13th-century Matsyendrasaμhitā, the earliest text associated with the Nāth tradition to teach a variety of āsanas, describes 13 seated āsanas, including three named after animals: mayūrāsana (peacock), kukkuţāsana (cock), and kūrmāsana (tortoise).

The use of the word āsana to describe any sort of physical posture appears to have become widespread by the early 14th century, when the Maithili Rasaratnākara used it (along with bandha) as a term to describe positions for sexual intercourse.

The circa 13th-century Dattātreyayogaśāstra and Vivekamārtaņďa both say that there are 84 lākh āsanas, but the former teaches only padmāsana (lotus posture), to which the latter adds siddhāsana (adepts’ posture). Both of these are taught in earlier texts, in particular in Śaiva works, although siddhāsana is known in the latter as svastikāsana (auspicious posture; Goodall, 2004, 349n730; the svastikāsana of later Haţha Yoga works is a slightly different posture).
The Haţhapradīpikā teaches 15 āsanas, of which seven are not seated postures, and marks the beginning of the proliferation and importance of such postures in the practice of yoga. It is also in the Haţhapradīpikā that practices that were originally not conceived of as āsanas first come to be included under its rubric. Thus śavāsana, “the corpse pose,” which is taught as one of the methods of layayoga in the Dattātreyayogaśāstra, becomes an āsana in the Haţhapradīpikā. In later texts Śaiva karaņas (physical practices taught in Śaiva Tantras, which are similar to to the mudrās of Haţha Yoga), Haţha Yoga mudrās, ascetic mortifications, Sufi practices, wrestling exercises, and Western bodybuilding and gymnastic poses all become āsanas. The benefits of āsanas vary accordingly. In the Haţhapradīpikā, āsana is said to lead to steadiness, health, and suppleness (aims not dissimilar from those of modern yoga); certain individual āsanas are said therein and in other texts to awaken kuņďalinī, destroy disease, make the breath enter the central channel, and increase the digestive fire.
The 17th-century Haţharatnāvalī is the first text to teach 84 individual āsanas. Descriptions of 84 āsanas are also found in the 18th-century Āsanayogagrantha (Gharote, 2006, lxiii) and Jogpradīpakā, and the early 19th- century Mahāmandir in Jodhpur (now commonly known as the Udai Mandir) has a frieze depicting 84 āsanas. To this day, traditional yoga practitioners will claim to know 84 āsanas. From the 18th century onward, the number of āsanas taught in texts and in oral traditions has increased beyond 84. The six-chapter Haţhapradīpikā teaches over 100 āsanas, the Śrītattvanidhi describes 122 (Sjoman, 1996), and B.K.S. Iyengar’s Light on Yoga teaches over 200.
he practice of breath control in Haţha Yoga has three sources:
1. an ancient (and not specifically yoga) tradition of regulated breathing, or prānāyāma, that is thought to get rid of karma and physical impurity;
2. a yoga principle that links the breath, the mind, and semen – by stopping one, the others are also stopped; and
3. specific methods of inhalation and exhalation known as kumbhakas (somewhat paradoxically, since kumbhaka in fact means the holding of the breath), which work on both the gross and the subtle bodies.
Many Haţha Yoga works teach a simple prāņāyāma in which the yogī is to inhale through the left nostril, hold the breath, exhale through the right nostril, inhale through the right nostril, hold the breath, and exhale through the left nostril. Different ratios of the lengths of each stage and different numbers of repetitions of the cycle are taught in different texts.

Through this practice, the nāďīs, or subtle channels of the body, are cleansed, enabling the breath and/or kuņďalinī to rise up the central channel and the mind to be stilled. For these latter benefits, the practice of kumbhaka, breath retention, is enjoined.

n the earliest systematized textual treatment, Haţha Yoga is identified with ten practices that assist in the preservation and raising of bindu, the essence of life, either through mechanical means or through the raising of the breath through the central channel. In Haţha Yoga’s classical synthesis in the Haţhapradīpikā, two of these practices, amarolī and sahajolī, were subsumed under the heading of another, vajrolī. To the resulting eight practices, which in the Haţhapradīpikā are all classified as mudrās, were added three more: mahāvedha, śakticālanī, and yonimudrā, making a total of eleven.

The purpose of śakticālanī and yonimudrā has always been to awaken kuņďalinī and make her rise up the central channel. In the Haţhapradīpikā, this is said to be the aim of all mudrās.
1. Mahāmudrā – Press the perineum with the heel of the left foot, stretch out the right foot, and hold it with both hands. Draw up the abdomen, put the chin on the chest, and inhale. After exhaling, swap the position of the feet and repeat the process.
2. Mahāvedha – This mudrā, which makes the breath enter the central channel, is first taught in the Amŗtasiddhi. Its technique therein has the yoga practitioner sitting with the soles of the feet pressed together, and the heels pressing the perineum. In later texts, the practitioners sit with one foot
under the perineum, lift themselves up with their hands, and then drop their perineum onto their heel.
3. Mahābandha – In its earliest Haţha Yoga formulation, in the Amŗta- siddhi, this mudrā is the same as the mūlabandha (on which see below). In later texts, to assume mahābandha, the yoga practitioner, after assuming the mahāmudrā position, puts the outstretched foot onto the opposite thigh.
4. Khecarīmudrā – The tongue is lengthened, so that it can be turned back and inserted into the cavity above the soft palate in order to seal bindu in the head, taste amŗta, or make kuņďalinī rise. In this latter aim, it is a practice similar to śakticālanī mudrā. (For a detailed study of khecarīmudrā, see Mallinson, 2007b.)
5. Jālandharabandha – Place the chin on the chest.
6. Uďďīyānabandha – Draw up the abdomen.
7. Mūlabandha – Contract the perineum region. This and the two preceding techniques are often grouped together as “the three bandhas.” They are to be practiced while holding the breath, and they are also sometimes prescribed, without being named, as adjuncts to other techniques, such as padmāsana.
8. Viparītakaraņī – The yoga prac- titioner inverts himself or herself, usually by assuming either a headstand or a shoulderstand.
9. Vajrolī – After ejaculation, semen or the commingled products of sexual intercourse are drawn upward through the urethra. Vajrolī is often grouped with the practices of sahajolī and amarolī, whose techniques are not always specified and, when they are, are taught differently in different texts. Sahajolī usually involves smearing the body with ash after intercourse; amarolī is the drinking or nasal application of one’s own urine.
10. Śakticālanī – The tongue is wrapped in a cloth and pulled in order to stimulate kuņďalinī (as indicated by the name of the practice: “[themudrā] that stimulates śakti”).
11. Yonimudrā – This practice, which is usually mentioned in passing in texts rather than explicitly taught, is the same as mūlabandha but is specifically oriented toward raising kuņďalinī.
aţha Yoga, like other methods of yoga, can be practiced by all, regardless of sex, caste, class, or creed. Many texts explicitly state that it is practice alone that leads to success. Sectarian affiliation and philosophical inclination are of no importance. The texts of Haţha Yoga, with some exceptions, do not include teachings onHatha Yoga.fig.2 metaphysics or sect-specific practices. To speak of “yoga philosophy” is to miss the point: yoga is a practical discipline aimed at attaining liberation. If duly practiced, it will work, irrespective of the practitioner’s beliefs. The lack of sectarianism in texts on yoga has made them readily adoptable by traditions other than those of their authors. Thus texts composed in a Nāth milieu could be used to compile the later Yoga Upanišads, and others were translated into Persian to satisfy Mughal interest in yoga. Yoga’s lack of sectarianism has also enabled its spread around the world today.
The intended audience of the texts of Haţha Yoga was most probably Brahmin men, as is the nature of Sanskrit texts. There are, however, references to women practitioners within the texts. In some texts, householders as well as renunciates are said to be able to practice Haţha Yoga, but the difficulty of many of its practices and the time required to master them, as well as the nature of their goal, liberation, meant that they were for the most part practiced by members of renunciate orders.


The ancient tradition of the ūrdhvaretās tapasvī (the ascetic whose seed is [turned] upwards), which is closely associated with the practice of yoga in texts such as the Mahābhārata, is likely to be the source of early Haţha Yoga, in which the preservation of bindu is paramount.

This relatively orthodox tradition has survived in ascetic orders such as the Daśanāmī saμnyāsīs and the Rāmānandīs. Onto the bindu-oriented Haţha Yoga was overlaid the layayoga of a Kaula tradition associated with siddhas such as Matsyendra and Gorakša, which came to be known as that of the Nāths. Its members practiced Śaiva magical arts such as alchemy (rasāyana) and the worship of goddesses known as yoginīs as well as kuņďalinīyoga and the other techniques of layayoga. The synthesis of the bindu- and kuņďalinī- oriented paradigms of yoga had its first truly systematic manifestation in Svātmārāma’s Haţhapradīpikā, which was so successful that it became the root text of Haţha Yoga for all traditions. The early Nāth yogīs’ willful adoption of bindu-oriented yoga was paralleled by their formation into a celibate ascetic order despite their origins in the rather less abstemious Kaula Tantrism. Many of today’s better-known schools of Haţha Yoga, such as Swami Satyananda’s Bihar School of Yoga and Swami Sivananda’s Divine Life Society, were established by gurus affiliated, albeit tenuously, with the Daśanāmī saμnyāsī order. The teachings on yoga of three students of T. Krishnamacharya, namely his son T.K.V. Desikachar, K. Pattabhi Jois, and B.K.S. Iyengar, have had the greatest influence on modern yoga. Their lineage, that of Śrīvaišņavism, is closely connected to the lineages of the first text to teach the Haţha Yoga mudrās (the []Dattātreyayogaśāstra) as well as the first texts to teach nonseated āsanas (Pāñcarātra Samhitās such as the Vimānārcanākalpa and Ahirbudhnyasaμhitā, and the Vasisthasaμhitā). Practice of Haţha Yoga among the Nāths is today almost nonexistent (Bouillier, 2008, 128).

Further Reading
Haţha Yoga – the Haţhapradīpikā, Śivasaμhitā, and Gherandasaμhitā – were uncritically edited and translated into English. These texts, arbitrarily selected, have formed the Haţha Yoga canon ever since, and studies of Haţha Yoga have been hindered by this limited view of the tradition.
Since the 1970s, a handful of critical editions of texts that teach the practices of Haţha Yoga have been published. Among the early works, one finds only the Khecarīvidyā and Śivasaμhitā. The Amŗtasiddhi has not been edited. The Vivekamārtaņďa has been edited (as the []Gorakšaśataka – the names of these two texts became confused) from just four of the hundreds of manuscripts available, and those of its earliest recensions were not consulted. The Dattātreyayogaśāstra, Yogabīja, Amaraughaprabodha, and Amaraughaśāsana have been published as transcripts of single codices. A translation of the Gorakšaśataka based on a single manuscript has recently been published.

Critical editions of two works, the Śivasaμhitā and Gherandasaμhitā, have been published with translations but without apparatus in the Yoga Vidya series (see
While guides to the practice of Haţha Yoga are legion, scholarly secondary literature is rare. Exceptions are C. Bouy (1994) on the relationship between Haţha Yoga texts and the Yoga Upanišads; S. Vasudeva (2004), which concentrates on Śaiva tantric yoga but is useful for understanding the context of Haţha Yoga; C. Kiss (2009) on the yoga of the early Nāths; D.G. White (1996) on the alchemist siddha tradition; the many encyclopedic works on Haţha Yoga practices published by the Lonavla institutes; the introduction to the Khecarīvidyā (Mallinson, 2007b); J. Mallinson’s articles on siddhi in Haţha Yoga (2011a) and the Goraksaśataka (2011b); and J. Birch (2011) on the meaning of haţha.
Alter, J., Yoga in Modern India, Princeton, 2004.
Amŗtasiddhi, mss. 1242 and 1243 at the Maharaja Mansingh Pustak Prakash, Jodhpur, n.d.
Avasthi, B., ed., Dattātreyayogaśāstra, Delhi, 1982.
Birch, J., ed., “Amanaskayoga,” BA thesis, University of Sydney, 2006. Birch, J., “The Meaning of Haţha in Early Haţhayoga, JAOS, forthcoming.
Bouillier, V., Itinérance et vie monastique: Les ascétes Nāth Yogīs en Inde contemporaine, Paris, 2008.
Bouy, C., Les Nātha-Yogin et les Upanisads: étude d’histoire de la littérature hindoue, Paris, 1994.
Brhatkhecarīprakāśa, ms. 14575 at the Scindia Oriental Research Institute Library, Ujjain, n.d.
Candrāvalokana, ms. D 4345 at the Government Oriental Manuscripts
Library, Chennai, n.d. Dattātreyayogaśāstra, mss. 1936 and
1937 at the Maharaja Mansingh
Pustak Prakash, Jodhpur, n.d. Dattātreyayogaśāstra, ms. 4107 at the
Oriental Institute Baroda, Vadodara,
Dattātreyayogaśāstra, ms. 6/4–399
at the Prajna Pathashala Mandal,
Mumbai, n.d. Dattātreyayogaśāstra, mss. 4369,
R5579, and 5585 at the Government Oriental Manuscripts Library, Chennai, n.d.
Digambarji, S., & P. Jha, eds., Haţhapradīpikā of Svātmārāma, Lonavla, 1970.
Gharote, M.L., Encyclopaedia of Traditional Asanas, Lonavla, 2007. Gharote, M.L., ed., Jogpradīpakā of
Jayatarāma, Jodhpur, 1999
Gharote, M.L., P. Devnath & V.K. Jha,
eds., Haţharatnāvalī, Lonavla, 2002. Goodall, D., The Parākhyatantra:
A Scripture of the Śaiva Siddhānta,
Pondicherry, 2004.
Iyengar, B.K.S., Light on Yoga, London,
Goraksaśataka, ms. R 7874 at the
Government Oriental Manuscripts Library, Chennai, n.d.
Haţhapradīpikā Siddhāntamuktāvalī, ms. 6756 at Maharaja Mansingh Pustak Prakash, Jodhpur, n.d.
Kiss, C., “Matsyendranātha’s Compendium (Matsyendrasaμhitā): A Critical Edition and Annotated Translation of Matsyendrasaμhitā 1–13 and 55 with Analysis,” diss., Oxford University, 2009.
Maheshananda, S., B.R. Sharma,
G.S. Sahay & R.K. Bodhe, eds., Vasišţhasaμhitā (Yogakāņďa), Lonavla, rev. ed. 2005
Maheshananda, S., B.R. Sharma, G.S. Sahay & R.K. Bodhe, eds., Haţhapradīpikājyotsnā of Brahmānanda, Lonavla, 2002.
Mallik, K., The Siddha Siddhānta Paddhati and Other Works of Nath Yogis, Poona, 1954
Mallinson, J., “Siddhi and Mahāsiddhi in Early Haţhayoga, in: K.A. Jacobsen, ed., Yoga Powers, Leiden, 2011a, 327–344.
Mallinson, J., “The Original
Gorakšaśataka,” in: D.G. White, ed., Yoga in Practice, Princeton, 2011b, 257–272.
Mallinson, J., ed., Śivasaμhitā, New York, 2007a.
Mallinson, J., The Khecarīvidyā of Ādinātha: A Critical Edition and Annotated Translation of an Early Text of Haţha Yoga, London, 2007b
Mallinson, J., ed., Gheraņďasaμhitā, New York, 2004.
Michelis, E. de, A History of Modern Yoga: Patañjali and Western Esotericism, London, 2004.
Mukund Ram Shastri, P., ed., Amaraughaśāsana of Gorakšanātha, KSTS 20, Srinagar, 1918.
Nowotny, F., ed., Das Goraksaśataka, Cologne, 1976. (This text is referred to as the Vivekamārtaņďa in this article; for the reasons why see Mallinson, 2011b.)
Peterson, P., ed., Śārńgadharapaddhati, Bombay, 1888.
Raghunathacakravartin &
S. Madhavacharya, eds., Vimānārcanākalpa, Madras, 1926.
Ramanujacharya, M.D., ed., Ahirbudhnyāsaμhitā, Madras, 1966.
Satyananda, Asana, Pranayama, Bandha and Mudra, Munger, 2003.
Shrivastav, R.L., ed., Yogabīja, Gorakhpur, 1982.
Singleton, M., The Yoga Body, New York, 2010.
Sjoman, N.E., The Yoga Tradition of the Mysore Palace, New Delhi, 1996.
Vasudeva, S.D., “Haμsamiţţhu: ‘Pātañjalayoga is Nonsense,’ ” JIPh 39/2, 2011, 123–145.
Vasudeva, S.D., The Yoga of the Mālinīvijayottaratantra, PIFI 97, Pondicherry, 2004
Vivekamārtaņďa, ms. 4110 at the Oriental Institute Baroda, Vadodara, n.d.
Yogani, Advanced Yoga Practices: Easy Lessons for Ecstatic Living, Morgantown, WV, 2004.
White, D.G., The Alchemical Body, Chicago, 1996.


About the Author


Dr James Mallinson is a scholar of the texts and practices of traditional yoga and yogis in India and an Associate at the oriental Institute , Oxford University.

Leave a Comment