The Philosophy of Classical Yoga

I

The Concept of God (īśvara)

The ontology of Classical Yoga, or kriyā-yoga, has three major foci, viz. īśvara, puruṣa and prakṛti. These are deemed irreducible ontic ultimates. The most distinctive feature of the ontology of Patañjali’s school of thought and, I wish to contend, of any form of hindu Yoga, is the concept of ‘the Lord’ or īśvara.

The word īśvara is a derivative of the verbal root (‘to rule’), current already at the time of the ancient vedic saṃhitās. Synonyms are īś, īśa and īśana, īśvara being the more prevalent form in later periods. It conveys the notion of a highest personal god, at times endowed with certain anthropomorphic characteristics but never totally divorced from the concept of the impersonal absolute, the brahman, of philosophical discourse. The term īśvara is ultimately bound up with the history of theism in India.

Repeated attempts have been made in the past to trace the evolution of this crucial religio-philosophical concept. One of the first scholars to apply himself to the study of the history of theism was M. Müller. He distinguished three principal stages, all of which can be evidenced still in the vedic age; they are (1) Polytheism, (2) Henotheism (or Kathenotheism), (3a) Monotheism and (3b) Pantheism.

Thus on the most archaic level M. Müller (1916⁴) envisaged a kind of theological pluralism in which the thirty-three known gods of the rgvedic pantheon were regarded as embodiments or abstractions of natural phenomena. On the basis of this diffuse conceptual stage the need arose for a unification of the multiple devas populating the heavens. According to M. Müller, the notion of the viśve-devas (‘all-gods’) was a gambit in this direction. Certain gods were identified with each other or coupled together, as in the case of Mitra-Varuna and Agni-Soma, etc. On the next stage, in M. Müller’s evolutionary scheme, a single god was invoked under the temporary forgetfulness of all other gods. For this phenomenon he devised the term Henotheism (also: Kathenotheism). From then on the development proceeded in a bifurcate line. On the one hand it gave rise to monotheistic conceptions and on the other hand to Pantheism with its impersonal absolute.

The entire problem was renewedly investigated by H. Jacobi (1923). In principle accepting M. Müller’s (1916⁴) classificatory model, he modified somewhat his formulation of the nature of Henotheism in that he preferred to regard it not so much as a direct pre-stage to Monotheism, but as a rejection of the gods as totally independent entities and thus as a preparatory stage for the development of the concept of an impersonal quintessence (or brahman) of the manifest world.

The concept of brahman (neutr.) was of first-rate importance in the religious and philosophical speculations of the post-vedic period, and, as S. Dasgupta (1963⁵, I, 20) remarked, it ‘has been the highest glory for the Vedānta philosophy of later days’. In one sense it is antipodal to the idea of īśvara, yet in another sense it can be said to complement it, or perhaps even partially define it. For in the formulation of the notion of a personal god the idea of the omnipresent and omni-temporal ground of being is never quite lost sight of.

The idea of a personal deity is anticipated in the ṛgvedic conception of the ‘unknown god’ (M. Müller’s phrase) eulogised in X.121, as also in the conception of Prajāpati, Dhātṛ, Viśvakarman, Tvaṣṭṛ and Puruṣa (see X.90). Whether or not one interprets these, according to some preconceived evolutionist system, as the culmination of a primitive polytheist medley, it is clear that by the time the bulk of the Mahābhārata had been composed the concept of īśvara was firmly lodged in the religious sector of Indian culture. The theism of the epic is largely analogous to that of the metric Upaniṣads, such as the Śvetāśvatāra- and the Kaṭha-Upaniṣad and not least the Bhagavad-Gītā. This highlights an interesting point, namely it brings out the close relation which exists between the concept of īśvara, Sāṃkhya onto-logical ideas and yogic practice. Their joint occurrence in the post-buddhist period is certainly remarkable and calls for an explanation.

B. Kumarappa (1934, 3), in a slightly different context, suggested that theological speculation was originally triggered off by the primary question ‘Whence this universe?’. He thus linked up theism with cosmological and etiological considerations, which would seem to have the supportive evidence of the many creation theories in the Upaniṣads. But perhaps this is merely half the full answer. A different solution to this problem is possible if one places proper emphasis on the fact that it is not only the more speculative Sāṃkhya which is bound up with the īśvara concept, but also the age-old experimental tradition of Yoga. Basing myself on R. Otto’s (1959) hypothesis of an innate capacity in man for numinous experiencing, I wish to propose that īśvara is essentially an experimental construct arrived at primarily on the basis of yogic self-absorption rather than pure theological ratiocination. In this respect it can be aligned with the other ontological categories of pre-classical Sāṃkhya and Yoga which, as I will show, are most appropriately understood as being phenomenological distillations of meditative-enstatic experiences. However, I hasten to emphasise that this line of argumentation in no way implies either an affirmation or a denial of the objective reference of any of these categories of experience.

It has not always been appreciated that theism is woven into the very fabric of hindu Yoga. Thus, in R. Garbe’s (1894) opinion, Yoga is a theistic reinterpretation of the nirīśvara (atheistic) tradition of ancient Sāṃkhya. He speculated (p. 50) that this acceptance of īśvara into Yoga was the likely result of an effort to make Yoga more acceptable to the popular strata of society. H. Oldenberg (1915, 281) probed further: ‘Did this belief originally pertain to Yoga as an essential element? Have Sāṃkhya and Yoga always been differentiated in the way the epic has it and as they are differentiated in their classical forms: as an atheistic and a theistic system respectively? This seems doubtful. The practice of Yoga obviously does not necessarily presuppose the notion of god [. . .]. Visible proof that a system greatly suffused with yogic elements could nonetheless reject the belief in god is supplied by the doctrine [. . .] of the Buddha.’

This stance has been challenged early on in the controversy by H. Jacobi (1923, 39), who wrote: ‘This assertion of īśvara has been interpreted as a concession of Yoga to Brahmanism, which is surely wrong; rather one should admire the audacity and the courage of a school of philosophy which, in the face of the prevalent atheism in philosophical and orthodox circles, dared to put forward the existence of īśvara [. . .] as one of its doctrinal axioms.’ H. Jacobi thus reaffirmed L. von Schroeder’s (1887, 687) contention that ‘Yoga has a distinct theistic character’.

This has been definitively confirmed by more recent research into the pre-classical configurations of the Sāṃkhya school of thought. In an outstanding contribution, K. B. R. Rao (1966) has conclusively demonstrated the intrinsic theistic nature of the pre-classical Sāṃkhya schools. His comprehensive study fully corroborates and consolidates F. Edgerton’s (1924, 8) findings: ‘Where, then, do we find that original atheistic view expressed? I believe: nowhere. A study of the epic and other early materials [. . .] has convinced me that there is not a single passage in which disbelief in Brahman or God is attributed to Sāṃkhya.’

H. Jacobi (1923) saw a connection between the employment of austerities (tapas) and the belief in īśvara. He pointed out that not infrequently the declared purpose of the fearful ascetic practices was to get the attention of a particular deity who, impressed and gratified with the tapasvin’s self-inflicted hardship and unflinching endurance, would bestow a boon on him. He mentioned in passing that in such a context the deity was generally known as varada or ‘bestower of the boon’. He speculated (p. 29): ‘Tor the popular conception at least, the grace of the deity was a necessary precondition for the recompense of ascetic exertion. It seems but natural that Yoga should adopt the recognition of īśvara into its system.’

This view is reiterated in many modern studies, especially on the history of religions. Thus N. Smart (1968, 30), a representative proponent of this misconception, wrote: ‘. . . Yoga has borrowed a concept from popular religion and put it to a special use.’ As he asserted elsewhere (1971, 163), Yoga is essentially an atheistic system. No reasons were supplied. At least H. Jacobi (1923) offered some kind of explanation even though it is unacceptable. For what his interpretation amounts to is the reduction of the conception of a personal god to one of two actors in a process of bargaining: the ascetic excels himself and is rewarded or ‘paid off’ by the deity. I do not contest that this may be exactly the essence of many of the ascetic ‘deals’ recorded in the epic. But I find it unsound reasoning to take this as a historical prelude to the act of grace (prasāda) spoken of in later Yoga. I prefer to understand such legends as folkloristic interpretations of a phenomenon which could well be a parameter of mystical experiencing: the ultimate crossing of the threshold of phenomenal existence interpreted as a transcendental act which appears to be initiated as it were from ‘outside’ or ‘above’.

The idea implicit in H. Jacobi’s (1923) suggestion that Patañjali in a way made a compromise to placate the orthodoxy is preposterous. Imputing to the famous Yoga teacher such hypocrisy, it is hardly surprising that his precise philosophical position has never been appraised adequately.

Less objectionable but similarly unconvincing is M. Müller’s (1916², 326) psychological explanation. Rejecting the historical argument according to which Patañjali merely sought to appease the orthodox brāhmanas, M. Müller instead suggested that it was the natural human craving for a first cause which led Patañjali to the postulation of īśvara. If this were correct one would expect īśvara to have at least one definite cosmological function; yet ‘the lord’ is neither the creator nor sustainer or destroyer of the universe. The ‘first cause’ of which M. Müller spoke is, in Patañjali’s system, the world ground or prakṛti, the eternally creative matrix of the manifest world.

Against the above historical and psychological explanations of the concept of īśvara, I wish to propose that its origins lie in the realm of yogic experiencing itself. This is also M. Eliade’s (1973³, 75) conclusion: ‘Patañjali nevertheless had to introduce Īśvara into Yoga, for īśvara was, so to speak, an experiential datum . . .’. This of course does not imply that Patañjali’s formulation of the concept is a creation ex nihilo. It is obvious from a perusal of the Mahābhārata, especially certain portions of the twelfth parvan, that the conceptualisation of īśvara in Classical Yoga has its epic antecedents.

Philosophically the most important treatment of the theistic component in epic Yoga is to be found in section XII.296¹ of the critical edition of the Mahābhārata. Here hiranyagarbha-yoga² is dealt with, which K. B. R. Rao (1966, 278) wrongly identified as the philosophy of the epic Yoga system par excellence. However, this slip does not detract from the general merit of his acute analysis of this particular branch of Yoga. On the basis of P. M. Modi’s (1932) earlier work, he succeeded in achieving a complete reinterpretation of the above passage, which has been lamentably misconstrued by F. Edgerton (1965) and others. He managed to reconstruct a good deal of the philosophy sketched in these extremely difficult and obscure verses.

Accepting, in principle, the general epic theories about the twenty-three evolutes of the unitary world-ground, the hiranyagarbha school of Yoga introduced the noteworthy distinction between the Self which has recovered its innate enlightenment, viz. the so-called buddhyamāna, and the ever-enlightened buddha or prabuddha. In comparison with the latter, i.e. god, the enlightened Self is said to be abuddhimān (see vs. 17). Thus there is no simple identification of the twenty-fifth tattva, viz. buddhyamāna, with the twenty-sixth, which is the supreme godhead. The latter principle is also referred to as īśvara, mahā-ātman and avyakta-brahman. The buddhyamāna is also called puruṣa and buddha (which confusingly enough is also applied to the twenty-sixth tattva). The twenty-fourth principle, which is the insentient world-ground, is known by the name of prakṛti, abuddha, avyakta and apratibuddha.

It is said of the buddhyamāna (see vs. 2) that it creates, upholds and withdraws the primary-constituents (guna) of the world-ground and that it ‘knows’ or apperceives the world-ground (see vs. 3) whilst itself being nirguna (see vs. 4) and hence ‘unknown’ by the avyakta. On the other hand, the buddhyamāna does not apperceive the lord (see vs. 6), who is pure, incomprehensible, eternal and always apperceiving (see vs. 7). This mahā-ātman or great being permeates both the visible and the invisible (see vs. 8). When the buddhyamāna or Self identifies itself with something that is external to its being, it is known as avyakta-locana (see vs. 10). Taking his cue from XII.296.18 (= XII.284.18 crit. ed.), K. B. R. Rao (1966, 282) interpreted this term as ‘wearing the spectacles of prakṛti’ or ‘seeing through the avyakta’ by means of the organ of cognition (which is buddhi) rather than understanding this interesting compound in the plain sense of ‘seeing the avyakta’.

The goal of this Yoga is naturally also quite different from that enunciated in the contemporaneous Sāṃkhya and Pāñcarātra schools, which advocate a merger of the phenomenal self with the transcendental Self. This difference is evident from such phrases as buddhatva (XXI.296.11), kevala-dharma (vs. 12) or kevalena samāgamya (vs. 13). These appear to imply that the buddhyamāna attains to the ‘estate’ of the twenty-sixth principle without becoming identical with it. In other words, īśvara always remains transcendent (para). He never becomes involved with any of the lower tattvas. Thus emancipation can be said to be a condition of the buddhyamāna qua the buddhyamāna in the ‘company’ (samiti) of the lord (see XII.296. 27 ff.).

The metaphysics of this prominent school of Yoga in epic times seemingly provided the paradigm for the peculiar ontology of Classical Yoga. This was first pointed out by P. M. Modi (1932, 81): ‘The idea of God in the Yoga System was not arrived at by superimposing it on an atheistic Sāṃkhya System with twenty-five principles, but by distinguishing the Jīva from God on practical grounds.’ This is endorsed by K. B. R. Rao (1966, 290): ‘Probably the Epic Yoga lays the inchoate foundation for the classical Yoga conception of a detached īśvara.’ However, he felt compelled to remark (p. 291) that the conception of īśvara in the ancient hiranyagarbha-yoga is ‘utterly naive and simple’, since it depicts god as ‘a motionless and frigid witness’ who is not even interested in the yogin’s struggle for emancipation. He also deemed the more activist conception of god as expressed in the Yoga-Bhāsya (1.25) a positive advance on this view. Evidently K. B. R. Rao’s criticism is somewhat biased.

Although no mention is made in the relevant epic passage of the lord’s soteriological function, one must nevertheless ask oneself why a need should have been felt to philosophically recognise the superlative status of īśvara if this concept would not somehow have had a compelling experiential basis. This line of argumentation would seem to be supported by the strictly pragmatic approach of Yoga, with its emphasis on experiment and personal verification. Nor is the absence of any reference in the above passage to the idea of grace or prasāda, which looms large in other contexts, a positive proof of its irrelevance in the yogic process as envisaged in hiranyagarbha-yoga.

A different hypothesis about the historical precursor of Classical Yoga was put forward by E. H. Johnston (1937). He proposed that ‘the Sāṃkhya side of Patañjali’s