The Hindu Doctrine of Transmigration: Its Origin and Background

in Vedic Cosmology and Ethics

If the inline PDF is not rendering correctly, you can download the PDF file here.

In1 1873 W.D. Whitney, the pioneer of American Indology, called the origin of the doctrine of transmigration “one of the most difficult questions in the religious history of India” (p. 61). Richard Salomon (1982, 410) denotes it as “the single greatest problem of Indological studies” in a review of Karma and Rebirth in Classical Indian Traditions (edited by Wendy Doniger O’Flaherty) in 1980. The latter publication was the outcome of three conferences held in 1976–1978 on the subject. These conferences raised rather than solved problems. I quote from the Introduction of the mentioned book: “Much of our time at the first conference … was devoted to a lively but ultimately vain attempt to define what we meant by karma and rebirth. The unspoken conclusion was that we had a sufficiently strong idea of the parameters of the topic to go ahead and study it, in the hope that perhaps then we would be able to see more clearly precisely what we had studied (rather like the woman who said to Abraham Lincoln, ‘How do I know what I think ’til I hear what I say?’)” (p. xi). Indeed a very practical approach.

The next problem raised by O’Flaherty’s colleagues was the question of “Abstract Theory versus Historical Explanation” (p. xii). After lengthy discussions they decided to follow both approaches. Again a very pragmatic solution. However, O’Flaherty’s survey of the discussions on “The Historical Origins of the Karma Theory” (6 pages) shows that a solution of the problem was hardly reached.

The historical origins were only treated in the first conference. O’Flaherty concludes her survey of the divergent views with the resigned statement: “Rather than looking for one central ‘source’ which was then embroidered by ‘secondary influences’ like a river fed by tributary streams, it would be better to picture the intellectual fountainhead of ancient India as a watershed consisting of many streams—each one an incalculably archaic source of contributing doctrines—Vedic, Ājīvaka [i.e. materialistic], Jaina, Dravidian, and tribal” (p. xviii). This metaphor actually amounts to the conclusion: “God may know what is the origin.”

Then Wendy Doniger O’Flaherty instigated the American Herman W. Tull to publish his thesis of 1985 in 1989 in a series edited by her, under the title The Vedic Origins of Karma. After reading this book my conclusion was that more than a century after Whitney’s statement the problems still were not satisfactorily solved.

I will not waste time with theoretical definitions. Rebirth or transmigration (Sanskrit saṁsāra) belongs together with karman (the deeds which cause this rebirth and determine its nature) and with mokṣa (the release from the cycle of rebirths) to one complex of concepts which mostly are studied together. So I cannot confine myself to the origin of rebirth as an isolated phenomenon.

The origin and background of this complex have raised several questions. It will be clear that I cannot answer all of them. Was the doctrine of karman originally a theory of causality which explained how every action has results? Did it function as a theodicy, an explanation of the evil in this world? Why did pessimism about life on earth arise, whereas originally the Vedic Indians liked this life? And above all: how did one arrive at the idea that man would return on earth? Many Indologists have regarded the repeated return of sun and moon as the basis of rebirth.2 However, this phenomenon is too universal. The typically Indian concept of cyclic time and of cyclic mundane periods (the yugas) is later than the doctrine of rebirth and therefore cannot serve as its starting point.

What have been the opinions of Indologists during the last thirty-five years?

In Gonda’s handbook (1960, 207) we find an incoherent enumeration of possible origins, introduced with the statement “Über die Ursprünge dieser für die ganze Folgezeit äußerst wichtigen Lehre lassen sich nur Mutmaßungen äußern” and concluded with “alle diese Faktoren haben zu ihrem Aufkommen und ihrer Verbreitung ohne Zweifel das Ihrige beigetragen. Vermutungen über nicht-arischen Ursprung … sind spekulativ.” In the second edition (19782, 207) the formulation of the problem was hardly modified.

In 1980 O’Flaherty, as we have seen, chose a more attractive phrasing of the problem without adding anything new. In the eleven pages of the rather unsatisfactory bibliography some important publications (especially about the origin) are missing, e.g. Paul Horsch (1971).

According to Horsch the doctrine developed out of Vedic thought, i.e. from the ideas of the Aryans who invaded India somewhere in the second millennium BCE.

In the same year 1971 Hyla Stunz Converse obtained her doctorate at Columbia University with a voluminous, but controversial and not completely satisfactory thesis in which everything new, creative and interesting was attributed to non-Aryans, proto-Dravidians and proto-Jains (three categories which would amount to the same). This thesis was not included in the mentioned bibliography.

Herman Tull, who defended the Vedic origin in 1989, refers to Paul Horsch (who did the same), but is silent on Converse. It is obvious that a real discussion of all the issues is still missing.3 It was a surprise to me to see my guru Gonda quoted in support of the Aryan as well as the non-Aryan origin in the theses of Tull (passim) and Converse (p. 8, n. 1). Gonda was rather cautious in his formulation of the problem of change and continuity in Ancient India and in this connection he acknowledged the process of adaptation that continuously took place, but I am sure that his predilection concerned the continuity and the Vedic origin and that he would have tried to prove it, if he had thought it were possible to do so.4

Since Horsch quoted almost all the available literature I use his article as a starting point. Horsch was primarily interested in the population and culture which would have developed the theory, rather than in the possible causes of the relevant ideas. He rejected the non-Aryan origins or even influences and thought “daß es sich dennoch um eine echte vedische Entwicklung handelt, deren Stufen bis in alle Einzelheiten eruiert werden können” (1971, 100). The non-Aryan influence would be entirely absent and apparently he equated the non-Aryans with primitive tribals. His observation: “Wo die Seelenwanderung bei den heutigen Primitivvölkern Indiens auftritt, weist sie deutlich hinduistischen Einfluss auf” (104, n. 9) seems to be based on the following way of reasoning: Since the modern, primitive tribals have adopted the doctrine from Hinduism, the Vedic, Aryan precursors of the Hindus cannot have adopted it from the primitive tribals in the most ancient period.

However, the situation is more complicated than sketched by Horsch. What is the meaning of the term Aryan in a discussion on ethnicity? Were the people who called themselves Aryans belonging to one, homogeneous race during the whole Vedic period? During the last decades (i.a. due to the results of archaeology) several Indologists have assumed that a process of acculturation took place in the most exciting period from about 1500 to 500 BCE. The denomination Aryan still referred to a linguistic and cultural unity, but this unity was no more racial (if it ever had been so) and linguistic borrowings5 seem to have been accompanied with other external influences. Unfortunately, it is difficult to decide how far the acculturation between the original Aryans and the autochthonous population (probably not exclusively consisting of primitive tribals) extended. Anyhow it is evident that developments within Vedism, resulting in the emergence of Hinduism, cannot exclusively be attributed to purely internal developments of the Aryan ideas (at least if Aryan is taken as Indo-European). The antithesis between Vedic and autochthonous is too simplistic, especially if we examine the late Vedic period.

Starting from the Vedic texts (indeed almost the only reliable, extensive material) Horsch tried to trace the source of all later developments in Vedism. A gradual evolution were discernible. However, it is as well possible that gradually external influences penetrated. This means that the philological proof of traceable developments may be less hard than philologists are used to assume. When speaking of external influences I do not follow the strict opposition of Vedic/Aryan and autochthonous/non-Aryan. The real opposition is between traditional, orthodox Vedism characterized by ritualism, and non-orthodox6 movements which need not have been entirely non-Aryan.

Against Herman Tull’s thesis of the Vedic origin of the doctrines under discussion it may be argued that Tull connects everything with Vedic ritualism and that the doctrine of karman (interpreted by him as originating from ritual) almost forms his single concern.7

I shall try to systematize the research about the possible Vedic origins and for the time being leave the point of Aryan and non-Aryan aside. Then I take three lines of approach, which of course cannot always be kept apart in practice:

  1. Terminology

  2. Ideological framework of the terms

  3. Textual evidence

1 Terminology

Terms like karman and mokṣa do occur in Vedic texts before the Upaniṣads and there might be a connection with their use in the classical doctrine of transmigration. The term saṁsāra is first used in rather late Upaniṣads, but there are other words and expressions referring to return and new birth which have induced some scholars to support the theory of Vedic origin.


The term karman is rather vague and denotes: deed, action, activity, ritual action, rite or even as a collective term ritual, ritualism. In the doctrine of transmigration it is supposed that actions have results for life after death on earth. Since according to the Vedic doctrine rituals result in a pleasant stay in heaven after death, the theory of causality implied by Vedic, ritual karman might have been the starting point for the classical doctrine of karman. Here, however, we have to make two critical remarks:

First, the ritual karman, as far as I can see, is exclusively positive. It concerns merits which produce a heavenly continuation of earthly life (rather than a rebirth in heaven). Bad actions and demerits do not belong to the sphere of Vedic ritual.

Second, one may ask whether the ethical aspect of the classical doctrine of karman (through good actions one becomes good, through bad actions bad) has any relation to the ideology of Vedic ritualism. In addition to worship of the gods Vedic rituals produce a comfortable life on earth and in heaven and in this respect are calculated investments rather than ethical, moral achievements.

Herman Tull (1989) tries to solve the problem of the antithesis between ritual and ethical karman by assuming that even in the old Upaniṣads the karman of the doctrine of transmigration was still ritualistic. The ethical karman were introduced afterwards, in the later Upaniṣads. In fact he only makes a chronological shift in order to save the Vedic origin. It is unclear, however, how the doctrine of karman could have been present in old Buddhism and Jainism as a borrowing from Vedism, if even in the older Upaniṣads it was still exclusively associated with Vedic ritual.

As far as the demerits of the negative karman in relation to Vedic ritual are concerned, Tull assumes that bad karman in Vedic literature (including the ancient Upaniṣads) simply refers to bad ritual and its resulting demerits. Bad and good karman would apply to the bad or good performance of the rites.

However, in the old Upaniṣads the qualifications of karman concern adjectives like beautiful, noble, fine, positive and their opposites, i.e. they do not seem to apply to rituals. Moreover, I think that poorly performed Vedic rituals did not exist, since mistakes could be expiated during the performance. The rituals were always good, since the scenario was fixed by the ritual sūtras. In distinction to performing arts the beauty of the sacrificial performance did not count.

The merits of the ritual are obtained on account of the bare fact that one organizes a ritual and knows its implications. These merits are moreover primarily obtained by the institutor of the sacrifice, the so-called Yajamāna, who hardly carries out any action and cannot be blamed for small mistakes made by his priests.

Now in the case of merits and demerits Vedic texts mostly use the terms sukṛta (good action or doing good) and duṣkṛta (bad action or doing wrong). Gonda (1966, 115 ff.) tried to show that sukṛta mainly (be it not exclusively) would denote the good performance of rituals. Tull (1989) elaborates this interpretation and makes it even refer to good karman. The duṣkṛta (the bad action) then would like bad karman denote poorly performed ritual and its resulting demerits and in this connection he refers to Gonda (p. 31). However, Gonda was wise enough not to equate duṣkṛta exclusively with bad ritual, as appears from his note 53 on pp. 126–128. If duṣkṛta is not to be equated with poorly performed ritual, it is not probable that the positive sukṛta would exclusively denote the well performed ritual. The fact that Gonda referred his treatment of duṣkṛta to a footnote consisting of two pages (97 lines) is to me an indication that somehow he realized the untenability of his interpretation. Footnotes suffering from elephantiasis without exception prove that the author is in trouble.

It is my conviction that neither sukṛta and duṣkṛta nor good and bad karman have anything to do with ritual. They denote merits and demerits. Of course, in ritualistic texts the merits resulting from good actions are mostly based on rituals.

In those ritualistic texts preceding the Upaniṣads, i.e. in the Brāhmaṇas, one may look for passages in which the term karman is used without ritual connotations. I have done so. This is not the occasion to present a lengthy discussion of the material.8 In ritualistic texts you may not expect many references to ethical karman. I have found some.9 Ethically good karman indeed secures a heavenly abode and the bad karman seems to result in punishment in hell, though the texts are rather vague on this point. There is no reference to a return on earth and nobody wants to be freed from his own karman (negative or positive).

So the term karman has a Vedic previous history, but ritual karman (the Vedic ideal) hardly suits the doctrine of transmigration which disqualifies the sacrifices. Ethical karman is barely found in the Vedic texts before the Upaniṣads. It is true that Vedic karman anyhow has results for life after death, but in most religions good conduct and good works are useful for the future (or only for future life after death).

1.2 Return and Rebirth (punarjanman)

In the older Vedic literature rebirth on earth is nowhere explicitly mentioned. Some scholars have assumed vague references to this doctrine, but nowadays serious scholars hardly believe that there is concrete evidence. Indeed, it is very improbable that such a vital issue would only have been treated in vague passages, which allow of other interpretations.

Only in some late Brāhmaṇas (the texts preceding the Upaniṣads) rebirth on earth is mentioned in a few passages. Mostly, however, the rebirth of the father in the son is meant.

The passage which according to some scholars forms the oldest proof of the doctrine of rebirth, would be found in the ŚB 1, 5, 3, 14. I quote Eggeling’s translation: “Now the spring, assuredly, comes into life again out of the winter, for out of the one the other is born again: therefore he who knows this, is indeed born again in this world.” Strikingly the translator Eggeling even does not dedicate a footnote to this historical moment. One may also ask whether actually the doctrine of rebirth plays a role here. From the point of view of terminology everything seems to be alright. However, there are two objections. Rebirth here is not based on karman, but on a particular knowledge and secondly it is even presented as a reward. Even Paul Horsch,10 who exclusively started from the Vedic origin of the doctrine, did not dare to regard this passage as a proof. One may rather assume with Horsch that rebirth in the son was meant here, since this is found in the same context.

There is no use in discussing some even more doubtful passages. However, I make one exception: Jaiminīya Upaniṣad Brāhmaṇa 3, 28, 4, a text to be situated on the borderline between ritualistic Brāhmaṇas and the more or less philosophical Upaniṣads. I will give a free and abridged translation: “If (being in heaven) one might wish: ‘May I be born here again,’ then one will be born again in the family one desires, be it in a Brahmin family or in a Kṣatriya family … . As to this Śāṭyāyani spoke: ‘This world is full of disease. And we also speak about yonder world and exert ourselves to reach it. Why throw away yonder world and try to return here? In this heavenly world (about which we are speaking now) one should be’.” It is evident that rebirth after death is meant here, but it is also striking that this rebirth is positively evaluated as a possible reward besides living on in heaven (the other option). On the other hand the well-known pessimism is introduced here. Apparently one had some idea about rebirth on earth. Horsch (1971, 144) calls this a “Vorstufe” rather than a reference to the theory, since free will and the doctrine of karman would be incompatible. One might as well formulate it differently and state that the author of this passage had some inkling of current ideas on rebirth and pessimism about life on earth, but still did not know the exact implications. Śāṭyāyani had received some information, he was on the right track, he was getting warm, we would say, but he still had no idea about karman and mokṣa and was only able to make a wonderful common sense statement like: “Why do we exert ourselves (i.e. through sacrifices) in order to reach heaven and after having reached heaven prefer life on earth?” Heaven, based on sukṛta, was his ideal rather than mokṣa. This mokṣa will be the next item to be discussed here.


The term mokṣa denotes release. From what did one want to be released? In the classical doctrine it was of course from rebirth and the karman which produces this rebirth and implicitly from life on earth. In the older Vedic literature the term mokṣa and the corresponding verb refer to other situations. Vedic man wanted to be freed from evil (pāpman or pāpa) which meant sin as well as ensuing distress. The metaphor used in this connection is freeing oneself or becoming freed from the noose or snare of evil (Rodhe 1946, 40). The same metaphor is also found in some later Upaniṣads for describing the classical release from the saṁsāra.11 From the terminological point of view there is continuity: mokṣa means becoming freed from the noose of evil, but this evil is in the older Vedic literature rather different from evil in the Upaniṣads.

In the older Veda untimely death and everything which produces this, such as disease and ultimately sin, are evil. Untimely death, however, is a far cry from rebirth.

In the latest portions of the ritualistic Brāhmaṇas we may find some sort of intermediate phase, namely the desire to be released from a renewed death in heaven, the so-called punarmṛtyu (redeath or second death).12 According to Vedic ritualistic thought heaven could be secured by sacrifices, but the late Vedic texts state that this life after death is not unlimited. One has to die again. The release from this second death thus might form a Vorstufe of the classical mokṣa, since in both cases immortality is reached.

However, many scholars do not emphasize the victory over or release from this redeath, but regard redeath as such as a precursor of rebirth. Out of the conception of redeath the doctrine of rebirth would have developed. Some even conclude that the transition from redeath to rebirth was a logical one, since after redeath in heaven automatically rebirth on earth would have to follow. The fact that no text place mentions rebirth as a stage following redeath should warn us not to apply our logic too easily. In all the passages on punarmṛtyu (redeath) only the release from this evil plays a role. This release therefore may form a Vorstufe of the doctrine of mokṣa and redeath is not a Vorstufe of rebirth.13

Terminologically there are Vedic starting points for mokṣa as well as for karman, but the return on earth is found only later, in the Upaniṣads. Now what about the ideological framework?

2 Ideological Framework

There is a friction between karman (action or ritual) and mokṣa (the release from the results of karman). If one assumes a Vedic origin of the doctrine of karman in the form of ritual, it should be taken into account that ritual karman itself actually also aims at salvation. Vedic ritual has more aims, but especially in later Vedic texts its main goal is the obtainment of life after death in heaven. The concept of mokṣa is likewise based on a doctrine of salvation. It is hardly imaginable that two antithetical doctrines of salvation could have been combined in the classical complex of concepts consisting of karman, rebirth and mokṣa.

Moreover, the classical karman doctrine sometimes has been too exclusively associated with ethics. This was even done by Yājñavalkya in his famous statement puṇyo vai puṇyena karmaṇā bhavati pāpaḥ pāpena “One becomes good by good action, bad by bad action (after death)” (BĀU 3, 2, 13). The real issue is not the improvement of one’s own position after death, but getting rid of the results of all actions. This point of view was already represented in TB 3, 12, 9, 7–8 na karmaṇā vardhate no kanīyānna karmaṇā lipyate pāpakena “he does not increase or decrease by karman … he is not polluted by evil karman.” One should be indifferent towards oppositions like good and bad and refrain from all activities which are associated with particular aims. This means that the karman doctrine actually is a doctrine of non-activity rather than being a doctrine of positive ethics. The mokṣa doctrine and the karman doctrine belong together and the karman doctrine which preaches non-activity in social life has to be associated with nivṛtti (“inactivity”) and Vedic karman (“ritual”) with pravṛtti. Therefore the classical doctrine of karman has connections with the so-called Śramaṇa tradition rather than having its roots in Vedic ritualism.14

In the classical doctrine karman and mokṣa (sometimes denoted as dharma and saṁnyāsa) are conflicting entities. In late Vedic texts, passages on overcoming punarmṛtyu seem to combine ritual karman with mokṣa in such a way that the performance of a particular rite produces some sort of mokṣa: one lives on forever and will no more become the victim of death. How could this situation form the starting point of a theory which teaches that all karman is an obstruction to mokṣa? Rather we should assume that some ritualists tried to adopt ideas on mokṣa and to adapt them to their own ritualistic views.

Originally one tried to obtain a personal continuation of life in heaven among the gods. The victory over redeath secures a personal, individual immortality, but this sort of immortality need not have any relation to the classical conception of mokṣa which is based on an identification of the soul or ātman with the highest, cosmic Principle (Brahman) or with the highest deity. This immortality is not personal and the released does not live on as an individual (separated from Brahman or god Brahmā).

Traces of the old ideals are still found in an Upaniṣad like ChU 8, 15, where one personally lives on in the Brahma-heaven rather than becoming merged with Brahman or with god Brahmā.

However, already in the ritualistic Brāhmaṇas there are some passages in which some sort of identification with the highest principle plays a role. According to ŚB 11, 5, 6, 9 after release from punarmṛtyu one will reach sātmatā (community of nature) with Brahman, which is almost the same as absorption into Brahman. An older text like ŚāṅkhB 21, 1 still did not reach this level of speculation and states that after having smitten away the evil of death (i.e. probably of redeath) man obtains identity of world and union with Brahman (i.e. one lives together with Brahman in the Brahman-heaven). An even more conservative view is found in other Brāhmaṇas15 where identity of world and union with several gods are mentioned; i.e. this union is no more than companionship.

It was the old Vedic ideal to live on among the gods in heaven with a complete body, even with genitals in order to continue the pleasures of earthly life. In late Vedic texts like ŚB 10, 4, 3, 10, however, we read that man may become immortal without the body. The body is even equated with pāpman (evil) that should be overcome in another Brāhmaṇa (JB 1, 252), in a passage also dealing with getting rid of punarmṛtyu (redeath). A later Jaiminīya text to be situated between the Brāhmaṇas and the Upaniṣads, JUB 3, 35–39, discusses punarmṛtyu and in that connection observes that the body is the victim of death and that the bodiless becomes immortal. The destruction of evil and getting rid of the body are mentioned together. Evil (pāpman) is no longer exclusively death, but especially the symbol of mortality, the body, of which one wants to become freed. Connecting these data with the reference to sātmatā (mergence and identity with Brahman) one has to conclude that in the latest layers of these ritualistic texts ritual was developing towards mokṣa in the classical sense.

Nevertheless it is uncertain whether we may speak here of an actual continuity. It is as well possible that influences gradually penetrated from external circles into the ritualistic tradition. The almost classical idea of mokṣa occurs only incidentally and in the latest Vedic passages, mostly in connection with passages on the release from punarmṛtyu.

This punarmṛtyu conception seems to have been influenced by the concept of punarjanman (rebirth). Rebirth takes place in endless repetition and one may assume that redeath (punarmṛtyu) is a non-recurring event; it is the final and ultimate end of a life started on earth and continued (without rebirth as a baby) in heaven. However, in a few passages we find a recurring punarmṛtyu, a problem hardly observed as such by almost all scholars. One may suppose that ideas about ever recurring rebirth have influenced some passages on the once-only redeath.

In the theory of mokṣa good works (including rituals) secure only limited goals. One may win a heavenly world, but has to return to earth. I would not exclude the possibility that the ritualists tried to secure their own position by introducing the topic of the victory over redeath, an evil which was never mentioned before and was introduced together with its countermeasure. Why would one so much emphasize a problem which never had been acknowledged as such before?

In several handbooks we read that at the end of the Brāhmaṇa period Vedic people started to have doubts on the eternity of the heavenly bliss obtained by their rituals. I doubt whether the ritualists themselves spontaneously became pessimistic on their own efficacy. Therefore one may assume that punarmṛtyu was not their invention. The passages on the victory over redeath seem to be ritualistic answers to external criticisms and doubts. The ritualistic mokṣa preached by the punarmṛtyu passages, however, did not secure the position of the ritualists. Ultimately the mokṣa of the mystics prevailed and therefore the topic of the punarmṛtyu almost eclipsed in the Upaniṣads as being outdated.

3 Textual Evidence

The locus classicus of the doctrine of transmigration is found in two Upaniṣadic parallels, BĀU 6, 2 and ChU 5, 3–10. Here the ever recurring origin of new life on earth is sketched by means of the ritualistic metaphor of five symbolic sacrifices starting with rain and ending with the seed that man offers into woman after having eaten the vegetarian food which had come into existence from the rain. The sixth sacrifice is the cremation of man generated at the fifth sacrifice. This human being rises upwards from the cremation fire along two possible paths. The one leads him back to earth by way of the rain, along the other he reaches heavenly immortality.16

From a philosophical point of view these passages are rather insignificant. The transmigration concerns seed rather than soul. Still these texts have played an enormous role in the Indian tradition of rebirth and release. Rebirth and release (and in one of the two versions even karma) occur together here for the first time.

In a more remote parallel, KauṣU 1, we see on the one hand further developments, such as the identification of the soul with god Brahmā or even with the Brahman, on the other hand there are connections with older passages in the Brāhmaṇas (JB) in which rebirth on earth and references to karman are entirely missing. There are even Vorstufen in older Vedic passages in which the exchange of fluid between heaven and earth is described in a similar way, but in which neither karman, nor rebirth, nor mokṣa play a role at all.

An attentive philologist would be inclined to see here interesting traces of a gradual development. However, it seems improbable that within a period of about hundred years the ritualists could have evolved from a doctrine of cyclic migration of fluid (rain from heaven transformed into libations which again produce rain) to a complex of concepts referring to karman, rebirth and mokṣa. The assumed continuity might as well be a gradual process of adaptation. New ideas are often traditionally formulated in borrowing circles and thus suggest more continuity than actually existing.

There are two indications for adaptation of external ideas by the ritualists. First the complex of rebirth and mokṣa (in one case in combination with karman) is ascribed to non-Brahmins, namely the ruling Kṣatriyas, in the introduction of these passages in which the Brahmins are almost humiliated by the Kṣatriyas. Secondly, these passages end with the conclusion that sacrifice and other good works produce rebirth, whereas leaving this world, retreating to the wilderness and other practices leading to salvation than sacrifices are said to result in immortality (i.e. mokṣa). One may also qualify oneself for this immortality by knowing the discussed doctrine and the doctrine is presented with ritualistic terms, but this looks rather like a compromise for those who accept the theory without drawing the ultimate consequence of leaving the world and becoming a wandering ascetic. Between an introduction in which the Kṣatriyas are glorious and a conclusion in which the renouncers win immortality there is not much honour left for the ritualists.

In his interesting studies on the hermits in Vedic literature Sprockhoff (1981) also refers to the discussed passages. It may be true that these text places do not prove much on the presence of hermits (vānaprasthas) and the Āśrama system, but it cannot be denied that these people in the wilderness (araṇya) are clearly distinguished from the householder ritualists in the village and that they have not left the world of the ritualists on account of old age. Sprockhoff (p. 85) should have explicitly associated them with the religiously motivated renouncers.

The exact and concrete historical value of such passages may be doubtful, but the message is clear: ritual and good works are meritorious but belong to the sphere of karman and therefore do not produce release.

The alliance between Kṣatriyas and renouncers may be variously explained. The theory about the leading role of the Kṣatriyas, long ago proposed by Garbe and later rejected by others, was revived (with modifications) by Horsch (1966, 432–443). However, he did not discuss the fact that both Kṣatriyas (as proclaimers of the new doctrines) and renouncers (as practicians) play a role in these texts. His observation that in this period the (in his view non-Brahminical) Śramaṇas were prominent and that in particular Kṣatriyas were represented among them (p. 465) does not explain the leading role of powerful kings in these passages. It is uncertain whether the non-priestly circles, which seem to have proclaimed or practised new methods of salvation, were non-Aryan or inspired by non-Aryans. I have already discussed the acculturation between the original Aryans and the autochthons, which resulted in a society of Aryans which racially was not homogeneous. In the period in which the complex of karman, rebirth and mokṣa became manifest in Vedic literature (i.e. in the Upaniṣads) Aryan culture had already penetrated the North-East of the Indian subcontinent. The Kṣatriyas who were in power there may have been Aryanized rather than pure Aryans. Their actual power made them unassailable as long as they accepted the principles of Vedic religion. They were in the position to attack the ritualists. The renouncers placed themselves outside the socio-religious system and therefore were likewise unassailable. As I observed long ago (1973, 216) “One of the most important factors in this development of Kṣatriya influence may have been the fact that discussions on the ritual and on religion in general (the brahmodyas) seem to have been delivered in the sabhā of the king who was not only the institutor of sacrifices, but also of debates.” This still fails to explain the link between kings and renouncers (and their doctrines).

A new explanation was put forward by Olivelle (1992, 36–38 and 1993, 60–62), who assumes that progressive Brahmins operating within the context of developing urbanisation challenged the conservative Vedic religion and changed it from within. On the question of the role of the kings he observes: “I think, however, that the proper, and certainly the more significant, questions are not why these upaniṣadic doctrines were created by kings … but why the proponents of these upaniṣadic doctrines ascribed them to kings … . I think … that the proponents of these doctrines must have found it advantageous to align their doctrines with the nobility in general and with kings in particular … . In general … I think that the alignment with the nobility must have served to distinguish these doctrines from the Vedic doctrines that were identified with Brahmins … . In this light, what appeared a problem for those scholars who upheld the noble provenance of upaniṣadic doctrines, namely why Brahmins should have preserved and handed down these stories that belittled them, ceases to be a problem at all … . Aligning with kings gave their doctrines a new status and prestige and served to distinguish them from the old doctrines” (1993, 61–62).

This would amount to stating that the role of the Kṣatriya was no more than that of a code word for progressiveness and rejection of the Vedic village culture associated with rituals. In this assumed game of tactics and strategics progressive Brahmins who if not living in the towns at least were visiting them and the courts of the kings, would have flattered and manipulated the kings in their attempts to innovate Vedism which was dominated by the villages and the ritual.

In this antithesis between village and town the role of the wilderness (araṇya) and the renouncers living in it still remains unclear. It is hardly probable that renunciation and retirement in the wilderness with theological aims would have exclusively been an invention of more or less urbanized Brahmins.

It is of course possible that the development of towns and courts gave some enterprising Brahmins, who visited them in spite of the dissuasions of the Dharma texts, the opportunity to become acquainted with revolutionary ideas developed outside the mainstream of Vedism. This seems also to be realized by Olivelle, who states: “Within Brāhmaṇism itself, it was the urban Brahmins who, in all likelihood, were most influenced … by the rising prestige and influence of non-Brāhmaṇical religious movements.” However, he continues with: “Most urban Brahmins probably remained within their tradition but challenged and changed it from within. It is these changes, and not the threat posed by non-Brāhmaṇical groups as assumed by many scholars, that I believe were the catalysts for the creation of inclusivistic institutions and theologies … that sought to integrate the emergent ascetic worldview and way of life into the Vedic culture.” (1992, 36).17

This looks like an attempt to rescue the initiative of Brahmanism and the Brahmins. One may, however, as well assume that the initiative was taken by some early “non-Brahmanical religious movements” and that the assumed urban Brahmins only acted as mediators between the more or less heretical movements and traditional Vedism.

A few traces of renouncement and of non-ritualistic practices and ideas are already found in the Vedic ritualistic texts. The opposition is between traditional Vedic ritualists and those who tried other ways of salvation rather than between Vedic and non-Aryan culture. I believe that the complex of karman, rebirth and mokṣa did not originate from the mainstream of Vedic religion, the ritual, though elements are discernible in late Vedic texts (often without coherence). The continuity assumed by some scholars cannot convincingly be proved and does not seem acceptable to me. Unfortunately our information for the older period is confined to ritualistic sources. Fortunately the Upaniṣads, though as texts connected with the ritual tradition, betray completely different influences in some passages. In several respects there is not a real break between Brāhmaṇas and Upaniṣads, but as far as transmigration and mokṣa are concerned there is a breakthrough, which was so sudden and fundamental that a gradual development out of the ritualistic view of life is hardly conceivable.

The question why the non-ritualists believed in rebirth and release cannot simply be answered. It may have some relation to the development from the optimistic view of life in the older Veda to the pessimism of the worldrenouncers. The causes of this pessimism have been variously interpreted: the climate weakened the Aryans or racial mixture produced the same result; socio-economic backgrounds might have created this pessimism. Even the rise of urban centres leading to disintegration of tribal security has been mentioned as a possible cause. Most of these hypotheses are not very convincing. They only try to explain what happened to the Aryans during the aftermath of their invasion. The most convincing explanation is that some socio-economic changes took place in the same period in which new religious ideas and ways of life developed. These changes involved a starting urbanization and concomitant new attitudes, some of which may have produced or increased the leaning toward asceticism and renunciation. For a survey of these aspects I refer to Olivelle’s treatment.18 However, village life was not replaced by urban life and traditional Vedism did not disappear. It may be true that new developments originated in the towns, but it looks as if these developments were brought about in spite of traditional Vedism rather than out of it.

The fact that the same names occur in the ritualistic Brāhmaṇa texts and in the more philosophic Upaniṣads has surprised scholars. Is it possible that someone like Yājñavalkya was interested in details of ritualism as well as in discussions on the ātman? I would not exclude the possibility that here again Vedic tradition has tried to adapt itself to other approaches by attributing non-ritualistic views to famous names of Vedic ritual experts. Thereby more continuity was suggested than actually existing.

First published in Indologica Taurinensia 23–24, 1997–1998, pp. 583–605.

This paper is an adaptation of a lecture published in Dutch several years ago: Oorsprong en achtergrond van de Indische wedergeboorteleer, Koninklijke Nederlandse Akademie van Wetenschappen, Mededelingen van de Afdeling Letterkunde, Nieuwe Reeks, Deel 55 no. 6, Amsterdam–New York–Tokyo, 1992, pp. 3–19 [225–239]. An abridged version was read at the Institute of Oriental Studies of the Russian Academy of Sciences (St. Petersburg Branch) in September 1996.

See e.g. Gonda (1960, repr. 19782, 207), who mentions “Das zyklische Denken, das geneigt ist, die Periodizität des Naturgeschehens auf das Dasein des Menschens und den Weltlauf zu übertragen” as one of the possible explanations. See also Horsch (1971, 115–116), who admits that the cosmic cyclism implies an eternal return of the same, which would not agree with variable rebirth based on variable karman, but still holds that cyclical return may at least have formed a catalysator for the development of the doctrine of transmigration.

Horsch (1971, 99) observed already: “Merkwürdigerweise ist die über 150 Jahre alte indologische Forschung gerade betreffs des Ursprungs dieser grundlegenden Lehre in eine Sackgasse geraten: anstelle einer communis opinio werden noch widersprüchlichsten Thesen verfochten.”

See also his introduction to 1965a, especially p. 13, 15, 20, etc.

See Kuiper (1991a) [provoking a discussion in Indo-Iranian Journal 38 (1995) between Das: 207–238 and Kuiper: 239–247].

According to Olivelle (1992, 22): “In the absence of an adequate definition of orthodoxy within the Brāhmaṇical tradition of this period … the division of conflicting theologies and modes of life into orthodox and heterodox is both anachronistic and utterly useless for historical purposes. The challenges to the mainstream Vedic views are found … across a broad spectrum of religious literature, including some of the most authoritative texts of Brāhmaṇism.” Olivelle wrote this regarding the phenomenon of renunciation, but it might as well apply to the doctrine of transmigration. I disagree with him insofar as we may equate orthodoxy with the mainstream of Vedic views and this mainstream is evident in the transmitted texts. The fact that incidentally opposition to this mainstream is found in these authoritative texts only shows the slow penetration of the still then unauthoritative ideas.

For an extensive discussion of Vedic karman in a non-ritualistic sense and of sukṛta not meaning “well performed sacrifice,” see Bodewitz (1993a and 1993b; this vol. ch. 19 and 18).

For my papers on this subject see previous note.

AB 7, 27, 1 “There sit those doers of an evil deed (pāpasya … karmaṇaḥ kartāraḥ), speakers of impure speech”; 7, 17, 4 “The evil deed (pāpaṁ karma) done by me … torments me” (said by someone who had sold his son in order to be sacrificed); TB 3, 3, 7 opposes honesty (ṛju-karman, sic!) to cheating (vṛjinam) together with two other ethical couples; in the same text (3, 12, 9, 7–8) it is said that the eternal greatness of a Brahmin is that he does not improve nor become worse by karman and that knowing the ātman one does not become polluted by evil karman; ŚB 13, 5, 4, 3 “The righteous Pārikṣitas … destroyed sinful work (karma pāpakam) by their righteous work (puṇyena karmaṇā, here indeed merits obtained by rituals)” (according to Horsch (1971, 140) the first occurrence of the term with ethical implications); in 11, 1, 5, 7 Evil, in the form of Vṛtra, keeps man from good actions (karma qualified with kalyāṇa and sādhu). For more instances and details and material from the Āraṇyakas see Bodewitz (1993a, 225–229; this vol. pp. 258–259).

Horsch (1971, 120); Converse (1971, 351) likewise rejects this passage as evidence.

See e.g. ŚvetU 1, 8; 2, 15; 5, 13; 6, 13.

See Bodewitz (1996b; this vol. ch. 10).

See also Pande (1978, 3).

See also Pande (1978).

See e.g. AB 8, 6, 10.

See Schmithausen (1994) and Bodewitz (1996a, this vol. ch. 9, of which the manuscript was sent to India many years ago so that unfortunately Schmithausen’s article could no more be consulted).

See also Olivelle (1993, 55–64).

Olivelle (1992, 29–38) and (1993, 33–64).

If the inline PDF is not rendering correctly, you can download the PDF file here.

Vedic Cosmology and Ethics

Selected Studies


Table of Contents

Index Card



All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 30 30 10
PDF Downloads 1 1 1
EPUB Downloads 0 0 0

Related Content