Pramāṇa · Search · About · API

Samnyasa Upanishads: Hindu Scriptures on Asceticism and Renunciation c5: Introduction

Saṃnyāsa Upaniṣads: Hindu Scriptures on Asceticism and Renunciation c5

Language
English

Contents

Introduction

1. The Saṃnyāsa Upaniṣads

c5.s1.2

The Upaniṣads form the concluding sections of the several Vedic collections handed down in the ancient Vedic schools and are therefore called *Vedānta*, which literally means the end of the Veda. The Upaniṣads, however, came to be viewed in many traditions not merely as the last books of the Vedas but also as the most important. The very term *Vedānta* was understood to mean not just the end but also the summit and crown of the Veda.

c5.s1.3

Many indigenous traditions divide the Vedic corpus into two sections: the section on rites (*karmakāṇḍa*), consisting of the earlier portions, and the section on knowledge (*jñānakāṇḍa*), consisting of the Upaniṣads. The theology of Vedic revelation within the mainstream of the Brāhmaṇical tradition regarded the Vedas to be without an author; they constitute the temporal manifestations of eternal and transcendent knowledge. As the ritual section contains infallible directions in ritual and moral matters, so the Upaniṣads contain the transcendent knowledge needed for human salvation. Within the soteriological traditions, such as Advaita Vedānta, that considered knowledge as the sole means of liberation (*mokṣa*), therefore, the Upaniṣads came to be regarded as its most authoritative source. As one medieval text puts it, “The Vedānta is contained in the Veda like oil in the seed.”¹

c5.s1.4

¹ *Muktikā Upaniṣad*, 1.9.

c5.s1.5

Because of the foundational nature of the Upaniṣads, later Hindu sects and theologies sought to find in them the revealed basis of and the ultimate justification for their doctrines and practices. The proponents understood their sectarian doctrines and practices to be the explication of the essential message of the Upaniṣads. Thus many Hindu sects called their own doctrines *Vedānta*.

c5.s1.6

The early Upaniṣads, composed many centuries before the rise of these sects and traditions, however, did not always directly support their often contradictory positions. Sectarian theologians, therefore, resorted frequently to hermeneutical strategies, interpreting the ancient Upaniṣads in ways that would provide support for their doctrines;² sometimes they composed new Upaniṣads. Given the lack of an acknowledged and closed canon of the Vedic corpus,³ these new texts were able to gain recognition, at least within the sects in which they were composed, as authentic Vedic documents, having the same sanctity and authority as other Vedic books. The number of Upaniṣads thus swelled to well over 100, and Upaniṣads continued to be composed down to modern times to support every sort of religious belief and practice.

c5.s1.7

² I have discussed some of these strategies in Olivelle 1986, 57–76.

c5.s1.8

³ For a survey of the problems relating to the canon and transmission of the Vedic texts, see Olivelle 1986, 66–76; L. Renou, *The Destiny of the Veda in India*, trans. D. R. Chanana (Delhi: Motilal Banarsidass, 1965). canon and its exegesis. . . . It [exegetical enterprise] is, at the same time, the most profoundly cultural, and hence, the most illuminating for what ought to be the essentially anthropological view point of the historian of religion and a conception of religion as human labor.” *Imagining Religion: From Babylon to Jonestown* (Chicago: University of Chicago Press, 1982), 43.

c5.s1.9

In the eyes of the believers, all Upaniṣads have the same authority. They are eternal and transcend history. Scholars who use them as sources for the reconstruction of Indian religious history, however, must approach them as human documents located in space and time. From a historical perspective, therefore, modern scholarship has customarily distinguished between the *major* (sometimes called *classical*) and the *minor* Upaniṣads. The former are generally older and were composed within the Vedic schools, whereas the latter are by and large later compositions, some of which are sectarian in nature.⁴ The Saṃnyāsa Upaniṣads belong to the latter category.

c5.s1.10

They are a collection of twenty texts written in Sanskrit. Their common characteristic is the theme of *saṃnyāsa*, or renunciation. The Saṃnyāsa Upaniṣads, however, do not constitute an indigenous classification of the Upaniṣads; no Indian list or collection

1.1 Editions, Translations, and Studies

c5.s1.t1.1

Deussen’s fivefold classification, with the addition of the Śākta-Upaniṣads, was followed by the Adyar Library when, under the direction of F. Otto Schrader, it drew up a plan to publish critical editions of all the minor Upaniṣads (Schrader, ii–iii). The first volume to be published in the series contained the Saṃnyāsa Upaniṣads, a collection of twenty texts critically edited by Schrader himself and published in 1912 with an original Sanskrit commentary, *Ṭippaṇī*, prepared by Schrader with the help of several pandits. The present translation is based on Schrader’s critical edition.

c5.s1.t1.2

Unfortunately, the editors of the remaining volumes of this series did not follow Schrader’s lead; the other collections were not critically edited. They were published without a proper critical apparatus and reproduce the version commented on by Upaniṣadbrahmayogin. The Saṃnyāsa Upaniṣads thus remain the only collection of minor Upaniṣads to have been critically edited.

c5.s1.t1.3

As part of the same series, the Adyar Library in 1929 produced another publication of the Saṃnyāsa Upaniṣads based on the version commented on by Upaniṣadbrahmayogin (see Dikshit 1929), a version far inferior to the critically reconstituted edition of Schrader. A reprint of it amounting in reality to a new edition was published in 1966.⁸ Individual texts of the collection have been published frequently in uncritical editions.⁹

c5.s1.t1.4

⁸ For differences in the two editions, see Sprockhoff 1990, 7–17.

c5.s1.t1.5

⁹ All these editions are listed by Sprockhoff (1976) at the beginning of his discussion of each Upaniṣad.

c5.s1.t1.6

In his introduction Schrader says that his translation of the Saṃnyāsa Upaniṣads “is practically ready for the press.”¹⁰ Unfortunately, it was never published. The Adyar Library recently published a translation of the Saṃnyāsa Upaniṣads by A. A. Ramanathan (Ramanathan 1978). Besides being an extremely poor and inaccurate translation, it is not based on Schrader’s critical edition but follows by and large Dikshit’s edition of the version commented on by Upaniṣadbrahmayogin.¹¹ I have been unable to consult the recent French translation of these Upaniṣads by Alyette Degrâces-Fahd referred to by Sprockhoff (1989, 138; 1990, 47, n. 150), who considers it an uncritical translation.

c5.s1.t1.7

¹⁰ Schrader, iii, n. 1. He repeats this claim elsewhere: Schrader, 80, nn. 1–2. Cf. Sprockhoff 1976, 10, n. 3.

c5.s1.t1.8

¹¹ For an assessment of Ramanathan’s translation, see my review in the *Journal of the American Oriental Society* 102 (1982): 228–229. A detailed critique of this translation has been made by Sprockhoff 1990. Individual texts, especially of the older group, have been translated into English and other European languages, but they are for the most part not based on Schrader’s critical edition. For a bibliography of these translations, see Sprockhoff 1976 under each Upaniṣad. The *Śāṭyāyanīya* was translated by me in Olivelle 1987, and the *Kaṭhaśruti* (together with the *Kaṭharudra* of the SR) by Sprockhoff 1990.

c5.s1.t1.9

Joachim Friedrich Sprockhoff, a student of Schrader, has made an excellent and detailed study of all the Saṃnyāsa Upaniṣads (Sprockhoff 1976). Sprockhoff’s work is indispensable for any serious study of these documents. It is neither possible nor desirable within the scope of this introduction to include all the details of his analysis. For more detailed and technical information on these documents, including their textual histories, I can do no better than refer the reader to Sprockhoff’s study, which ought to be regarded as a standard work of reference on Brāhmaṇical renunciation. Besides being the only group of minor Upaniṣads to have been critically edited, the Saṃnyāsa Upaniṣads remains also the only group to have been subjected to such a thorough historical and philological study.

c5.s1.t1.10

It is my hope that this first complete English translation of the critical editions of all the Saṃnyāsa Upaniṣads will make that valuable collection accessible to a wider group of people interested in the history of Indian religion and culture.

1.2 Dates

c5.s1.t2.1

The Saṃnyāsa Upaniṣads share with other ancient Indian texts a common problem: it is impossible to date them with any degree of precision or certainty. Neither the texts nor their manuscript traditions contain any information regarding their authors or the dates and places of their composition. Except for citations in other datable works, we have no external evidence regarding their dates. Almost all these texts, furthermore, are not original compositions. They draw extensively from other Brāhmaṇical sources, such as the classical Upaniṣads, the epics, and the Purāṇas.¹² Many of them, moreover, are composite texts, containing sections taken from older works or from other Saṃnyāsa Upaniṣads. Some of the later Upaniṣads appear to be expansions of older ones. Others, such as the *Laghu-Saṃnyāsa* and the *Kuṇḍikā*, as well as the *Kaṭhaśruti* and the *Kaṭharudra* (which occurs only in the SR), appear to be different recensions of the same text. Each text, therefore, contains several layers that may be separated by hundreds of years. All of this makes any dating of these documents tentative and provisional at best.

c5.s1.t2.2

¹² Sprockhoff (1976 especially 277–295) has dealt extensively with the composition and the sources of the Saṃnyāsa Upaniṣads. He gives all the sources and parallel passages in seventeen detailed tables (319–377) and in an extremely helpful diagram on p. 280.

c5.s1.t2.3

In his critical edition, Schrader arranged these Upaniṣads in a manner that in his view approximated their relative dates: “The texts edited in this volume fall into two groups, an older one and a younger one. The older one comprises the first six texts, the younger one Maitreya and the succeeding Upaniṣads. Āśrama-Upaniṣad cannot well be included in either of the groups; it seems to stand exactly on the dividing-line” (Schrader, xxvi).

c5.s1.t2.4

Sprockhoff’s exhaustive study generally confirms this broad division. Sprockhoff, however, has attempted to date individual Upaniṣads and even the different strata within them. He assigns the older group to the last few centuries before the common era. The *Āśrama Upaniṣad* was composed around 300 c.e. The texts of the younger group for the most part belong to the medieval period. Sprockhoff assigns the *Nāradaparivrā-jaka* to around 1150 c.e. and the *Śāṭyāyanīya* to 1200 c.e. Most of the others belong to the fourteenth and the fifteenth centuries.¹³

c5.s1.t2.5

¹³ For details on the dates of each Upaniṣad, see the appropriate chapter of Sprockhoff 1976.

c5.s1.t2.6

A clue to the dating of the older group is found in a controversy recorded in all except the *Laghu-Saṃnyāsa*. These texts attempt to answer the objection that if a renouncer cuts his topknot and discards his sacrificial string, he can no longer be regarded as a Brahmin (see below 4.3). Now the abandonment of the string as well as any objection to such an abandonment have significance only if it was either an obligation or at least a universal custom among Brahmins to wear such a string. It is no doubt true that in medieval and modern times the wearing of a sacrificial string was regarded as the hallmark of a Brahmin. The older Brāhmaṇical documents, however, do not record a rule or custom that required Brahmins to wear a sacred string at all times. The term *yajñopavīta*, which in later times came to mean the sacrificial string, is used in them to refer to a particular mode of wearing the upper garment during ritual acts.¹⁴ It is nowhere mentioned that a garment or a string should be worn in that manner at all times.

c5.s1.t2.7

¹⁴ The garment or string goes over the left shoulder and hangs under the right arm. Other ways of wearing it are called *prācīnāvīta* (over the right shoulder and under the left arm), used during rites for the dead, and *nivīta* (hanging from the neck).

c5.s1.t2.8

That the custom of wearing a string was a late development is confirmed by the fact that the oldest texts are silent on the investiture of a boy with the sacred string at his initiation, an investiture that later became its central element. After examining all the evidence, Kane concludes:

c5.s1.t2.9

From the above passages, from the fact that many of the gṛhyasūtras [“treatises on domestic ritual”] are entirely silent about the giving or wearing of the sacred thread in upanayana [“Vedic

c5.s1.t2.10

initiation''] and from the fact that no mantra is cited from the Vedic Literature for the act of giving the yajñopavīta (which is now the centre of the upanayana rites), while scores of vedic mantras are cited for the several component parts of the ceremony of upanayana, it is most probable, if not certain, that sacred thread was not invariably used in the older times as in the times of the later smṛtis and in modern times, that originally the upper garment was used in various positions for certain acts, that it could be laid aside altogether in the most ancient times and that the cord of threads came to be used first as an option and later on exclusively for the upper garment. (*sic*; Kane, HDh, 2:291)

c5.s1.t2.11

If Kane is right in dating the custom of wearing a sacred string at all times to the period of the *smṛtis*, that is, roughly the first few centuries of the common era—and I believe he is—then the texts of the older group, with the exception of the *Laghu-Saṃnyāsa*, must be younger by several centuries than assumed by Sprockhoff. The earliest date to which these texts in their present form can be assigned is the first few centuries of, rather than before, the common era. It is possible, of course, that sections of these composite documents belong to an earlier period.

c5.s1.t2.12

With regard to the *Śāṭyāyanīya Upaniṣad*, it may be possible to add some more information bearing on the date to that given by Sprockhoff. This text was clearly the product of a Vaiṣṇava sect, in all likelihood the Śrī-Vaiṣṇava founded by Rāmānuja. As far as I know, the earliest citation of this Upaniṣad is found in Varadācārya's *Yatiliṅgasamarthana*.¹⁵ Varadācārya (1165–1275 C.E.) was the grandson of Rāmānuja's sister's son (Olivelle 1986, 23–24). Rāmānuja's elder contemporary and teacher was Yādava Prakāśa. Tradition has it that Rāmānuja converted his former teacher from a non-dualist form of Vedānta to his Viśiṣṭādvaita philosophy.¹⁶ Yādava's literary activities, therefore, must have taken place in the second half of the eleventh century C.E.

c5.s1.t2.13

¹⁵ See the edition and translation of this text in Olivelle 1987.

c5.s1.t2.14

¹⁶ Vedānta Deśika (13–14th cent. C.E.) in his *Śatadūṣaṇī* refers to the conversion of Yādava Prakāśa: see Olivelle 1987, 82.

c5.s1.t2.15

He wrote an extensive treatise on renunciation called *Yatidharmasamuccaya* (Olivelle 1986, 22), which generally supports the Śrī-Vaiṣṇava position on renunciation, especially on the disputed issue of the insignia of a renouncer. He fails, however, to cite the *Śāṭyāyanīya Upaniṣad*, even though it would have given him plenty of scriptural ammunition against the Advaita position. It is quite unlikely that Yādava would have been ignorant of or deliberately ignored such an important scriptural text of his tradition, a text that is cited prominently a few generations later by Varadācārya. The likely conclusion is that the *Śāṭyāyanīya* was either composed or attained the status of an Upaniṣad sometime between Yādava and Varada, probably in the early part of the twelfth century.

1.3 Context: Brāhmaṇical Literature on Renunciation

c5.s1.t3.1

The Saṃnyāsa Upaniṣads did not arise in a vacuum; they were part of a broader literary tradition concerning renunciation and related topics both within the Brāhmaṇical mainstream and in non-Brāhmaṇical traditions such as Buddhism and Jainism. Here I want to present a brief conspectus of that broader tradition that provides the context within which these documents were composed and need to be studied.

c5.s1.t3.2

There is sufficient evidence for the existence of organized ascetical institutions¹⁷ in northern India probably during the sixth—and for certain by the fifth—century B.C.E. At a very early age, moreover, at least some of these ascetical institutions produced literary works often ascribed to their founders. At first they were, in all likelihood, composed and transmitted orally. Examples of such texts are found both within and outside the Brāhmaṇical tradition.

c5.s1.t3.3

¹⁷ I use the term *institution* in the broadest possible sense. We have little evidence regarding how large these institutions were or how they were organized. From early Buddhist evidence it appears that each ascetic group coalesced around a teacher who became its corporate head and, at least in the case of institutions known to us, such as Buddhism and Jainism, provided both an ideology and rules of conduct for its members.

c5.s1.t3.4

Whatever may be said about the dates of their extant canons of scripture, the Buddhist and Jain traditions—and possibly other extinct non-Brāhmaṇical sects such as the Ājīvika—produced a considerable body of literature not long after their founding. These literary products often claim to record the very words of the deceased founders of their respective sects and were thus infused with their personal authority, becoming thereby sacred and authoritative. The early ascetic literature of non-Brāhmaṇical sects thus became the nucleus of their sacred scriptures, which paralleled—and which may have even been composed in imitation of—the Vedic scriptures, the model par excellence of sacred text in India.

c5.s1.t3.5

Many of the earliest of these literary products appear to have been codes of ascetic conduct; this is indeed a feature common to all ancient ascetical literature in India. Given that the main features of the renunciatory life are by and large uniform across all the traditions, it appears likely that these texts for the most part codified existent patterns of ascetic behavior, even though each sect modified them to suit its own needs and also included rules peculiar to it. An early example of such a code is the Buddhist *Prātimokṣa*, a list of over 200 monastic rules.¹⁸ It is also likely that early ascetic texts also included the rituals and ceremonies that accompanied the entry of a person into the ascetic life; initiation rites occupy a prominent position in the extant versions of these texts.

c5.s1.t3.6

¹⁸ For a discussion of the *Prātimokṣa* and Buddhist monastic discipline, see S. Dutt, *Early Buddhist Monachism*. 2d ed. (London: Asia Publishing House, 1960); W. Pachow, "A Comparative Study of the Prātimokṣa," *The Sino-Indian Studies*, VI(1951), pts. 1–2; C. S. Prebish, *Buddhist Monastic Discipline: The Sanskrit Prātimokṣa Sūtras of the Mahāsāṃghikas and Mūlasarvāstivādins* (University Park, Pa.: Pennsylvania State University Press, 1975).

c5.s1.t3.7

That this early ascetic literature should have focused on ascetic practice should come as no surprise, for in India throughout the ages proper conduct has counted for more than ideological purity. As we shall see (2.5), what a person did—and even where one did it—was considered inseparable from the broader issues of world view and religious ideology; indeed, the latter was often reducible to the former. Within Brāhmaṇism itself, moreover, the bulk of the religious literature of this period was produced mostly for very practical reasons: either for the proper performance of Vedic or domestic rituals (the *Śrauta*- and *Gṛhya-sūtras*) or to inculcate proper personal and social behavior (the *Dharma-sūtras*). Indeed, all the so-called ancillary Vedic texts (*Vedāṅgas*) have very practical purposes.

c5.s1.t3.8

Within Brāhmaṇism itself the earliest literature with an ascetic thrust is found in the classical Upaniṣads. Passages with a strong ideological slant in favor of the ascetic worldview and way of life are found in the earliest of the Upaniṣads, such as the *Bṛhadāraṇyaka* and the *Chāndogya*. Some of them will be examined in the next section. It is unclear, however, whether these passages were produced within ascetic institutions and later incorporated into the Upaniṣads or merely reflect the influence of ascetic ideologies on some segments of the ancient Vedic schools.

c5.s1.t3.9

The earliest evidence we possess pointing to the existence of independent ascetical codes within the Brāhmaṇical tradition comes from the grammatical treatise *Aṣṭādhyāyī* of Pāṇini, commonly assigned to the fourth century B.C.E. In explaining a particular type of derivative noun meaning "proclaimed by," Pāṇini (IV.3.110–111) refers to the *bhikṣusūtras*—codes of conduct for mendicants—proclaimed by Pārāśarya and Karmandin.¹⁹ That these were not generic texts intended for all but codes that regulated the life of sociologically distinct groups of mendicants can be gathered from the reference to "Pārāśarin mendicants" (*pārāśariṇo bhikṣavaḥ*) as a distinct group by Patañjali, generally assigned to the second century B.C.E., in his commentary, *Mahābhāṣya* (IV.2.66), on Pāṇini’s grammar. This evidence, coming as it does from grammarians intent on finding examples for grammatical rules among words in common use, is historically very significant and certainly more reliable than any information we may find in the normative literature.

c5.s1.t3.10

¹⁹ For a detailed discussion of these, see Ajay Mitra Shastri, "The Bhikṣusūtra of Pārāśarya," *Journal of the Asiatic Society* (Calcutta), 14, nos. 2–4(1972): 52–59 (issued May 1975). I agree with Shastri that, despite some contradictory evidence from the poet Bāṇa, Pāṇini considered these *sūtras* to have been Brāhmaṇical works.

c5.s1.t3.11

We know very little, unfortunately, about these early Brāhmaṇical authors of ascetic texts.²⁰ It is not even clear whether their composition and transmission were carried out within or outside the Vedic schools. There is considerable evidence, especially in the epics, nevertheless, for a close association between the early Sāṃkhya school of philosophy and the renunciatory mode of life. Ancient Sāṃkhya teachers, such as Kapila and Pañcaśikha,²¹ are portrayed as wandering mendicants. Although I am inclined to believe that the earliest Brāhmaṇical literature on renunciation originated among such Sāṃkhya renouncers, there is insufficient evidence at present to demonstrate that conclusion with any certainty, especially because none of these early ascetic texts has been preserved.

c5.s1.t3.12

²⁰ There were also parallel codes dealing with forest hermits. Baudhāyana (fourth century B.C.E.), for example, explicitly refers to a *vaikhānasaśāstra* (“Treatise for Hermits” or “Treatise by Vikhanas”): BDh 2.11.14.

c5.s1.t3.13

²¹ Some scholars have identified Pañcaśikha with Pāṇini’s Pārāśarya on the basis of a reference in the *Mahābhārata* (12.308.24) to Pañcaśikha as belonging to the lineage of Pārāśarya (*pārāśaryagotra*). See Shastri, 55.

c5.s1.t3.14

One possible reason for the disappearance of independent texts on asceticism may have been the absence of organized ascetic or monastic institutions within Brāhmaṇism until the early medieval period, institutions that would have facilitated the preservation and the transmission of the ancient texts. Another and possibly more important reason may have been the Brāhmaṇical theology of scriptural authority.

c5.s1.t3.15

By the last few centuries before the common era, Brāhmaṇical theologians had developed the principle that scriptural authority resided solely and primarily in the Veda and derivatively in texts referred to as *smṛtis*. Both these categories of sacred texts, but especially *smṛti*, were very elastic and without recognized canons or boundaries. “*Smṛti*” itself included many categories of texts, themselves without definite boundaries: ritual texts (*Śrauta-* and *Gṛhya-sūtras*), epics, Purāṇas, and the like. As time went on, more and more texts, and classes of texts, especially of sectarian provenance, were added to this category. What all the *smṛti* texts have in common, however, is that they are viewed as representing the Vedic tradition, which is the context for transmitting and understanding the Vedas. According to Brāhmaṇical theology, *smṛtis* derive their authority from the Vedas on which they are based. For regulating social conduct and individual morality, the central *smṛti* texts were the *Dharma-sūtras* and their later counterparts, the *Dharma-śāstras*, that are often and significantly referred to simply as *smṛtis*. They are *smṛti* par excellence.

c5.s1.t3.16

Lacking an independent scriptural tradition, such as those created in Buddhism and Jainism, the only way ascetic codes could acquire scriptural authority within Brāhmaṇism was by being incorporated into the elastic categories of Veda or *smṛti*. The Saṃnyāsa Upaniṣads are examples of ascetic texts that were either incorporated into or deliberately composed to fit the Vedic category of “Upaniṣad.”

c5.s1.t3.17

The *dharma* literature represented by the *smṛtis*,²² however, contains the bulk of the ancient Brāhmaṇical literature on asceticism. There are four extant *Dharma-sūtras* ascribed to Gautama, Baudhāyana, Āpastamba, and Vasiṣṭha and belonging roughly to the last four or five centuries before the common era. Each of them has a section containing the rules of conduct of both renouncers and forest hermits. These are given within the context of the *āśrama* system that permits a person to live a religious life as a student, a householder, a hermit, or a renouncer.²³ It appears likely that these rules were taken from ascetic codes; the *Dharma-sūtras* frequently quote from unnamed sources.²⁴ Gautama and Baudhāyana, moreover, present the *āśrama* system as the view of an opponent. These rules on asceticism are in prose and set in the aphoristic genre of *sūtra*; in this they do not differ from the other sections of the *Dharma-sūtras*. With the exception of the earliest strata within the Saṃnyāsa Upaniṣads, the ascetic rules contained in the *Dharma-sūtras* are the earliest extant ascetic literature within the Brāhmaṇical tradition.

c5.s1.t3.18

²² Under this rubric I include not only the ancient *Dharma-sūtras* and the later *Dharma-śāstras*, but also the *dharma* portions of the epics and Purāṇas.

c5.s1.t3.19

²³ GDh 3.1–36; BDh 2.11.6–34; 2.17–18; ĀpDh 2.21.1–17; VDh 10.1–31. For further discussion, see 2.6.

c5.s1.t3.20

²⁴ See, for example, GDh 3.1; BDh 2.11.8; 2.17.30; 2.18.13; VDh 10.2, 20.

c5.s1.t3.21

The *Dharma-śāstras*, belonging roughly to the first five or six centuries of the common era, as well as the sections dealing with *dharma* in the epics and the Purāṇas, follow the *sūtras* in incorporating rules of ascetic behavior within the context of the *āśrama* system.²⁵ These texts, however, are written for the most part in verse, and versification also characterizes the ascetic literature of this period. Most of these texts are now extinct. There is a staggering number of citations in medieval ascetic texts from these lost *Dharma-śāstras*.

c5.s1.t3.22

²⁵ MDh 6.33–86; YDh 3.56–66; ViDh 96; VaiDh 1.9; 2.6–8; 3.6–8. The *Mahābhārata* and the Purāṇas also contain discussions of ascetic life, mostly within the context of the *āśrama* system. Several of these passages, such as the dialogue between Bhṛgu and Bharadvāja (MBh 12.184), appear to be remnants of extinct *Dharma-sūtras*.

c5.s1.t3.23

We have to wait until the medieval period, after the organization of Brāhmaṇical monastic orders, traditionally ascribed to the Advaita theologian Śaṅkara, for the composition of independent ascetic texts. They were probably composed within monastic establishments as handbooks for the use of monks and generally fall within the category of *dharma* literature known as *paddhati*, handbooks dealing with specific topics mostly of a ritual nature.²⁶ They did not form part of Veda or *smṛti*, and thus their authority did not transcend the boundaries of their sects. This sociological background also explains the sectarian biases they reveal, even though all claim to hand down the Vedic *dharma* on asceticism and cite profusely from Vedic and *smṛti* sources. Some of them are even polemical in nature.²⁷

c5.s1.t3.24

²⁶ The *paddhati* itself can be viewed as a subcategory of the *nibandha* class of literature. *Nibandhas* are not *smṛti* and therefore not part of scripture. They are medieval digests of law in the broadest sense of the term and include all the topics covered in the *Dharmaśāstras*.

c5.s1.t3.25

²⁷ For such polemical works, see Olivelle 1986 and 1987.

c5.s1.t3.26

Within the Brāhmaṇical mainstream, the earliest independent work on renunciation outside the Saṃnyāsa Upaniṣads that has come down to us is a work called *Yatidharmasamuccaya*, written by Yādava Prakāśa (eleventh century C.E.).²⁸ Yādava, according to tradition, subscribed to some form of non-dualist Vedānta but was later converted by his own pupil, Rāmānuja; the Śrī-Vaiṣṇava tradition claims him as one of its own. He is cited by later Śrī-Vaiṣṇava authors, and Vedānta Deśika says that only Yādava's "treatise on the *dharma* of renouncers is unanimously accepted by the learned, just as Manu's law book and the like."²⁹ The medieval period saw the proliferation of similar compendia of ascetic rules and customs, although none can match the depth, comprehensiveness, and lucidity of Yādava's work.³⁰

c5.s1.t3.27

²⁸ See above 1.2 and n. 16. I am currently engaged in preparing a critical edition and translation of this work. Its third chapter is included in Olivelle 1987.

c5.s1.t3.28

²⁹ *Yatiliṅgabhedabhaṅgavāda* of the *Śatadūṣaṇī*. See Olivelle 1987, 82.

c5.s1.t3.29

³⁰ Kane (HDh 1: 989–1158) lists over eighty such works. Many more have come to light since Kane published his work in 1930. Almost all these works exist only in manuscript and, therefore, are unavailable to all but the most assiduous of scholars. Among the ones published are Viśveśvara Sarasvatī's *Yatidharmasaṃgraha* (cf. YDhS) and Vāsudevāśrama's *Yatidharmaprakāśa* (ed. and trans. Olivelle 1976 and 1977).

c5.s1.t3.30

One of the few, and clearly the most important, independent medieval works on renunciation that is not a mere handbook is Vidyāraṇya's (literary activity 1330–1385 C.E.) *Jīvanmuktiviveka*. Written from an Advaita Vedānta standpoint, it deals with the state of a renouncer who has achieved the liberating knowledge of Brahman and contains a commentary on the *Paramahaṃsa Upaniṣad*.

c5.s1.t3.31

The Saṃnyāsa Upaniṣads fall broadly into the pattern of development traced above. The earlier texts are largely written in prose and often exhibit the pithy style of the *sūtra* genre; this is most evident in the *Nirvāṇa Upaniṣad*. Some of them probably existed as texts of the *sūtra* tradition before they were classified as Upaniṣads.³¹ The *Āśrama Upaniṣad*, for example, has many recensions, some of which are considered as *smṛtis*.³² Some of the later documents within the Saṃnyāsa Upaniṣads, such as the *Nāradaparivrājaka Upaniṣad*, on the other hand, show all the characteristics of medieval legal compendia (*nibandha*). These later Upaniṣads are collections of passages, some in prose but mostly in verse, from older *smṛtis* and other sources. Unlike the *nibandhas*, however, they do not reveal their sources. It appears likely that these later documents were composed deliberately in a manner that would allow them to be perceived as Upaniṣads.

c5.s1.t3.32

³¹ For more details of the prehistory of the early Saṃnyāsa Upaniṣads, see Sprockhoff 1976. Sprockhoff (1987) has drawn special attention to the similarities between the *Kaṭhaśruti Upaniṣad* and a section of the *Mānavaśrautasūtra*.

c5.s1.t3.33

³² For the textual history of this Upaniṣad and its relation to the *Kāṇvāyana-* or *Kātyāyana-smṛti*, see Sprockhoff 1976, 120–124.

c5.s1.t3.34

The fact that the major monasteries of early medieval India—monasteries where the old ascetic texts were preserved and new ones produced—belonged to the Advaita Vedānta tradition may explain the strong Advaita leaning of most of the Saṃnyāsa Upaniṣads. Old texts with different views may not have been preserved; others may have been recast in the Advaita mold. The only Upaniṣad of our collection with a non-Advaita, or even an anti-Advaita, orientation is the *Śāṭyāyanīya*, and it was undoubtedly composed within the Śrī-Vaiṣṇava camp, possibly to counter the Advaita bias of the others.

2. Renunciation and Society: The Inner Conflict of Tradition

c5.s2.2

As the Saṃnyāsa Upaniṣads need to be located within the broader literary tradition of Brāhmaṇism, so the renunciatory life and ideals of India can be adequately studied only in their historical context and within the framework of their interaction with societal religion and ethics. That the ascetic way of life and the so-called life-negating and world-denying ideals and ideologies are central elements or even defining characteristics of Indian religions is often taken as axiomatic by scholars and lay observers alike. Such assumptions, however, often create the impression of a monolithic religion unrelated to historical vicissitudes. Albert Schweitzer's evaluation of the Indian religious ethos and its presumed contrast to the Hellenistic, or more generally the Western ethos is representative:

c5.s2.3

When Hellenic thought turns towards world and life negation, it is because in the end it begins to have misgivings about the world and life affirmation which for centuries had seemed a matter of course. . . . In the thought of India, on the other hand, world and life negation does not originate in a similar experience. It is there from the very beginning, self-originated, born as it were in a cloudless sky. And it is there in a very curious guise. Unlike Hellenic world and life negation, it does not claim

c5.s2.4

to be generally accepted, but remains on good terms with world and life affirmation, which it allows to exist alongside of it.¹

c5.s2.5

¹ Albert Schweitzer, *Indian Thought and its Development*, trans. Mrs. Charles E. B. Russell (Boston: The Beacon Press, 1960), 18. The book was originally published in 1936.

c5.s2.6

Schweitzer is wrong on both counts. First, world-denying ideals and ways of life were not "there from the very beginning" in India. They were not "self-originated" but arose at a particular time in history—probably around the sixth century B.C.E.—against the background of momentous socioeconomic changes along the Ganges valley. Second, world renunciation in India did not always remain "on good terms" with society-oriented ideologies. It took a long period of domestication for the renunciatory ideals to be incorporated into the fabric of mainstream religion, and even this incorporation, as the evidence of the Saṃnyāsa Upaniṣads themselves demonstrates, did not altogether erase their irresolvable conflict. The reality of Indian religions is more complex than Schweitzer's observations suggest.

c5.s2.7

In assessing the emergence of renunciatory ideologies and ways of life, scholars have been generally divided into two camps; some see it as a natural or, to use Heesterman's term, an *orthogenetic* development of the Vedic tradition, while others regard it as a new phenomenon that challenged and transformed the Vedic religion.

c5.s2.8

Heesterman's work remains the best argued and most developed exposition of the former position.² He sees the emergence of renunciation not "as a protest against brahmanical orthodoxy" but as "the orthogenetic, internal development of Vedic thought," the final and logical outcome of the internal maturation of the Vedic sacrificial theology that sought ultimate autonomy and transcendence (Heesterman 1964, 24).

c5.s2.9

² See, for example, Heesterman 1964. This essay is included in his recent collection, Heesterman 1985. This view is also evident in the writings of Madeleine Biardeau (for example, in Biardeau and Malamond, 1976). For an accessible account of this controversy, see Kaelber 1989, 101–124.

c5.s2.10

The opposite view has been argued most forcefully by Louis Dumont (1960) in his seminal essay “World Renunciation in Indian Religions.” Even though I am willing to admit lines of continuity between renunciation and sacrificial religion, I believe that Dumont is right in viewing it as a “new” element that at least initially challenged and contradicted many of the central premises of sacrificial theology. Nothing in history, after all, emerges ex nihilo, and Vedic religion provided the backdrop both positively—for example, in providing key concepts and a technical vocabulary—and negatively for the emergence of renunciation. Yet, I believe—and the following study, I trust, will demonstrate—that when the evidence is examined completely it does point to a profound conflict between the two, a conflict that cannot be adequately explained if renunciation was in fact “an orthogenetic development of Vedic thought.” The claim frequently made in later sources that the Brahmin householder is the ideal renouncer far from supporting Heesterman’s position appears to reflect the incorporation of renunciatory values into Brāhmaṇical institutions and theology; often it is mere rhetoric.³

c5.s2.11

³ These claims of Brahmin householders are taken at their face value by A. Wezler, *Die wahren “Speiseresteesser” (Skt. vighasāśin)*. Akademie der Wissenschaften und der Literatur. (Wiesbaden: Kommission Steiner Verlag, 1978). Bodewitz (review of Wezler in *WZKSA* 24 (1980): 239–242) is right, I think, in caricaturing this type of Brāhmaṇical rhetoric by comparing it to a right-wing politician’s characterization of employers as the “true workers,” because they do not just work from nine to five but have to continuously think about their business.

c5.s2.12

Although I consider world renunciation to be a “new element” that challenged the Vedic world, I do not subscribe to the view put forward by some scholars that ascetic modes of life were non-Aryan in origin. These scholars argue that the emergence of renunciation as a major force in religion signals the growing influence of non-Aryan populations on the Aryans as they migrated east along the Ganges valley.⁴ Such an hypothesis is clearly untenable if we confine ourselves to the available evidence. It is obvious that the Vedic society contained large numbers of people whose roots were non-Aryan and that their customs and beliefs must have influenced the dominant Aryan classes. It is quite a different matter, however, to attempt to isolate non-Aryan customs, beliefs, or traits at a period a millennium or more after the initial Aryan migration into India. The mere absence of a practice or belief in the extant Vedic documents is not sufficient to call it non-Aryan. The Vedic literature provides us with merely a peephole into the society of that period; it does not give us a complete account of the religious life of the Aryans, who, moreover, should not be considered to have been a monolithic group. That some customs or ideas we observe at a later period are absent in the early Vedic texts is no proof of their absence at that time, much less of their non-Aryan provenance.

c5.s2.13

⁴ See, for example, S. K. Belvalkar and R. D. Ranade, *History of Indian Philosophy*, vol. 2: *The Creative Period*, 2d ed. (Delhi: Oriental Books Reprint Corporation, 1974), 79–81, 400–403; G. C. Pande, *Studies in the Origins of Buddhism* (Allahabad: University of Allahabad, 1957), 251–261; G. C. Pande, *Śramaṇa Tradition: Its History and Contribution to Indian Culture* (Ahmedabad: L. D. Institute of Indology, 1978). S. Singh, *Evolution of the Smṛti Law: A Study in the Factors Leading* *to the Origin and Development of Ancient Indian Legal Ideas* (Varanasi: Bhāratīya Vidyā Prakāśana, 1972), 176–185.

c5.s2.14

A culture, furthermore, develops and changes over time not only through external influences but also through its own inner dynamism and because of social and economic changes. New elements in a culture, therefore, need not always be of foreign origin.

c5.s2.15

In the absence of an adequate definition of orthodoxy within the Brāhmaṇical tradition of this period, moreover, 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, as we shall see, across a broad spectrum of religious literature, including some of the most authoritative texts of Brāhmaṇism. Opposition to accepted views of a tradition, after all, can and often does come from within that tradition. Thus, when I assert that renunciation was a "new" element, I do not mean that it was either non-Aryan or even non-Brāhmaṇical.

c5.s2.16

The subtitle of this chapter—"The Inner Conflict of Tradition"—is a reference to the conflict between societal and renunciatory values and ideas that in one form or other persists throughout the tradition, a conflict observable in the Saṃnyāsa Upaniṣads themselves.⁵ In what follows I want to explore the background of and possible reasons for the historical emergence of renunciatory ideologies and their initial conflict with the Vedic world.

c5.s2.17

⁵ I have deliberately adopted the tile from Heesterman's (1985) volume. Heesterman sees it primarily as a conflict between the renunciation-oriented Brahmin and the this-worldly king. I hope to show that the deeper conflict within Indian religions is between the renunciatory life and ideals (whether they are espoused by Brahmins or Buddhists) on the one hand, and the socioritual order sponsored by priests and rulers alike. This conflict, moreover, is not between concrete individuals but between ideal types. The conflict, therefore, did not historically prevent alliances between rulers and renunciatory sects; clearly other-worldly people and ideas can be and have been put to good this-worldly use. See, for example, my "The King and the Ascetic: State Control of Asceticism in the *Arthaśāstra*," *The Adyar Library Bulletin* 51 (1987): 39–59.

2.1 Vedic Theology: Sacrifice and Marriage

c5.s2.t1.1

The Vedic religion had reached great complexity and sophistication by the sixth century B.C.E. Its central element both practically and theologically was the Vedic sacrifice. By this time it had become extremely intricate and expensive, requiring an array of ritual specialists. The development of the Vedic sacrificial theology paralleled that of the sacrifice; finding ever-new meanings and cosmic significances in ritual acts, this theology created both a cosmology and a soteriology centered on the sacrifice. The Vedic sacrificial theology was based on two significant claims. First, creation itself resulted from the sacrifice. Second, gods themselves attained immortality⁶ through the sacrifice, setting an example for humans to follow.

c5.s2.t1.2

⁶ The exact meaning of immortality (*amṛta*) in these ancient texts is not clear. As Collins (1982, 42–44) points out, in this context the term probably means freedom from death and the indefinite continuation of existence, a meaning quite different from the deathless state envisaged within the later concept of *mokṣa* ("absolute freedom"), a state that transcends both life and death. In the latter view, life and death are inseparable correlatives constituting our present state of *saṃsāra* (“life subject to repeated births and deaths”), a concept that by definition assures the indefinite continuation of existence.

c5.s2.t1.3

The *Ṛg Veda* (10.90) in the celebrated Puruṣa hymn presents the sacrifice as the primordial cosmogonic principle. The Brāhmaṇa literature came to identify the sacrifice with a new and somewhat artificial creator god, Prajāpati, which identification became the cornerstone of later sacrificial theology. Prajāpati is both the sacrifice personified and the primordial performer of the sacrifice. From Prajāpati as sacrifice and sacrificer flows the entire creation: heaven and earth, gods and humans.⁷

c5.s2.t1.4

⁷ For an accessible discussion of the Brāhmaṇical theology of sacrifice and for further bibliography on the subject, see Kaelber 1989.

c5.s2.t1.5

As the fountain of creative power, the sacrifice is also the source of immortality. The gods were the first to obtain the knowledge of the sacrifice. By performing it correctly they attained immortality. The Vedic seers discovered that same knowledge, and the Vedas are its repository. The sacrifice, then, is the source of immortality for humans, as it was for gods. It is connected, moreover, with the archetypal activity of the creator god Prajāpati, as well as of gods and seers. This activity, recorded in myth and reinforced by Brāhmaṇical theology, therefore, becomes doubly paradigmatic for humans.

c5.s2.t1.6

The obligation to imitate the gods by performing sacrifices is grounded also in a practical concern: humans seek to be like gods, to attain the world of gods, in other words, to become immortal. Brāhmaṇical theology, however, also gives that obligation a cosmic dimension. As the cosmos first emerged from the sacrifice, so it must be continually recreated and renewed through the sacrifice if it is not to lapse into chaos and death. The king, as the hub of the social cosmos and as the chief sacrificer in society, occupies a central position in the sacrificial recreation of the cosmos. As Heesterman observes:

c5.s2.t1.7

. . . the rājasūya [“royal consecration”] seems to be an abridgment of what originally must have been an unremitting series of yearly ceremonies with the object of regenerating the universe.

c5.s2.t1.8

The king took a central place in it. The universe had yearly to be recreated and so had the king who, like the common śrauta [“Vedic”] sacrificer, incorporated the cosmos.⁸

c5.s2.t1.9

⁸ J. C. Heesterman, *The Ancient Indian Royal Consecration*. Disputationes Rheno-Trajectinae, no. 2 (The Hague, 1957), 10.

c5.s2.t1.10

In the Vedic sacrifice, therefore, Brāhmaṇical theology combined the cosmological and the soteriological; the individual and social obligation to perform sacrifices was thereby doubly reinforced. The focus of this obligation fell in a special way on the daily fire sacrifice called *agnihotra*. Most Vedic sacrifices were offered on special occasions or on specific days of the ritual calendar. Given their elaborate nature and the enormous expense they entailed, few in all likelihood could have afforded to perform the major Vedic sacrifices. The relatively simple daily offering of *agnihotra* thus began to assume an ever-increasing importance among theologians as the paradigmatic sacrifice. The obligation to offer it daily assumed a central position within the sacrificial ethic; a person has the duty to offer it every day during his entire life. As one text puts it:

c5.s2.t1.11

It is assuredly a long sacrificial session upon which they enter who offer the *agnihotra*. This is indeed a session to be broken off only through old age. For one is absolved from it either by old age or by death.⁹

c5.s2.t1.12

⁹ ŚB 12.4.1.1; *Jaiminīya Brāhmaṇa*, 1.51.

c5.s2.t1.13

A corollary to the obligation to offer the daily fire sacrifice is the duty to maintain one or more of the sacred Vedic fires burning at all times in one’s house. These ritual obligations defined the Brāhmaṇical ethic regarding normative behavior encapsulated in the term *dharma*, which for all Indian religions indicated the “right way” of both behavior and ideology. From a symbolic and sociological point of view, the presence of a sacred fire at home placed that household squarely within the Brāhmaṇical *dharma*.

c5.s2.t1.14

A second and less obvious corollary of sacrifice is the obligation to marry, for only a married man accompanied by his legitimate wife is qualified to be a sacrificer. The two constitute a single sacrificial persona. “A man who has no wife,” says the *Taittirīya Brāhmaṇa* (2.2.2.6), “is not entitled to sacrifice.” The intimate and unbreakable union between husband and wife in the sacrifice, the *Śatapatha Brāhmaṇa* (12.8.2.6) says, is like that between truth and faith, and between mind and speech: wherever the husband is, there also is the wife.

c5.s2.t1.15

Next to sacrifice, therefore, the obligation to get married and procreate children (especially a son) is central to Brāhmaṇical theology, which regards the family—father, mother, and son—as the only complete person. The wife is said to be one half of the husband, because so long as a man is without a wife he remains childless and incomplete.¹⁰ “Therefore, having found a wife,” says the *Aitareya Āraṇyaka* (1.2.5), “a man considers himself to be, as if, more complete.”

c5.s2.t1.16

¹⁰ TS 6.1.8.5; ŚB 5.2.1.10.

c5.s2.t1.17

Apart from her role in ritual activities, the wife’s importance is directly related to her being the mother of her husband’s son. It is, however, not just the child that she begets; the husband himself is born again in her as the son. This rebirth and continuation of the husband is said to be the true wifehood of the wife:

c5.s2.t1.18

The husband enters the wife;
Becoming an embryo he enters the mother.
In her become a new man again,
He is born in the tenth month.¹¹
A wife is called wife,
Because in her he is born again.¹²

c5.s2.t1.19

¹¹ In ancient India the period of human gestation was taken to be ten months.

c5.s2.t1.20

¹² AitB 7.13. The authors play on the word *jāyā* (“wife”), which is derived from the verbal root *jan* (“to be born”). The later law book of Manu (9.13) says: “The husband, after conception by his wife, becomes an embryo and is born again of her; for that is the wifehood (*jāyātva*) of a wife (*jāyā*), that he is born (*jāyate*) again by her.”

c5.s2.t1.21

“The father,” says the *Śatapatha Brāhmaṇa* (12.4.3.1), “is the same as the son and the son is the same as the father.” Consequently, one should desire a good wife, thinking "let my self come into existence in something good," and guard her continuously "lest in my womb, in my world somebody else come into existence" (*Jaiminīya Brāhmaṇa*, 1.17).

c5.s2.t1.22

The cardinal point in the Brāhmaṇical theology of religious life that the son constitutes the immortality of the father is already expressed in the Ṛg Vedic (5.4.10) prayer: "Through offspring, O Agni, may we attain immortality." The Brāhmaṇas never tire of repeating this: "In your offspring you are born again; that, O mortal, is your immortality" (*TB*, 1.5.5.6).

c5.s2.t1.23

The Brāhmaṇical conceptions of immortality as freedom from physical death and of the family as the true and complete person are reflected in the belief that a man's immortality is found in his son. The family line continues in the son despite the death of the father. As the son survives after the father's death, so the father in his son survives his own death. Upon his death, the son takes the place of the father as the ritual head of the household.¹³ This appears to be the meaning of the statement that a father is born again in his son. This new birth frees him from the death that must eventually end the life begun at his first birth. In a very significant way, therefore, the family is what guarantees human immortality.

First 107 of 545 passages. Read on

Linked texts (56)