This paper begins with the identification and analysis of some earliest textual references to Pāśupata ascetics, their tenets, and behaviours. Then it inquires into the genesis of Pāśupatism by analysing some critical passages of the Pāśupatasūtra, going beyond Kauṇḍinya’s Bhāṣya. It analyses relevant passages from the Jaiminīyabrāhmaṇa, Mahābhārata, and Brahmāṇḍapurāṇa, and shows how the Vedic govrata or anaḍudvrata has been first adopted and then adapted in Pāśupata tradition, and how Indra, a deity associated with the original vow, comes to be depicted as the primaeval observer of the newly defined pāśupatavrata. It argues further that the conception of the celestial bull as a divinity and the idea of imitating the bull’s behaviour to please that divinity are at the heart of the Pāśupata praxis, in all stages of its development. This paper also argues in favour of recognising “Megasthenes’ Heracles” as Indra, and the Sibae people mentioned in his report as the Śibis.
It is only in the latter-half of the fourth century ce that some inscriptions related to Pāśupatism appear in Bāgh and Mathurā.1 Early textual references to the Pāśupatas, and particularly to their tenets, too, are extremely rare. It is only in the Mukha section of the Niśvāsatattvasaṃhitā that a large number of sūtras from the Pāśupatasūtra, the highest authority of the Pāśupatas, are paraphrased.2 Next to this, we can mention the Madhyamakahṛdaya of the sixth century Buddhist scholar Bhavya where a few of the Pāśupata sūtras are paraphrased and briefly criticised.3 The original Skandapurāṇa, which most probably belongs to the same period, is also aware of the Pāśupata vow and Yoga.4 The Mahābhārata does not contain any clear reference to the Pāśupata vow or knowledge except in the latter part of the Nārāyaṇīya section.5 In the present state of research, this is all that we know from early times, let us say, before the seventh century.6
In the first part of this paper I shall collect a few early references to the Pāśupata ascetics, their tenets and behaviours as well, either implicit or explicit, which I have not seen reported anywhere so far. These references come from Bharata’s Nāṭyaśāstra, which is generally placed between 200 and 400ce, the Lalitavistara (ca. 400ce), and Vasubandhu’s Abhidharmakośabhāṣya, which also can be placed around 400ce.7
Early Textual References to Pāśupatas
Let me first take two references from the Nāṭyaśāstra referring directly to the Pāśupatas. These simply tell how the ascetics of different orders, including the Pāśupatas, should be presented in a drama:
8In cases of the ascetics of the Brāhmaṇa, Jaina, and Buddhist order, and also the Yatis and Pāśupatas, the dress should be made as it suits their respective observances and [also] matches their disposition in the world (lokasvabhāvajaḥ). One should prepare Tāpasas in tatters, bark, and hide. The clothes of ascetics of the Parivrājaka, Brāhmaṇa and Buddhist order are prescribed to be dyed in brown-red colour, but for the Pāśupatas many clothes of various colours are required.
9The gait of the Pāśupatas should be characterised by proud moving.
Apart from these two passages, there is at least one implicit description of the Paśupata ascetic in the Nāṭyaśāstra. When the text tells how to act out madness, it sounds very much like the text has Paśupata ascetics in view; for it speaks of smearing ashes and using, carrying, and decorating oneself with used garlands (nirmālya)—traits closely connected with the Pāśupatas. Here is the passage:
Madness, of course, arises due to the [adverse] determinant emotional conditions (vibhāva) like separation from desired persons, loss of property, injury, excess of [any or all of] the three [corporeal humours:] wind, bile, and phlegm. One should act it out by ways of causeless laughing, weeping, and crying out loud; by speaking nonsense, [now] lying down, [then] sitting, standing up, running, dancing, singing, [and] reading; by smearing ashes and dust on the body, by using, carrying, and decorating oneself with grasses, used garlands, filthy clothes, rags, clay pots, bowls, and platters; with [these] many unsettled movements and imitations which are [in this context] the consequent emotional conditions (anubhāva).10
Now I turn to the Lalitavistara. Here Pāśupata ascetics are not explicitly mentioned, but when the text complains about wrong notions of those people it regards as deluded, we cannot fail to notice that some of these notions are exclusively related to the Pāśupatas or their skull-bearing Kāpālika offshoots.
By annointing onself with ashes, ink,11 the ‘dark’ substance of used garlands, dust, soil, and mud; by growing body hair, mustache, hair, and nails; by bearing rags, and by carrying skeleton and skull, … by carrying charcoals, minerals, red robes, and three staffs; by keeping the head shaven, and by carrying a water-pot of certain style, skull-cups, and skull-staff, the[se] deluded people hold that purity is achieved.12
These people annointing their body with ashes and also making use of used garlands, particularly in this period of time, must be Pāśupatas.13
All these references give us an impression of the outward appearance of Pāśupatas,14 but those from the next text, the Abhidharmakośabhāṣya, refer mainly to their ideology and behaviour, and so are more significant. This text makes the following remark, when it explains that various notions arise from the thirst caused by one’s feelings regarding good desirable things (iṣṭavedanātṛṣṇā). Among them are the notions of ‘conduct’ and ‘observance’. As it explains:
15 ‘Conduct’ means abstinence from bad conduct, [and] ‘observance’ means the observance of [behaving like] a dog, or a bull, or the like. And [also], as the Nirgrantha and other like-minded ascetics [say]: ‘[an ascetic] becomes naked, does not have any cloth.’ This is an elaboration [going beyond the main statement].
[It also includes] adoption of [the rule of] holding a staff and a hide, [that of] keeping matted hairs and smearing ashes, [and that of] keeping a set of three staffs and shaving the head, and of other similar ones, seen among the brahmins, Pāśupatas, and Parivrājakas, and other similar groups [respectively].16
Here is another passage from the same text (bhāṣya on 2.64d) which criticises a position that holds ‘the Lord’ to be the cause of the world. Vasubandhu appears to be satirical here; he cites a verse from some authoritative text of the criticised group without naming it,17 and mocks their Lord Rudra for his fierce and ghastly aspect:
18 [The world does] not [arise] from the Lord or the like, because of the sequence of [things] and other similar factors.
If the sole cause [of the world] were the Lord or anything else, the entire world would have come into existence at once. But we can observe that things arise in a [gradual] sequence. […]
And if the Lord creates people afflicted with a lot of plagues in places like the hell, and gets pleased with that, then let there be obeisance to the Lord [with full awareness that he is] of such nature! And, here is a well-stated verse about him:
Since he burns down, since he is sharp, since he is violent, since he has heat, since he is the consumer of flesh, blood and bone-marrow; therefore, he is called Rudra.19
Moreover, as is evident, if one [entity] is accepted as the cause of the world, then the clearly visible individual contribution (puruṣakāra) of other entities is denied. If one postulates the Lord as a causative agent along with the causes, this would merely be a statement made in devotion, because the Lord, being different from the causes, is not seen to have any function in the rise of the [world]. …
As we know from the previous passage that Vasubandhu definitely knows the Pāśupatas, we are in a relatively secure position to say that the people criticised in this passage for holding Rudra to be the cause of the world are the Pāśupatas. However, if we rely on the commentary of Yaśomitra, this Lord could be Mahādeva, Vāsudeva, or Puruṣa or others. There is no doubt that Yaśomitra wants to engage here all those groups of people who believe in this or that form of the Lord—even Vasubandhu might have intended so, but since Vasubandhu gives a ghastly description of Rudra and mocks him, we may say that he is aiming mainly at the Pāśupatas.
Once again, when Vasubandhu needs to criticise people’s ‘attachment to specific modes of conduct and observances’ that stems from their viewing of a non-cause as the cause and a non-path as the path, he remembers the same people, who regard Maheśvara as the cause (hetu) of the worlds. As he says (Abhidharmakośabhāṣya 5.7–8):
20 Viewing a non-cause as the cause and a non-path as the path defines [one’s] ‘attachment to [specific modes of] conduct and observances.’ For instance, Maheśvara is not the cause of worlds but a man regards him as the cause; or else, Prajāpati or somebody else [he regards so]. Entering into fire or water and so on are not the cause of [attaining] heaven but man takes them as the path [leading there]. Mere modes of conduct and observances as well as the knowledge of Sāṃkhya, Yoga and the like do not make the path of liberation, but a man regards them as the path. …
Also this [attachment], which makes one seek the attainment of heaven by entering into fire or water and so on, or [seek] purity through certain conducts and observances, has to be abandoned by means of realising sorrow [as the nature of everything in the world]. For, this is the text of the [Jñānaprasthāna]-śāstra:21 “Furthermore, these [are] the people, who have this kind of view and speak this way, ‘[as soon as] this personal-soul lives, taking on a bull’s behaviour, [or] the behaviour of a wild animal, [or] the behaviour of a dog, with that he is purified, is liberated, [and] goes beyond happiness and sorrow, [namely,] attains transcendency beyond happiness and sorrow.’ If someone regards a non-cause as the cause, [then he has] attachment to conducts and observances, and [this view] should be abandoned by means of realising sorrow [as the nature of everything in the world].” This much is an elaboration [going beyond the main statement].
What is mentioned as govrata in the first passsage of Vasubandhu I cited earlier is gośīla in the Jñānaprasthāna excerpt contained in the above passage, and what is called gośīla and mṛgaśīla in the latter is called godharma and mṛgadharma in Pāśupatasūtra 5.18. However, it is clear from their respective contexts that the same observance is intended with three different but nearly synonymous terms govrata, gośīla, and godharma. It appears that the Pāśupatas have much to do with this vow but it cannot be limited to them and until now no positive statement connecting this vow and the Pāśupatas had come to light.
Pāśupatasūtra 5.18 prescribes the godharma and mṛgadharma as an essential part of the Pāśupata vow. In the scheme of Kauṇḍinya or possibly even in that of the version of the Pāśupatasūtra available to him and handed down to us, this observance appears to be a part of the final stage of the Pāśupata vow. Moreover, Kauṇḍinya tells us that Pāśupatasūtra 5.18 intends to prescribe adoption of the welcome aspects of the bull’s or a wild animal’s behaviour. The reality looks different if we read the corpus of Pāśupatasūtras critically and carefully and without being blinkered by the commentary of Kauṇḍinya. Therefore, in the next section I will gather available hints, and reading the Pāśupatasūtras this way, investigate how crucial the vow of taking on the behaviour of a bull is in the development of Pāśupata asceticism.
The Archaic Govrata as the Pāśupata Vow
I find that the Pāśupatasūtra does originally prescribe that a Pāśupata should adopt the bull’s behaviour from the beginning till the end of the Pāśupata vow. He has to attend the Lord everyday with an offer of the bull’s bellowing (huḍuṅkāra) among other things (see PS 1.8 and Kauṇḍinya’s Bhāṣya thereon). In the final stage, too, he has to adopt the behaviour of a bull or a wild animal.22 However, in the intermediate stage of his vow, according to Kauṇḍinya’s reading of the related sūtras (PS 3.11–15) and his interpretation of them, a Pāśupata has to enact certain actions so that people abuse him. Abused this way and suffering therewith, he takes away their righteous deeds and transfers his bad deeds to their account.23 Now, these actions do not at first blush appear related to the bull’s behaviour, but we should not hastily turn this first impression into a conclusion.
As Oberlies (2000: 178) has already pointed out, a very similar set of prescriptions meant for the same purpose of attracting others’ abuses and thus trasferring one’s bad karman to the abusers is found in the Taittirīyabrāhmaṇa (II.3.9.9). There one can find parallels for some of the Pāśupatasūtras prescribing a set of actions for the Pāśupatas in the intermediate stage. Oberlies has briefly dealt with two of these sūtras: pretavac caret and krātheta vā (Pāśupatasūtra 3.11–12). He appears to suggest that Kauṇḍinya’s reading as well as interpretation of the first sūtra is secondary, and so is Kauṇḍinya’s interpretation of the second.24 But now I believe that Kauṇḍinya’s reading as well as interpretation of the whole set of sūtras 3.11–15 involves revision and confusion too.
Before I elaborate, let me make it clear that I am not saying Kauṇḍinya has misinterpreted these sūtras—the way Kauṇḍinya interprets them was most probably the way they were understood in his tradition. What I want to say is that these sūtras originally meant specific things related to a specific vrata, but as soon as Pāśupatism was presented in a modified and philosophised way, perhaps long before Kauṇḍinya, the original meaning was forgotten.
Let me first discuss Kauṇḍinya’s reading of these sūtras. The reading of Pāśupatasūtra 3.11 is pretavac caret, but its parallel in the Taittirīyabrāhmaṇa passage reads préva calet. The Taittirīyabrāhmaṇa reading is straightforward for a reader aware of pre-classical usages, while Kauṇḍinya must resort to secondary meanings while interpreting the sūtra reading. In this situation, as Oberlies has suggested, we can suppose that originally the sūtra reading was not different from the Taittirīyabrāhmaṇa reading, and that it was eventually changed into the present reading: pre
It is clear from the context that all sūtras in this set are concerned with the simulated behaviour, and the Taittirīyabrāhmaṇa parallel reveals as much. By necessity, such behaviour calls for enactment rather than true engagement. In Pāśupatasūtra 3.11, -vat of pretavat can somehow suggest enactment of the prescribed action, but in other sūtras a sense of enactment is missing. I therefore propose that vā in each of the following four sūtras 3.12–15: krātheta vā, spandeta vā, maṇṭeta vā, śṛṅgāreta vā, was originally iva, exactly as in the Taittirīyabrāhmaṇa parallel. Orthographically, it is very easy to explain the shift from krātheteva together with a daṇḍa to krātheta vā and so on.26
The last of these sūtras (3.15) reads śṛṅgāreta vā but its parallel in the Taittirīyabrāhmaṇa passage reads śṛṅgāyeteva. The Pāśupatasūtra reading here is quite problematic. First, śṛṅgāreta as a denominative optative is irregular. Second, even the noun śṛṅgāra is not attested in the Vedic corpus, and so, the form śṛṅgāreta cannot go back a long way in time. On the other hand, the denominative root śṛṅgāy- is regular and fits the original context of the adoption of the bull’s behaviour. As in the previous case, it seems likely that the available sūtra reading results from a revision of the original text which was closer to the Taittirīyabrāhmaṇa reading, in a redefined context.
Now let me venture to restore these sūtras (PS 3.11–15) to their archaic form, and interpret them as they might have been understood in an earlier phase:
*preva caret, krātheteva, spandeteva, maṇṭeteva, śṛṅgāyeteva.27
‘[The ascetic] should enact thrashing about, he should enact injuring [others], he should enact kicking or twitching of his limbs, he should enact getting agitated/hobbling, he should enact butting.’
These prescriptions correspond indeed with a bull’s behaviour. The first prescription is neutral, but, together with the following actions, it can easily be associated with a bull. In a different cultural context, the bull is well-known for thrashing about, and injuring people28 in narrow streets. The third prescription, spandeta, has been rightly interpreted by Kauṇḍinya in the sense of twitching of limbs, but he adds that this action is enacted with thought, will, and effort. This enhanced interpretation might well have been valid at the time of Kauṇḍinya29 but not at an earlier age. In fact, as Geldner (1951: I.421) has pointed out, the root spand in early Vedic texts is used only of the kicking of a cow or bull.30 Even when a new meaning developed, it did not depart much from this original meaning and remained associated with behaviour of cattle: it is their twitching of limbs. They are twitching their limbs all the time in order not to allow the biting insects. Thus, whether it denotes kicking or twitching, the root spand is primarily associated with cattle,31 and inclusion of this particular action in the set of prescriptions makes our case very strong. This original meaning now goes well with the preceding prescription of imitating injuring people, and also with the following ones.
The next prescription in the row is somewhat unclear. The root maṭ/maṇṭ is not attested anywhere else except this sūtra and the Taittirīyabrāhmaṇa passage parallel to it. Kauṇḍinya interprets it in the sense of hobbling, whereas the Taittirīya commentators, Bhāskara and Sāyaṇa, take it in the sense of getting agitated.32 This meaning fits the bull’s behaviour. Moreover, I find that the whole set of actions prescribed here involves some kind of aggressive behaviour. This argument, too, is in favour of the interpretation of Bhāskara/Sāyaṇa. The last of the prescribed actions, butting with horns, does not need any interpretation as it involves the typical culmination of a bull’s agitation.33 Thus, we can be sure that, following these prescriptions, a Pāśupata ascetic is supposed to imitate all these actions, including butting with his head as if he has horns, and to behave like a bull. This analysis makes it clear that imitating the bull was something prescribed for more than one stage of the Pāśupata religious life. This compells us to consider that the Pāśupata practice started in its original form as some sort of godharma, or even as paśudharma if generalised.
The Pāśupatas ritually adopted the bull’s behaviour, regarding themselves as the cattle of their Lord, and thus cultivated devotion to Rudra, ‘the Lord of Cattle’ (paśupati). Originally, this must have been their intention in all ways and throughout all phases of their life after accepting Pāśupatism. This was true at the time of composition of the Pāśupatasūtra. But by the time of Kauṇḍinya the Pāśupata vow had somehow become moderated and divided into stages, and the godharma observance was circumscribed and attached only to the final stage of Pāśupata practice. The prescriptions requiring one to adopt the bull’s behaviour were transformed into something suitable to the modified notion of Pāśupatism. Consequently, what was practiced in the initial and intermediate stages was no longer recognised as godharma.
How to Behave like a Bull?
We have thus managed to reveal that the godharma observance underlies all stages of Pāśupata practice. Armed with this finding, let us now investigate what might be the meaning intended by the Sūtrakāra of godharma. As I mentioned before, Pāśupatasūtra 5.18 requires an accomplished Pāśupata practitioner (siddhayogin) to adopt the behaviour of a bull or a wild animal in the final stage of his vow, but neither the sūtra text nor Kauṇḍinya’s Bhāṣya give the details of this behaviour. Rather, Kauṇḍinya says that what the sūtra intends to prescribe is the adoption of the welcome aspects of the bull’s behaviour but not the natural beastly aspects. This defensive attitude of Kauṇḍinya only makes his case more suspicious. In reality, the sūtra text simply enjoins the adoption of the behaviour of the bull. Unless we have some hint from elsewhere, it is impossible to decide what is implied and what is not. Let us therefore turn elsewhere to gather some information on this vow of the bull. I start with a passage from the Mahābhārata (5.97.12–14) which, describing some pātāla, relates Bhūtapati Maheśvara and the observers of the govrata, and then provides a brief definition of that vow:
Here did Bhūtapati, the great lord (Maheśvara) of all living beings, practise the highest austerity for the sake of the prosperity of all living beings. Here do brahmins practising the govrata, emaciated by their studies and customs, die and win heaven, to dwell [there] as great sages. [A brahmin] is invariably lying just anywhere, is fed with just anything [by way of food], and covered with anything [by way of clothing]; he is called in this [pātāla] ‘one practising the govrata.’34
At this point, I would like to reiterate that there is only one reference to the Pāśupata knowledge in the main text of the Mahābhārata and that it falls in the latter part of the Nārāyaṇīya section. Otherwise, the Mahābhārata does not bear witness to the Pāśupatas or their dogmas (but see footnote 5 above). Given this near silence, the significance of the above passage attesting to the group of brahmins who observed the govrata is all the higher. In view of their behaviour, we must identify these brahmins as the forerunners of the Pāśupatas. We may conjecture that the Mahābhārata does not know the Pāśupatas as Pāśupatas but as a group of brahmins observing the govrata to please Maheśvara. In this connection, we should not forget that as late as the last quarter of the fourth century ce, in the Mathurā pillar inscription, our Pāśupatas are still called Māheśvaras.
The above passage is significant also for the definition of the govrata practised by these brahmins. There is another “purified version” of govrata described in some Purāṇas and Dharmaśāstra texts which amounts to dedicating oneself to the service of cows,35 but the definition of govrata in the passage above, though it is very much generalised, suggests something quite different. For more details of the original govrata, I propose to read a passage from the Brahmāṇḍapurāṇa.
Let me start with the background story needed to understand the passage with which we are concerned: Bṛhaspati, the priest of the gods, by force slept with his elder brother Uśija’s wife when she was pregnant. When he was discharging his semen, the son of his brother in the womb asked him to stop. This angered the uncle and he cursed the baby in the womb that he would be born blind. Then, because of the curse, the blind seer who would be named Dīrghatamas was born.36 He had no family of his own, and so, lived in the house of his brotherly cousin Śaradvat, a son of Bṛhaspati.
The story of Dīrghatamas’ birth is narrated in the Mahābhārata (I.98.6–32) and also in the Bṛhaddevatā (4.11–23), though, in these texts, the elder brother of Bṛhaspati has been named as Utathya/Ucathya. The story in these texts contains two salient points: Dīrghatamas was born blind due to the curse of his uncle Bṛhaspati, and when he was old he was thrown into a river by his own people. In addition to these points, the Mahābhārata states that Dīrghatamas was rescued by King Bali and was put to produce children for the king from his queen, while the Bṛhaddevatā tells that the sage produced kṣatriya children after the river threw him on to the bank near the country of Aṅga. In the Brahmāṇḍapurāṇa, however, the same story is told with an additional plot that explains why people had boxed and thrown Dīrghatamas into a river. This additional plot is of interest to us here:
37 Then that son of Uśija, who was widely talked about and was like Bṛhaspati in virility, had no family of his own (ūrdhvaretās); and all the more for that reason, he lived in the house of his brother[ly cousin]. He received [there] the teaching of godharma from the lordly bull, a son of the divine cow Surabhi.
His brother[ly cousin from] the paternal uncle (bhrātā pitṛvyaḥ) then built a house.38 While he was living there, one day an unexpectedly arrived bull, offspring of Surabhi, fed on the bunch of kuśa grass spread on the ground for the purpose of the new moon sacrifice. Dīrghatamas soon held that bull by its horns as it was struggling. Held by him, it was unable to move from one step to the other.
39 Then the bull spoke to him, “O best of mighty men, please release me! O my son, though I have been carrying the god Tryambaka, since I have come to the earth I have not encountered a mighty man like you anywhere! Please release me and choose a boon in return for your friendship.”
Upon being spoken to thus [the man] replied, “Being alive, [when let loose], you will injure me.40 Therefore I will not let you go, you four-footed devourer of others’ property!”
Then the bull replied to Dīrghatamas: “My dear, we have neither fatal sin nor theft. We do not distinguish at all what is to be eaten and drunk, and what is not. And, o brahmin, we truly do not [distinguish] what should be done and what not, nor who is fit for sexual relation and who not. We are not sinners, o brahmin, because all this is known from the tradition as the nature of bulls.”
41 Having [thus] heard about bulls, Dīrghatamas got really confused and released him. Then, while pleasing that son of cow with Vedic devotion,42 he accepted the ‘vow of the bull’ by the grace of that king of bulls. Agreed and devoted to it, he took that [teaching] to heart. Then, as his mind was dull due to destiny, he lusted after the wife of Autathya Junior (yavīyas),43 while she was struggling and weeping. Regarding this act as a sign of haughtiness, Śaradvat did not excuse him, because Dīrghatamas had lusted after a daughter-in-law on the strength of godharma. Regarding this as an act of misapprehension, having reflected on it, and foreseeing what was going to happen,44 the great soul (Śaradvat) despised45 Dīrghatamas, and with his eyes red in anger told him: “As you harass a daughter-in-law on the strength of godharma, you do not know who is fit for sexual relation and who is not. I here disown you as a wicked man! You go your own way. Since you are old, blind, dependent, and wicked, therefore you are abandoned. I regard you as a wicked man!”
Then he had the idea of doing this cruel act: after rebuking him in various ways and cuffing him with both hands, he put him in a wooden box and threw [it] into the waters of the river Ganges.
This passage makes clear that the vow of godharma involves breaking of dietary and sexual restrictions with the intention of going beyong morality. But it appears that there, too, some rules were still to be observed. Otherwise, Śaradvat would not have complained that Dīrghatamas had made a mistake by approaching his daughter-in-law.
Another important piece of information obtained from this passage is that the observance of godharma is linked with Vedic ritualism. Following this clue, as I explored the Vedic literature, I have found in the Jaiminīyabrāhmaṇa the following detail about the observance of the govrata that is attached to the Vedic gosava sacrifice:
46 The observance of that [one-day Gosava sacrifice] is [this]: He should [sexually] approach (upa+iyād) his mother, sister, or a lady of his own clan (sagotra). Having leaned down close [to the source] he should sip water [directly with his mouth], and having leaned down close [to the ground] he should cut grass [with his teeth].47 Wherever he feels the urge to evacuate faeces, right there he should evacuate. He wins the world of the [celestial] draft-ox.
Janaka Vaideha once wished to perform this [sacrifice]. As it is reported, brahmans then were sitting in front of him [in the court]. He asked them: “What is the stoma [for it]?”
Sudakṣiṇa, the son of Kṣema, told: “[Experts] say it is not only the threefold Stoma!”48
“What is its observance? What is the sacrificial gift?”
He explained [all] this to him.
The other said: “I definitely wish [to pay] the gifts but I do not wish for its observance.”
Thus he did not dare to perform that sacrifice. But Puṇyakeśa, son of Yodhena, the king of the Śibis, performed this sacrifice. In the assembly hall he felt the urge to evacuate faeces. Evacuating it right there he said: this sacrifice was really meant for an elderly man.
Only an elderly man should perform this sacrifice. For, all this is permitted to an elderly man. This is definitely a sacrifice [meant] for an elderly man. One should perform this [sacrifice] in the final years of one’s life.
Now we know that breaking of sexual and dietary restrictions was an essential part of this Vedic vow, and also that, given the troublesome nature of the vow, it was restricted to the final years of one’s life. According to Mylius (1976), this advice was ignored by the Śrautasūtras as soon as it became the king’s sacrifice. However, this advice accords well with the inclusion of the godharma in the final stage of the Pāśupata vow. I think it probable that crossing the boundary of sexual morality was symbolically performed only in old age or the final stage of the vow, even if the other rites of the govrata were carried out life-long. As I mentioned before, there is a hint in the Pāśupatasūtra as to the nature of this vow. As sūtra 5.20 claims, an accomplished Yogin is not defiled by any of his actions, nor by any of the fatal sins (pātaka). A statement like this would not have been made if trespassing against normal ethical rules were not involved in this vow. It appears that some secret of the vow of godharma was known even to Bhavya, a sixth-century Buddhist scholar, as he cites the same sūtra and sharply criticises the Pāśupatas for their wrong behaviour. Let me cite him here:
Savants regard what is sufficient for burning the kleśas as the [real] knowledge. Thus, a man of knowledge cannot commit sin, because the cause of sin does not exist in him. [However,] these people who say that an accomplished Yogin is not defiled by any action or even by any of the fatal sins, they have themselves fallen off the right path and have made others, too, to fall off.49
The Śibis, Govrata, and Megasthenes’ Report
In the Jaiminīyabrāhmaṇa passage cited above, we find that Vaideha Janaka declines to perform the sacrifice once he becomes aware of the observance associated with it, but that the King of Śibis comes forward and observes it. We can understand that it is natural for Vaideha Janaka, as a representative of Vedic orthodoxy, to refuse performing a sacrifice that is bundled with a vow of a socially problematic nature, but it is not so easy to infer why the king of Śibis came forward to perform it. For this, we should listen to the Greek historian Megasthenes who provides the following information:
They said also that Sibae were descended from those who accompanied Hêrakles on his expedition, and that they preserved badges of their descent, for they wore skins like Hêrakles, and carried clubs and branded the mark of a cudgel on their oxen and mules.50
So also when they came among the Sibai, an Indian tribe, and noticed that they wore skins, they declared that the Sibai were descended from those who belonged to the expedition of Heraklês and had been left behind: for besides being dressed in skins, the Sibai carry a cudgel, and brand on the backs of their oxen the representation of a club, wherein the Makedonians recognised a memorial of Heraklês. But if any one believes all this, then this must be another Heraklês,— not the Theban, but either the Tyrian or the Egyptian, or even some great king who belonged to the upper country which lies not far from India.51
From these passages, we learn that the Macedonians had seen the people of Sibae/Sibai tribe dressed in skins, carrying a club, and their cattle branded with the mark of a club. The Jaiminīyabrāhmaṇa passage cited before tells us that it was a king of the Śibis who came forward to perform the govrata. Now if we put these two facts together, we can say that the Śibis were typical cattle-raisers, who could somehow be called proto-Pāśupatas after their character, but in view of Vedic orthodoxy they were Vrātyas. They observed the govrata and worshipped some deity who carried a club in his hand by mimesis, whom the Macedonians wanted to identify as Heraklês because of the club and his temple as ‘a memorial of Heraklês’.52 They imitated their Lord and carried themselves a club, and also branded the mark of a club on the backs of their cattle.
Indra is the Deity of the Śibis
Now the question arises: who was the deity venerated by the Śibis in the so-called memorial of Heracles? Let us first try arguing in favour of Rudra. I have already shown how close is the behaviour of the Śibis with the original Pāśupata vow modelled after the Vedic vow of the govrata. We are also told in the Jaiminīyabrāhmaṇa that the king of the Śibis performed the govrata when the king of Videha had refused to do so. Thus, the Śibis in their capacity as the performers of the govrata can be called forerunners of the Pāśupatas. One might therefore venture to identify the deity worshipped by the cattle-raising Śibis as Rudra, who is recognised as Paśupati, the Lord of Cattle, already in Vedic passages like TS 1.8.10, MS 2.6.6, and ŚB 5.3.3.
On the other hand, it is also reported by Megasthenes that the Macedonians had seen the same deity being worshipped by the Śūrasenas in Mathurā. The Śūrasenas, the city of Mathurā, and the deity with a club in his hand—all these points are present in Kṛṣṇa stories, and this has led some scholars to identify this deity deemed as the Indian Heracles by the Macedonians as Kṛṣṇa. On logical grounds, this deity with the same attribute and characterisation, worshipped by the Śibis as well as the Śūrasenas, cannot be identified differently. So, he can be either Kṛṣṇa or Rudra.
But there are two more options available: Asko Parpola has recently renewed the proposal of James Tod (1873) that this deity should be identified as Baladeva, an important deity in early Viṣṇuism who was in later times made the elder brother of Kṛṣṇa.53 Similarly, Dahlquist (1962) has argued extensively against the earlier proposals of identification of this deity as Kṛṣṇa and Śiva, and has proposed that Megesthenes’ Indian Heracles may have been Indra on the ground that the two figures have a lot in common.54 Dahlquist was pointing to the right direction, but it appears that nobody listened to him because of the lack of compelling evidence, and because of his many other erroneous arguments.
I find that Indra is the strongest among these four candidates. The resemblance between Indra and Heracles is well-known, at least amongst scholars of Comparative Mythology. The classicist Wilamowitz (1895: Vorwort, VIIIff) was the first to point out this resemblance, followed by Leopard von Schröder (1914); but it was Franz Rolf Schröder (1957) who compared these warrior figures together with Thor and showed convincingly that the three have many attributes in common. This was a welcome discovery for Dumézil (1969), who wanted to see ‘the warrior’ behind all individual warrior figures. It is not difficult to imagine that the Macedonians, who were accustomed to the figure of Heracles, might have confused the figure of Indra with that of Heracles because of their mutual resemblance.
Beyond this resemblance, there are some other indicators which point in the same direction. Kṛṣṇa-mythology preserves stories of Kṛṣṇa’s victory over Indra and also of the latter’s rivalry and conflict with Kṛṣṇa and finally the acknowledgement of Kṛṣṇa’s divine status.55 All this tells that the Kṛṣṇa cult is tactically positioning itself as superior to the Indra cult, and it indicates that the Śūrasenas in the Mathurā area were probably worshipping Indra when that place was visited by the Makedonians.
In the Indian epics and Purāṇas, Indra either comes to glorify the gods of new cults or he is in conflict with them and is subjugated. In Buddhist myths also, the Buddha is often depicted as being glorified by Brahman and Indra, and in early sculptures, he is shown flanked by these two. It seems to me that this is a strategy: Brahman Prajāpati is the central figure of the mainstream Vedic world focusing on the fire sacrifice, and Indra appears to be the central figure of the marginal Vedic world focusing on a kind of devotionalism based on vows and conducts. As they both are venerating the Buddha, the Buddha transcends them both.
Now we return back to the Śibis, and to our original question whom they were worshipping. Since they appear to be observing the govrata, there is greater likelihood that they are worshipping Indra. For, the Jaiminīyabrāhmaṇa passage we read earlier provides indirect evidence for the hypothesis that Indra is the deity of these govratins in that it associates the govrata with the gosava, a sava type of Vedic sacrifice in which Soma is pressed and Indra is invited.56 In fact, there is a passage in the Pāśupatasūtra itself which speaks openly of Indra’s association with the Pāśupata vow. Pursuing the clue provided in this passage, we can tentatively probe deep down to what might have been the foundational layer of this vow. So, in the next section I read this passage of three sūtras and pursue its clues.
Indra Observes the Vow!
Pāśupatasūtra 4.10–12: indro vā agre asureṣu pāśupatam acarat, sa teṣām iṣṭāpūrtam ādatta, māyayā sukṛtayā samavindata.
‘Indra actually (vai)57 observed the Pāśupata [vow] amongst Asuras in the very beginning; he took their merits of the sacrificial acts and of the sacrificial charities,58 and attained [his goal] with well-performed magical skill (māyā).’
This passage presents Indra as the primaeval observer of the Pāśupata vow and claims that he observed it amongst the Asuras. We cannot take this statement literally but must take it seriously and investigate it critically.59 Since we have already seen that the Pāśupata vow is based on the Vedic govrata, we can guess that what Indra is said to have performed might be the original govrata or some other Vedic vow, similar in nature and more archaic, which can serve as the archetype of the Pāśupata vow.
We do indeed find such a thing in an Atharvavedic hymn dedicated to the celestial draft-ox (anaḍutsūkta) found in both Śaunaka (AVŚ 4.11) and Paippalāda (AVP 3.25) Saṃhitās. This hymn speaks about the vow of the draft-ox and relates that Indra assumed the form of a draft-ox and observed ‘the vow’ for the gods. Let me present here, following the Śaunakasaṃhitā,60 the second, third and seventh stanzas of that hymn:
anaḍvā́n índraḥ sá paśúbhyo ví caṣṭetrayā́ṃ chakró ví mimīte ádhvanaḥ |bhūtáṃ bhaviṣyád bhúvanā dúhānaḥsárvā devā́nām carati vratā́ni ‖índro jātó manuṣyèṣv antárgharmás taptáś carati śóśucānaḥ |suprajā́ḥ sánt sá u dāré ná sarṣadyó nā́śnīyā́d anaḍúho vijānán ‖índro rūpéṇāgnír váhenaprajā́patiḥ parameṣṭhī́ virā́ṭ |viśvā́nare akramata vaiśvānaré akramatānaḍúhy akramata |só ’dṛṃhayata só ’dhārayata ‖
61 The draft-ox [is] Indra; he appears [distinguished] from among cattle (sá paśúbhyo ví caṣṭe); the mighty one (śakrá) measures out the three ways; milking (duh) the past, future, and existing things, he observes all the vows (vratá) for the gods.
[He is] Indra [himself] born among human beings. The heated hot-drink (gharmá) [he is, and] moves about (car) shining brightly. Being one with good offspring he shall not, on the other hand, run into the cleft [on the path] (dārá),62 who shall not partake of (aś) the draft-ox63 with the understanding [of this reality].
Indra by form, Agni by carrying (váha), Prajāpati, Parameṣṭhin, Virāj. In Viśvānara he strode, in Vāiśvānara he strode, in the draft-ox he strode; he made [himself] firm, he sustained.
The last of the above stanzas appears to allude to some myth which is unclear here. However, we can find this myth told in some detail in the sixth anuvāka of the 17th Book of the Paippalādasaṃhitā (AVP 17.27–29). These three sections—it would not be right to call them hymns—are concerned with the vow of the draft-ox (anaḍudvrata). Here the story of Indra’s slaying of Vṛtra is told, the celestial draft-ox is identified as Indra, and the observer of the vow of the draft-ox who would know these details and observe the vow is lauded. In this passage we can even find some allusions to the anaḍutsūkta.64 These two Atharvaveda passages establish the celestial draft-ox as one of Indra’s forms. They also attest to the fact that Indra once learnt ‘the great vow’ from the celestial draft-ox, observed it, and lived among wild and domesticated animals.
Now let us go back to the Pāśupatasūtra statement that Indra observed the Pāśupata vow while he was living among the Asuras. I think that the Pāśupatasūtra is alluding to the myth narrated in AVP 17.27–29. Even though there we are not told anything about the nature of the vow Indra observed before foresaking the Asuras, we can see that the vow Indra observed after slaying Vṛtra, fulfilling the expectation of the gods, is comparable to the original Pāśupata vow of imitation of the bull’s behaviour.
We can now say with confidence that this is the episode to which the Pāśupatasūtras 4.10–12 are alluding. When the Sūtrakāra makes the claim that “Indra actually performed the Pāśupata vow,” he is clearly identifying the anaḍudvrata/ govrata as the Pāśupata vow. Once such an identification is made, it is natural for a Pāśupata to take Indra in their fold and conceive him as the first performer of the Pāśupata vow. This claim is very shrewd in the sense that it downgrades the original deity of the old cult to a mere follower of the new cult, standing down in front of the new lord. In this way, Rudra takes over Indra and poses himself firmly as the Lord of Cattle. There are some indications that in a transitional period Rudra was still under the shadow of Indra. Inclusion of Indra’s epithets, for example balapramathana, in the second brahmamantra of the Pāśupatas suggests this.65
There is one more clue which proclaims this transition. Rudra as the Lord of the historical Pāśupatas is not depicted with the weapons commonly associated with Rudra: a bow and arrows, the Pināka or a Pāśupatāstra. Even when Pāśupatism has been modified and reintroduced, it has branched into various sub-systems, and the Lord is given new names in these sub-systems, he is always depicted with a staff or club in his hand: as Caṇḍa in an earlier stage,66 as the Lord-incarnate Lakulīśa in the form of the divine teacher, or even as Musalīśa, the lordly divine teacher of Somasiddhāntins.67 In my analysis, this is so because the Pāśupata cult emerges from the remnants of a cult of Indra, and the figure of the Lord they have adopted is calqued upon Indra. The club in the hand of these figures is derived from Indra’s vajra, which should originally be a club-like weapon with tiny spikes all around but since early times some have apparently understood it as a club, for in Avestan vazra means a club (cf. Mayrhofer 1996: s.v. vájra). He is still Indra in appearance, though he is reintroduced as Rudra, and this is emphasised by the descriptive names Lakulīśa and Musalīśa.
On the other hand, our rediscovery of Indra as the celestial bull, and his association with the vow of the bull in all three levels of anaḍudvrata, govrata, and Pāśupata vrata provide the necessary evidence for the existence of Indra’s popularity from Middle Vedic times to the time of the composition of the Pāśupatasūtras. So without hesitation, we can now say that the deity seen by the Makedonians in different parts of India and deemed as Indian Heracles was very likely to have been Indra.
The findings presented in this article lead us beyond the boundaries of the Pāśupatism of the Pāśupatasūtra into the realm of the Atharvavedic anaḍudvrata. The distinctive feature of this vow is that the deity venerated by its observers is the celestial draft-ox, who is identified with Indra and Prajāpati and is recognised as the one standing highest (parameṣṭhin), who is related to the All-Gods. Likewise, the ultimate reward of this observance is the world of the celestial draft-ox itself, beyond the world of the sun, in other words neither the world of Indra nor that of Rudra.
Putting the information from the Jaiminīyabrāhmaṇa account of the govrata and the Greek accounts of the Śibis, we can guess that this vow of adoption of the bull’s behaviour was a well recognised if not actually commonplace phenomenon among certain peoples in the margins of Vedic society. The bull was a sort of totem of pastral tribes, and by mimesis these people worshipped their god in the form of the archetypal bull.
It appears that for these people it was a vow for life, a substitute of orthodox Vedic ritualism, but it was at some point accommodated in the scheme of mainstream Vedism by equating the divine bull with Indra/Prajāpati. The anaḍutsūkta from the Atharvaveda itself (AVŚ 4.11.11)68 provides evidence for an early ritualisation of this vow, as it talks of the twelve nights of the observance. The anaḍudvrata of the Atharvaveda has obviously found its place in the brahmanical ritual world as govrata, as it is evidenced in the Jaiminīyabrāhmaṇa 2.113. Now this is associated with the one-day Soma-pressing Vedic sacrifice of Gosava,69 and the duration of this observance, according to the Śrautasūtras, is either one year or just twelve days. As the Jaiminīyabrāhmaṇa describes, the person who observes this vow imitates a draft-ox, and even sleeps, at least symbolically, with his own mother, sister, or a lady of his own Gotra; and in this way breaking the boundaries of purity and impurity, vice and virtue and so on, he attains the world of the celestial draft-ox.
The reward of the vow is still the world of the celestial draft-ox (anaḍuho lokaṃ) in the Jaiminīyabrāhmaṇa, but in the Śrautasūtras it has been altered. Now it is svārājya: the king’s sovereignty or the kingdom of heaven, and it is a king’s choice to perform this sacrifice and observe the vow. Various Śrautasūtras bear witness to this development,70 and some of them attest to the fact that this vow was eventually renamed as paśuvrata.
Then in the next stage, about the time of composition of the Pāśupatasūtra, Rudra was identified as the lord of cattle,71 the vow of paśuvrata was again renamed as the pāśupata-vrata, and Indra, the deity of the govrata/paśuvrata, was downgraded to the rank of the first observer of this vow. In this new model, too, this vow was originallly a life-long vow, as it was in the case of the Śibis, but possibly some arcane elements of the vow were practiced only in old age, a tendency already suggested in the Jaiminīyabrāhmaṇa. Over the ages, as the notion of Pāśupatism became gradually transformed, the Pāśupatasūtra was rewritten, and the original form of the vow got covered up under the new elements. The practice of the adoption of the bull’s behaviour remained there, but it became more and more restricted.
In all these stages, however, the conception of the celestial bull as god and the way of his worship remained the same. Though he was identified in various stages differently as Indra, Prajāpati and Rudra, his worshippers venerated him by imitating his behaviour. The origin of the Pāśupata praxis lies in an ancient imitatio dei practice in which the worshiper pleased his god, the archetypal, by mimesis. Thus, the findings of this paper lend support to Hans Bakker’s recently published thesis that, like in many other eastern as well as western traditions, imitation of god is at the heart of the Pāśupata praxis.72
The roots of this kind of practice of adoption of a domesticated or wild animal’s behaviour, if we may be permitted to generalise in this way, can be found reaching beyond ancient Indian culture. Ingalls long ago pointed out similarities between Pāśupata practices and those of the Greek cynics who adopted the behaviour of dogs.73 Now with our knowledge of the original version of the Pāśupata vow and its background in the govrata-based cult of Indra, we may observe that the proto-Pāśupatas look even more similar to the Greek cynics. In fact, some Buddhist texts are aware of both of these practices: imitation of the bull’s behaviour and also that of dogs behaviour, and they put them together in one rank. The Kukkuravatiyasutta from the Majjhimanikāya (II.1.7) presents a govatika together with a kukkuravatika. They are observing their vows, and have adopted the behaviour of a bull and a dog respectively. The Buddha tells them that as they are cultivating bullness and dogness, the state of mind of these animals, they will go to hell or become reborn as animal. They are alarmed at this and take refuge in the Buddha. In accord with this narrative, the Jñānaprasthāna excerpt cited in Vasubandhu’s passage I quoted on p. 9 talks about both of these vows, and adds the vow of imitation of a wild animal’s behaviour in the middle of these two. The Pāśupatasūtra does not mention the imitation of dogs, though it mentions the imitation of a wild animal’s behaviour along with that of a bull. The Lalitavistara mentions other animals in relation with this kind of vow:74 ‘by means of the vow of taking on the behaviour of a bull, any wild animal, a dog, boar, or elephant, … stupid people understand that purity is achieved.’ The Yogācārabhūmi does the same.75 However, we have to be careful here, because there is no trace of the vow of imitation of the bull’s behaviour in ancient Greek sources or of that of the dog’s behaviour in Vedic texts. As Ingalls suggested, for the time being, it is “best to regard the two cults as parallel.”76
Abhidharmakośa & Bhāṣya of Vasubandhu with Sphuṭārthā Commentary of Yaśomitra. ed. Dwārikādās Śāstri. Bauddha Bharati Series 5–9. Benares: Bauddha Bharati, 1987.
Acharya, Diwakar. 2005. “The Role of Caṇḍa in the Early History of the Pāśupata Cult and the Image on the Mathurā Pillar Dated Gupta Era 61” in Indo-Iranian Journal 48: 207–222.
Acharya, Diwakar. 2011. “Pāśupatas” in Brill’s Encyclopedia of Hinduism, Vol. III: 458–466. Leiden: Brill.
Atharvavedasaṃhitā in Paippalāda recension (AVP). Electronic Text redacted by A. Griffiths, in joint collaboration with A.M. Lubotsky, M. Witzel, and T. Zehnder.
Atharvavedasaṃhitā in Śaunaka recension (AVŚ). Second Revised Edition. ed. R. Roth & W. Whitney. Berlin: Ferd. Dümmlers Verlagsbuchhandlung, 1924. For a Translation, see Whitney 1905.
Bakker, Hans. 2000. “Somaśarman, Somavaṃśa and Somasiddhānta. A Pāśupata tradition in seventh–century Dakṣiṇa–Kośala: Studies in the Skandapurāṇa III” in Ryutaro Tsuchida and Albrecht Wezler (eds.): Harānandalaharī. Volume in Honour of Professor Minoru Hara on his Seventieth Brithday: 1–19. Reinbek.
Bakker, Hans. 2010. “The gospel of Kauṇḍinya: The descent of God in Gujarat and the practice of imitating God” in Jitse Dijkstra, Justin Kroesen & Yme kuiper (eds.): Myths, Martyrs, and Modernity. Studies in the History of Religions in Honour of Jan N. Bremmer: 517–529. Numen Book Series 127. Leiden: Brill.
Bakker, Hans. 2011. “Origin and Spread of the Pāśupata Movement: About Heracles, Lakulīśa and symbols of masculinity” in Bertil Tikknen & Albion M. Butters (eds.): Pūrvāparaprajñābhinandanam. Volume in Honour of Klaus Kartunen: 21–37. Studia Orientalia 110. Helsinki: Finnish Oriental Society.
Bhattacharya, Vidhushekhara. The Yogācāra-bhūmi of Ācārya Asaṅga. Part I. Calcutta: University of Calcutta, 1957.
Bisschop, Peter. 2006. Early Śaivism and the Skandapurāṇa: Sects and Centres. Groningen Oriental Series, Vol. XXI. Groningen: Egbert Forsten.
Bisschop, Peter. 2006a. “The sūtrapāṭha of the Pāśupatasūtra” in Indo-Iranian Journal 49: 1–21.
Bisschop, Peter. 2010. “Once Again on the Identity of Caṇḍeśvara in Early Śaivism: A rare Caṇḍeśvara in the British Museum?” in Indo-Iranian Journal 53: 233–249.
Bṛhaddevatā. The Bṛhaddevatā: Text Reconstructed from the Manuscripts of the Shorter Recension with Introduction, Explanatory Notes, and Indices. Muneo Tukunaga. Kyoto: Rinsen Book Co., 1997.
Brahmāṇḍapurāṇa. ed. J.L. Shastri. Delhi: Motilal Banarsidass, 1973.
Caland, W. 1919. Das Jaiminīya-Brāhmaṇa in Auswahl: Text, Übersetzung, Indices. Amsterdam: Koninklijke Akademie van Wetenschappen.
Dahlquist, Allan. 1962. Megasthenes and Indian Religion: A Study in Motives and Types. Uppsala: Almquist & Wiksells.
Dumézil, Georges. 1969. The Destiny of the Warrior. Translated by Alf Hiltebeitel from the French original: Heur et Malheur du guerrier. Aspects de la fonction guerrière chez les Indo-Européens, Presses Universitaires de France, 1969. The University of Chicago Press, 1970.
Ehlers, Gerhard. 1988. Emendationen zum Jaiminīya-Brāhmaṇa (Zweites Buch). Indica et Tibetica, 14. Bonn: Indica et Tibetica Verlag.
Gaṇakārikā of Ācārya Bhāsarvajña with the Ratnaṭīkā and four appendices. ed. Chimanlal D. Dalal. Gaekwad Oriental Series, 15. Baroda 1920. Reprint 1966.
Geldner, Karl Friedrich. 1951. Der Rig-veda aus dem Sanskrit ins Deutsche übersetzt und mit einem laufenden Kommentar versehen. 3 Volumes. Harvard oriental Series, 33–35. Cambridge, Massachusetts: Harvard University.
Goodall & Isaacson 2007. “Workshop on the Niśvāsatattvasaṃhitā: The Earliest Surviving Śaiva Tantra?” in NGMCP Newsletter 3: 4–6.
Hara, Minoru. *1966. Materials for the Study of Pāśupata Śaivism. Unpublished Dissertation. Harvard University.
Hara, Minoru. 2002. Pāśupata Studies. ed. Jun Takashima. Publications of the de Nobili Research Library Volume, 30. Vienna: Sammlung de Nobili.
Hara, Minoru. 2003. “Ashes” in Cracow Indological Studies 4–5: 251–267.
Ingalls, Daniel H.H. 1962. “Cynics and Pāśupatas: the Seeking of Dishonor” in Harvard Theological Review 55.4: 281–298.
Iyanaga, Nobumi. 1983. “Daijizaiten” in Hôbôgirin. Sixth Fascicule, Paris: Librrarie d’Amerique et d’Orient/ Tokyo: Maison Franco-Japonaise, pp. 713–-765.
Jaiminīyabrāhmaṇa of Sāmaveda. ed. Raghu Vira and Lokesh Chandra. Second Revised Edition. Delhi: Motilal Banarsidass, 1986.
Kauśikasūtra of Atharvaveda, with extracts from the commentaries of Dārila and Keśava. ed. Maurice Bloomfield. Indian Reprint Delhi: Motilal Banarsidass, 1972.
Lalitavistara. Ed. P.L. Vaidya. Buddhist Sanskrit Texts, 1. First edition. Darbhanga: The Mithila Institute, 1958.
Lindtner, Christian. 1997. “Bhavya on Mīmāṃsā” in Aspects of buddhism: Proceedings of the international Seminar on Buddhist Studies: 91–123. Warsaw.
Madhyamakahṛdaya of Bhavya. ed. Chr. Lindtner. Adyar Library Series, 123. Chennai: The Adyar Library and Research Centre, 2001. See also Lindtner 1997.
Mahābhārata. ed. V. Sukthankar, with the cooperation of S.K. Belvalkar, A.B. Gajendragadkar, V. Kane, R.D. Karmarkar, P.L. Vaidya, S. Winternitz, R. Zimmerman, and other scholars and illustrated by Shrimant Balasaheb Pant Pratinidhi. (Since 1943 ed. S. Belvalkar). 19 Vols. Poona: Bhandarkar Oriental Research Institute, 1927–1959.
Maitrāyaṇīsaṃhitā. Die Saṃhitā der Maitrāyaṇīya-Śākhā. ed. Leopold von Schroeder. Viertes Buch. Wiesbaden: Franz Steiner, 1972.
Majjhima-nikāya. Vol. I. ed. V. Trenckner. Pali Text Society Text Series no. 60. London: The Pali Text Society. Reprint, 1979.
Manusmṛti. Manu’s Code of Law: A Critical Edition and Translation of the Mānavadharmaśāstra. Patrick Olivelle. New York: Oxford University Press, 2005.
Mayrhofer, Manfred. 1996. Etymologisches Wörterbuch des Altindoarischen. Volume II. Heidelberg: Carl Winter Universitätsverlag.
Mylius, Klaus. 1976. “Der Gosava” in Archív orientální, Vol. 44:1: 43–53. Praha: Czechoslovak Academy of Sciences.
Nāṭyaśāstra: Text with Introduction, English Translation and Indices in Four Volumes. N.P. Unni. Second Edition. Delhi: Nag Publications, 2003.
Niśvāsatattvasaṃhitā. NAK MS 1–227, NGMPP Reel No. A 41/14. Palm-leaf, Folios 117, early Nepalese (Kuṭilā) script. There are two apographs available, both in Devanagari and on paper: NAK MS 5-2401, NGMPP Reel No. A 159/18, and Welcome Institute for the History of medicine, London, Sanskrit MS I.33. I have used Goodall’s edition in progress.
Oberlies, Thomas. 2000. “Kriegslisten und ungeziemendes Benehmen: Die askesepraktiken der Pāśupatas” in Harānandalaharī: Volume in Honour of Professor Minoru Hara on his Seventieth Birthday: 175–191. Reinbek Dr. Inge Wezler.
Parpola, Asko. 2002. “Pandaíē and Sītā: On the Historical Background of the Sanskrit Epics” in Journal of the American Oriental Society 122.2.
Pasadika, bhikkhu. 1989. Kanonische Zitate im Abhidharmakośabhāṣya des Vasubandhu. Göttingen: Vandenhoeck & Ruprecht.
Pāśupatasūtra. With Kauṇḍinya’s Pañcārthabhāṣya Commentary. ed. R. Ananthakrishna Sastri. Trivandrum Sanskrit Series, 143. Trivandrum: University of Travancore, 1940.
Ramesh, K.V. & Tiwari, S.P. 1990. A Copper-plate Hoard of the Gupta period from Bagh, Madhya Pradesh. New Delhi: Archaeological Survey of India.
Rau, Thomas. 1983. “Sundry Remarks on the Printed Text of Jaiminīya-Brāhmaṇa” in Surabhi: Sreekrishna Sarma Felicitation Volume: 141–144. Tirupati: Felicitation Committee.
Sakamoto-Gotō, Junko. 2000. “Das Jenseits und iṣṭā-pūrtá- “die Wirkung des Geopferten-und-Geschenkten” in der vedischen Religion” in Indoarisch, Iranisch und die indogermanistik: Arbeitstagung der Indogermanischen Gesellschaft vom 2. bis 5. Oktober 1997 in Erlangeen: 475–490. Wiesbaden: Reichert Verlag.
Sanderson, Alexis. 2006. “The Lakulas: New evidence of a system intermediate between Pāñcārthika Pāśupatism and Āgamic Śaivism.” Ramalinga Reddy Memorial Lectures, 1997. In: The Indian Philosophical Annual 24 (2006): 143–217.
Śatapathabrāhmaṇa. The Çatapatha-Brāhmaṇa in the Mādhyandina-Çākhā with extracts from the commentaries of Sāyaṇa, Harisvāmin and Dvivedāṅga. ed. Albrecht Weber. Berlin 1849.
Schmithausen, Lambert. 1992. “A Note on Vasubandhu and the Laṅkāvatārasūtra” in Asiatische Studien XLVI.1: 392–397. Bern: Peter Lang.
Schröder, Franz Rolf. 1957. “Indra, Thor und Herakles” in Zeitschrift für Deutsche Philologie 76 (1957): 1–41. Berlin: Erich Schmidt Verlag.
von Schröder, Leopard. 1914. Herakles und Indra. Eine mythenvergleichende Untersuchung. Denkschriften der Kaiserlichen Akademie der Wissenschaften in Wien. Philosophisch-historische klasse. 58, 3–4.
Schwanbeck, & McCrindle, J.W. 1877. Ancient India as described by megagasthenês and Arrian.
Taittirīyabrāhmaṇa with the commentary of Bhaṭṭa Bhāskara Miśra. Vol. II: Aṣṭaka II. ed. R. Shama Sastry. Mysore: University of Mysore, 1921. Reprint: Delhi: Motilal Banarsidass, 1985.
Taittirīyabrāhmaṇa with the commentary of Sāyaṇa. ed. Nārāyaṇa Śāstrī Godbole. Vols. I–III. Ānandāśrama Sanskrit Series 37. Ānandāśrama Sanskrit Press, 1979.
Tod, James. 1873. “Comparison of the hindu and Theban Hercules. Illustrated by an Ancient Hindu Intaglio” in Transactions of the Royal Asiatic Society of Great Britain and Ireland 3: 139–159.
Whitney, William D. 1905. Atharva-veda Samhita: Translated with a Critical and Exegetical Commentary. Revised and brought nearer to completion and edited by Charles Rockwell Lanman. in 2 Vols. with continuous pagination. Harvard Oriental Series 7–8. Cambridge, Mass.: Harvard University Press.
von Wilamowitz-Moellendorff, Ulrich. 1895. Euripides Heralkes. Second edition. Vol. 1. Berlin: Weidmannsche Buchhandlung.
Yaśomitra. See Abhidharmakośa.
A short version of this paper was presented in the Tantra and Āgama section of the 14th World Sanskrit Conference held at Kyoto University in September 2009. Since then it has gone through several changes and expansion. On the other hand, I have kept my analysis of AVP 17.27–29 aside for a future article.
I have benefitted from critical comments of Dominic Goodall and Arlo Griffiths on earlier drafts of this paper. I have also received helpful suggestions from Hans Bakker, Alexis Sanderson, Albrecht Wezler, and Yuko Yokochi. I express my sincere gratitude to all these scholars.
In seven copper-plate inscriptions of King Bhuluṇḍa dated between Year 50–59 (369–378 ce, supposing that the year used in these plates is the Gupta era), some places of worship of the Pāśupatas and their teachers are mentioned (see Ramesh and Tewari 1990: nos. 3, 5, 6, 9, 10, 12, 13). In the Mathurā pillar inscription of Gupta year 61 (380ce), some Māheśvara teachers whose names end in vimala are mentioned. Though they are not called Pāśupata, their lineage stems from Bhagavant Kuśika who is undoubtedly linked with the origin of Pāśupatism. For a brief account of Pāśupata history, see Acharya 2011. For various issues related to the early beginnings of Pāśupatism, see Bakker 2010 and 2011, Bisschop 2006 and 2010, Sanderson 2006, Acharya 2005, Ingalls 1962.
See Sanderson 2006, Goodall & Isaacson 2007. Similarly, the Pāśupatavidhi of the Atharvapariśiṣṭas paraphrases a large number of sūtras from the Pāśupatasūtra; See Bisschop & Griffiths 2003. But, like any of chapter of the Atharvapariśiṣṭas, it is very difficult to tell which time period this chapter belongs to.
First reported in an unpublished 1998 lecture handout of Prof. Alexis Sanderson, as cited in Bisschop 2006a: 15. In the ninth chapter of this text (vv. 9.62, 114–115), Pāśupatasūtras 5.11–12, 17–19, 24 and 26 are paraphrased.
The Nārāyaṇīya section of the Mahābhārata (12.337.59) lists Pāśupata knowledge along with Sāṃkhya, Yoga, and Pañcarātra, and states that this Pāśupata knowledge was taught by Śiva, Śrīkaṇṭha, the son of Brahman, husband of Umā, the lord of living beings (Mahābhārata 12.337.62). Except this chapter, the term pāśupata appears 14 times in the whole of the critical edition of Mahābhārata, but always as the name of Śiva’s powerful arrow. However, later in this article I will argue that the Mahābhārata knows of the forerunners of the Pāśupatas or proto-Pāśupatas not yet known by this name; see below, pp. 14–15.
If I extend my search to the Appendices to the main text of the critical edition of Mahābhārata, the same term appears 15 times more as the name of Śiva’s arrow, but at five places it signifies some special knowledge or Yoga. Not in the first three but the other two places some extra information is found. The 28th supplement to the 12th Book (ll. 401–409) puts in the mouth of Śiva that the atyāśramavrata and Pāśupata Yoga he is to impart to Dakṣa was created at an ancient time. Similarly, in the 15th supplement to the 13th Book (ll. 4344–4348), it is put in the mouth of Śiva that at the time of creation of the Pāśupata Yoga he had collected four Brahmins from four stages of life and explained it to them. They had received it from his South face and spread in the world. The same passage (ll. 4351–4352) mentions that all those devotees of Śiva smearing their body with ashes should be regarded as Pāśupatas, and also that the ascetics bearing scull-cups in their hands wander in the world reflecting on the scripture of Pāśupata Yoga. These two passages contain interesting information but there is higher probability that they are comparatively late. Therefore, I have limited myself to the main text of the critical edition of Mahābhārata.
It is true that the date of the Niśvāsatattvasaṃhitā is not yet well-fixed, but it cannot be later than the seventh century. See Goodall & Isaacson 2007.
This text can be dated about the end of the fourth century or a few decades later, see Schmithausen 1992.
muninirgranthaśākyeṣu yatipāśupateṣu ca ‖ vratānugas tu kartavyo veṣo lokasvabhāvajaḥ | cīravalkalacarmāṇi tāpasānāṃ tu yojayet ‖ parivrāṇmuniśākyānāṃ vāsaḥ kāṣāyam iṣyate | nānācitrāṇi vāsāṃsi kuryāt pāśupateṣv atha ‖ (Nāṭyaśāstra 22.125–127ab)
Nāṭyaśāstra 7.83ff: unmādo nāma iṣṭajanaviyogavibhavanāśābhighātavātapittaśleṣmaprakopādibhir vibhāvair utpadyate. tam animittahasitaruditotkruṣṭāsambaddhapralāpaśayitopaviṣṭotthitapradhāvitanṛttagītapaṭhitabhasmapāṃsvavadhūlanatṛṇanirmālyakuchelacīragha-ṭakapālaśarāvābharaṇadhāraṇopabhogair anekaiś cānavasthitaiś ceṣṭānukaraṇādibhir anubhāvair abhinayet.
After reading this passage in a straightforward way, Hara (2003: 265) has understood that “the application of ashes to one’s body is one of the types of behaviour peculiar to a madman.” According to him, “the Pāśupata ascetics who besmear their body with ashes intentionally imitate the behaviour of madmen (unmatta) in their pursuit of ātmaśauca.” However, I think that the argument should be the other way round: the Nāṭyaśāstra knows that the Pāśupata ascetics imitate the behaviour of a madman, and therefore, when it needs to provide instructions for acting out madness, it describes a Pāśupata ascetic. All other points are not compelling: they could be part of a description of any madman, but it becomes compelling when the use of the garlands taken off after offering to a deity (nirmālya) is mentioned. An ordinary madman does not seek this specific substance.
-masi- (meaning ink) is very likely to be a misreading for -rāśi-. In that case, it will be combined with the preceding word, and the combination of bhasmarāśi- will mean ‘the heap of ashes.’
Lalitavistara Parivarta 17, (Vaidya p. 183): … bhasmamasinirmālyoddhṛtatamorajapāṃśupaṅkaparimrakṣaṇaiś ca lomamuñjakeśanakhacīvarapañjarakaraṅkadhāraṇaiś ca … aṅgāradhātukaṣāyatridaṇḍamuṇḍikakuṇḍikakapālakhaṭvāṅgadhāraṇaiś ca śuddhiṃ pratyavagacchanti saṃmūḍhāḥ.
Armed with this much information from these texts, we may feel tempted to identify ‘the twice-born in the habit of lying in ashes’ featuring in the Buddhacarita (7.51) as some kind of Pāśupata ascetic. As Johnston remarks, ‘Bhasmaśāyin shows that he was a Śaiva ascetic’ (1936, II (translation): 101, fn. 51). Hara (1966: 19–20 and 2003: 251, fn. 1) also refers to this passage but cautiously as ‘the earliest reference to the ash-besmeared ascetic in Sanskrit literature.’ However, I find that this ascetic is recommending prince Siddhārtha to resort to Arāḍa, who is depicted as a Sāṃkhya teacher. This internal evidence, if not stemming from Aśvaghoṣa’s poetic freedom, does not allow us to identify this ascetic as a sectarian Pāśupata or even as a Śaiva.
We can gather the additional piece of information from the first passage, namely that the Pāśupatas as well as other ascetics appeared differently on different occasions, depending on the vows they observed at particular occasions.
śīlaṃ dauḥśīlyaviratiḥ. vrataṃ kukkuragovratādīni. yathā ca nirgranthādīnāṃ nagno bhavaty acelaka iti. vistaraḥ—brāhmaṇapāśupataparivrājakādīnāṃ ca daṇḍājina-bhasma-jaṭā-tridaṇḍamauṇḍyasamādānam. (Abhidharmakośabhāṣya 3.28ab)
Yaśomitra connects the above mentioned religious groups with the respective observances and also mentions that, because the term ādi is included in the compound, one should incorporate here other silimar religious groups like the Kāpālikas. If so, there will be some problem in respective distribution of the rules to the religious groups. However, it is important to note that Yaśomitra knows the Kāpālikas as a distinct group.
neśvarādeḥ kramādibhiḥ ‖
yadi hy ekam eva kāraṇam īśvaraḥ syād anyad vā, yugapat sarveṇa jagatā bhavitavyaṃ syāt. dṛśyate ca bhāvānāṃ kramasaṃbhavaḥ. […] yadi ceśvaro narakādiṣu prajāṃ bahubhiś cetibhir upasṛṣṭāṃ sṛṣṭvā tena prīyate namo ’stu tasmai tādṛśāyeśvarāya. sugītaś cāyaṃ tam ārabhya śloko bhavati—yan nirdahati yat tīkṣṇo yad ugro yat pratāpavān | māṃsaśoṇitamajjādo yat tato rudra ucyate ‖ iti. ekaṃ khalv api jagataḥ kāraṇaṃ parigṛhṇatā ’nyeṣām arthānāṃ pratyakṣaḥ puruṣakāro nihnutaḥ syāt. sahāpi ca kāraṇaiḥ kārakam īśvaraṃ kalpayatā kevalo bhaktivādaḥ syāt, kāraṇebhyo ’nyasya tadutpattau vyāpārādarśanāt.…
Prof. Sanderson has kindly informed me that the Chinese commentator Puguang (Sino-Jap. Fukō) comments on this verse in his commentary ad Abhidharmakośabhāṣya composed circa 650–655ce, which can be found as No. 1821 in the Taisho edition of the Chinese Tripiṭaka. There Puguang tries to elaborate the ideas and myths he understands are lying behind this verse. Prof. Sanderson has been kind to share his English translation of the related portion of Puguang’s commentary which he “accessed through a French translation given by Iyanaga Nobumi in his excellent article ‘Daijizaiten’ (=Maheśvara) in Hôbôogirin (pp. 713–765).” He has also added the Sanskrit, in parentheses behind the glosses and what he conjectures to be the Sanskrit of the technical terms in square brackets. I cite it from his email received on the 28th of January, 2010. It covers from line 22 of column A to line 15 of column B on page 728 of Iyanaga’s article:
The heretics who cover themselves with ash (zuke-gedo, i.e. the Pāśupatas …) teach that the god Īśvara transcends the three planes and has three bodies: 1. A Body of Essence [dharmakāyaḥ] which fills the Essence Plane [dharmadhātuḥ]; 2. A Body of Fruition [sambhogakāyaḥ] which resides at the summit of the Material Plane [rūpadhātuḥ], in the palace of the god Īśvara: this is the god Maheśvara who is treated in Buddhist texts. He has three eyes and eight arms; his body is 1600 leagues high; 3. A Metamorphosis Body [nirmāṇakāyaḥ], which changes its appearance in accordance with the six Destinies [gatiḥ], in which he converts [beings] in various ways. The verse (cited in the Bhāṣya) describes this Metamorphosis Body. “ Harsh” (tīkṣṇaḥ) and  “incendiary” (nirdahati)—There were [once] three Asuras, masters of the three worlds (sangokudo, Tripura). Flying through the air they passed over the god Īśvara. He, not tolerating [this offence], shot a flaming arrow at the three worlds, which burned them at a stroke. This arrow, burning and  “harsh” has [therefore]  “burned” the three worlds.  “Fearsome (ugraḥ),  he oppresses always” (pratāpavān)—On a dragon [nāgaḥ] he threads human skulls (dokuro, kapāla) and with this adorns the top of his head (or, according to the variant of edition B [otsu]): his head and his neck). Furthermore, he wraps snakes around his arms, and kills an elephant to obtain his blood-drenched hide, which he wears inside out.  “He loves [to drink] blood, and to eat flesh and marrow” (māṃsaśoṇitamajjādaḥ)—This explains how he feeds himself.  [“This is why he is called Rudra”]—Sacrificers of today [without considering him for these reasons to be cruel] on the contrary offer him sacrifices; this is why they call him Rudra.
The most interesting bit in this piece is the statement about the three bodies of Īśvara phrased in Buddhist terms. Since he belongs to the seventh century it is quite probable that the commentator knows the Mantramārgic concept of the three bodies of Śiva: Niṣkala, Sakalaniṣkala, and Sakala but to render them comprehensible to his audience, he indicates to them using Buddhist terms. The question of mutual influence on this issue is not easy to answer. Although both show a tripartite formation, they are based on two totally independent and original concepts.
ahetau hetudṛṣṭir amārge mārgadṛṣṭiḥ śīlavrataparāmarśaḥ. tad yathā maheśvaro na hetur lokānām, taṃ ca hetuṃ paśyati prajāpatim anyaṃ vā. agnijalapraveśādayaś ca na hetuḥ svargasya, tāṃś ca mārgaṃ paśyati. śīlavratamātrakaṃ sāṃkhyayogajñānādayaś ca na mārgo mokṣasya tāṃś ca mārgaṃ paśyati. …
yas tarhi jalāgnipraveśādibhiḥ svargopapattiṃ paśyati śīlavratena vā śuddhim, so ’pi duḥkhadarśanaprahātavya eva. eṣa hi śāstrapāṭhaḥ— “ye caivaṃdṛṣṭaya evaṃvādino yad ayaṃ puruṣapudgalo gośīlaṃ samādāya vartate mṛgaśīlaṃ kukkuraśīlam, sa tena śudhyati, mucyate, sukhaduḥkhaṃ vyatikrāmati, sukhaduḥkhavyatikramaṃ cānuprāpnoti. akāraṇaṃ kāraṇataḥ pratyeti śīlavrataparāmarśo duḥkhadarśanaprahātavyaḥ” iti vistaraḥ.
See Pasadika 1989: citation no. 366 for the identification of this citation from the Jñānaprasthānaśāstra, one of the Sarvāstivāda Abhidharma texts believed to be composed around 150bce.
On this issue of transfer of merit, see Hara 1994 (republished as article no. 8 in Hara 2002). However, he has not investigated into the Vedic roots of this concept. At a later occasion, I intend to make a separate investigation into the Vedic roots of this concept.
When a daṇḍa is by mistake associated with the preceeding akṣara, it can be interpreted as the mātrā of ā. Once this happened, the pṛṣṭhamātrā of e can easily be ignored.
One can still argue in favour of vā, and say that it can be taken as the inclusive ‘or.’ Indeed, in the version available to Kauṇḍinya, it should have been that way, but not in the original form.
Kauṇḍinya takes the root krath- in the sense of snoring. This interpretation gets support from the usage of the same root in the sense of rattling in some Āyurveda texts (Oberlies 2000: 178, fn. 78). So could one alternatively take it to refer to snoring loudly, perhaps with a rattling sound, which fits for a bull, too. However, I prefer Bhāskara/Sāyaṇa’s interpretation ‘to hurt or injure,’ because this meaning is commonly attested and is equally appropriate to the context.
This enhanced explanation of the meaning of the root spand could have served as the ground for philosophisation of spanda in later times in the Śaiva domain as the ultimate reality, the vibration of consciousness.
Geldner’s footnote to his translation of ṚV 4.3.10c: “spand wird im Veda nur vom Ausschlagen der Kuh während des Melkens gebraucht: AV. 8,6,17 (spandanā́), Ait. Br. 5,27,7. So auch hier von dem Stier, der nach dem beliebten Paradoxon in d gemolken wird.”
It is obviously the case as late as the time of Śaṅkara. Commenting on BĀU(K) II.3.6, when he needs to describe how people commonly refer to a bull, he spontaneously uses this root to describe its typical behaviour, whether we interpret it here as kicking or twitching of limbs: gaur asau spandate śuklo viṣāṇīti yathā loke nirdiśyate…
In favour of this meaning, they identify the root as maṭi unmāde but such a root is not found in the available Pāṇinian Dhātupāṭha.
In peculiar contexts, the denominative root śṛṅgāy can mean ‘to feel horny’, and most probably, this interpretation was behind the shift of Pāśupatasūtra reading from śṛṅgāyeta to śṛṅgāreta. In this situation, one can argue that in our prescription too, from the beginning, this figurative meaning will have been valid. Though this can be made to fit with a bull’s behaviour, I do not find this preferable to the literal meaning of butting which comes to mind as a characteristic defining an animal more naturally than its sexual behaviour.
atra bhūtapatir nāma sarvabhūtamaheśvaraḥ | bhūtaye sarvabhūtānām acarat tapa uttamam ‖ atra govratino viprāḥ svādhyāyāmnāyakarśitāḥ | tyaktaprāṇā jitasvargā nivasanti maharṣayaḥ ‖ yatratatraśayo nityaṃ yenakenacid āśitaḥ | yenakenacid ācchannaḥ sa govrata ihocyate ‖
Brahmāṇḍapurāṇa II.74.46cd–50ab: athauśijo bṛhatkīrtir bṛhaspatir ivaujasā ‖ ūrdhvaretās tataś cāpi nyavasad bhrātur āśrame | godharmaṃ saurabheyāt tu vṛṣabhāc śrutavān (chatavān Ed.) prabhoḥ ‖ tasya bhrātā pitṛvyas tu cakāra bhavanaṃ tadā | tasmin hi tatra vasati yadṛcchābhyāgato vṛṣaḥ ‖ darśārtham āstṛtān darbhāñ cacāra surabhīsutaḥ | jagrāha taṃ dīrghatamā visphurantaṃ tu śṛṅgayoḥ ‖ sa tena nigṛhītas tu na cacāla padāt padam |
It is possible that this brother cousin, who seems to be called Śaradvat, is in fact a half-brother: Dirghatamas and Śaradvat are fathered by two brothers, Bṛhaspati and Uśija, and their mother is the same.
Brahmāṇḍapurāṇa II.74.50cd–55: tato ’bravīd vṛṣas taṃ vai muñca māṃ balināṃ vara ‖ na mayāsāditas tāta balavāṃs tvadvidhaḥ (tadvidhaḥ Ed.) kvacit | tryaṃbakaṃ vahatā devaṃ yato yāto (jāto Ed.) ’smi bhūtale ‖ muñca māṃ balināṃ śreṣṭha pratisnehaṃ varaṃ vṛṇu | evam ukto ’bravīd enaṃ jīvaṃs tvaṃ me krathiṣyasi (kva yāsyasi Ed.) ‖ tena tvāhaṃ na mokṣyāmi parasvādaṃ catuṣpadam | tatas taṃ dīrghatamasaṃ sa vṛṣaḥ pratyuvāca ha ‖ nāsmākaṃ vidyate tāta pātakaṃ steyam eva ca | bhakṣyābhakṣyaṃ na jānīmaḥ peyāpeyaṃ ca sarvaśaḥ ‖ kāryākāryaṃ ca vai vipra gamyāgamyaṃ tathaiva ca | na pāpmāno vayaṃ vipra dharmo hy eṣa gavāṃ śrutaḥ ‖
The available reading, kva yāsyasi, is emended to krathiṣyasi. With this emendation this sentence becomes syntactically perfect and the following sentence logical. I am grateful to Prof. Yuko Yokochi for this suggestion.
Brahmāṇḍapurāṇa II.74.56–64: gavāṃ nāma sa vai śrutvā saṃbhrāntas tam amuñcata | bhaktyā cānuśravikayā gosutaṃ vai prasādayan ‖ prasādato vṛṣendrasya godharmaṃ jagṛhe ’tha saḥ | manasaiva tad ādadhre tadvidhas tatparāyaṇaḥ ‖ tato yavīyasaḥ patnīm autathyasyābhyamanyata | viceṣṭamānāṃ rudatīṃ daivāt saṃmūḍhacetanaḥ ‖ avalepaṃ tu taṃ matvā śaradvāṃs (suradvāṃs Ed.) tasya nākṣamat | godharmaṃ (godharma Ed.) vai balaṃ kṛtvā snuṣāṃ sa hy abhyamanyata ‖ viparyayaṃ tu taṃ dṛṣṭvā śaradvān pravicintya ca | bhaviṣyam arthaṃ jñātvā ca mahātmā tv avamatya tam ‖ provāca dīrghatamasaṃ krodhāt saṃraktalocanaḥ | gamyāgamyaṃ na jānīṣe godharmāt prārthayan snuṣām ‖ durvṛttaṃ tvāṃ tyajāmy eṣa gaccha tvaṃ svena karmaṇā | yasmāt tvam andho vṛddhaś ca bhartavyo duranuṣṭhitaḥ ‖ tenāsi tvaṃ parityakto durācāro ’si me mataḥ | sūta uvāca | karmaṇy asmiṃs tataḥ krūre tasya buddhir ajāyata ‖ nirbhartsya caiva bahuśo bāhubhyāṃ parigṛhya ca | kāṣṭhe samudge (koṣṭhe samudre Ed.) prakṣipya gaṅgāmbhasi samutsṛjat ‖
This reference to Vedic (ānuśravika) devotion suggests that this verse aims at alluding to the govrata carried out during the performance of a gosava. Furthermore, I feel that the expression gosutaṃ vai prasādayan might also be translated as “performing the gosuta, i.e. gosava.” In that case, we could therefore argue in favour of reading the first pāda differently: gavāṃ nāma save instead of gavāṃ nāma sa vai of the edition, and translate that as “in the sacrifice associated with bulls.” However, this interpretation would disturb the story. So, I guess that the fact of Dīrghatamas’ adoption of godharma in a session of the gosava is known from some tradition to the redactor of this story and he has slightly modified it in order to fit it in the story.
This story emphasises on the point that Dīrghatamas assaulted a daughter-in-law to satisfy his sexual desire. The story puts in the mouth of Śaradvat the view that Dīrghatamas has misapprehended the rule of godharma. In this way, this story appears not to be criticising the vow itself but rather its misapprehension.
Autathya Junior seemingly is a son of Śaradvat, because the latter calls his wife a ‘dauther-in-law.’ This Autathya is distinguished as ‘Junior’ apparently to avoid confusion. For, in the Mahābhārata and Bṛhaddevatā the father of Dīrghatamas is named as Utathya/Ucathya, and thus Autathya, a patronimic derived from this name, can refer to any member of Utathya’s extended family, and in fact, it is used to refer to Dīrghatamas himself in other contexts. Since Śaradvat and Dīrghatamas are brothers, either half-brothers or cousins, Autathya Junior’s wife is their daughter-in-law, and Dīrghatamas’ lust for her amounts to incest.
This refers to Dīrghatamas’s involvement in the production of Bali’s heirs, a plot found in the conclusion of this story; cf. Brahmāṇḍapurāṇa II.74.65–94.
In view of the Pāśupata practice of soliciting contempt (avamāna), we must pay attention to the component of contempt present in this story. Here Dīrghatamas is despised by Śaradvat (cf. Brahmāṇḍapurāṇa II.74.60d: avamatya). This is in confirmity with Pāśupatasūtra 3.3 which tells that a Pāśupata is ‘despised’ (avamataḥ) by others. Commenting on this sūtra, Kauṇḍinya quotes two verses which are in slightly modified form known from older sources, the Mahābhārata (12.222.20–21) and the Manusmṛti (2.162–63).
Jaiminīyabrāhmaṇa 2.113: tasya vratam. upa mātaram iyād upa svasāram upa sagotrām. upāvahāyodakam ācāmed upāvahāya tṛṇāny āchindyāt. yatra yatrainaṃ viṣṭhā vindet tat tad vitiṣṭheta. anaḍuho ha lokaṃ jayati. tena haitena janako vaideho iyakṣāṃ cakre. tam u ha brāhmaṇā abhito niṣeduḥ. sa ha papraccha ka stoma iti. sa hovāca sudakṣiṇaḥ kṣaimir nāyaṃ trivṛd evāhur iti. tasya kiṃ vrataṃ kā dakṣiṇā iti. tasmā u haitat provāca. sa hovāca ud evāsya dakṣiṇā āśaṃse vrataṃ tv evāsya nodāśaṃsa iti. teno ha sa nopadadharṣa yaṣṭum. teno ha puṇyakeśo yaudhenir īje śaibyo rājā. taṃ ha sabhāyām eva viṣṭhā viveda. sa ha tad evodāvṛṇāna uvāca sthavirayajño vāva kilāyam āsa. sthavira evānena yajñena yajeta. sthavirasya hy evedaṃ sarvam anujñātam iti. sa haiṣa sthavirayajña eva. tena haitenottaravayasy e[va] yajeta.
Rau (1983) suggested to correct upāvahāya to upāhvarya in this passage. But that does not seem right to me. First of all, the meaning of the verbal combination upāvahā-, ‘to lean down close,’ fits better in this context, because here the sacrificer who imitates the bull is supposed to lean down close to the water-source and drink water directly with his mouth. On the other hand, the suggested combination of upāhvar-, which is not attested, must remain close to the meaning of the attested combination of upahvar- which is connected with the movement of a river. It means ‘to reach or approach the end by an uneven and deviative course, curving and turning here and there,’ and this meaning does not fit our context. Besides, Vasubandhu quotes a passage (bhāṣya on Abhidharmakośa 4.68cd) from some otherwise unknown Brāhmaṇa or Śrautasūtra text as he criticises Brahmanical perversions. That passage agrees with our passage up to this point. There the descriptive noun upahā derived from the same root hā- with the preverb upa is used. This supports the reading of the edition, upāvahāya, formed from the same root and prefix, with one more prefix ava in the middle. Let me cite and translate that passage which Vasubandhu quoted:
yathoktam: brāhmaṇo gosaveneṣṭyā saṃvatsaragovratī bhavati | upahā udakaṃ cūṣati tṛṇāni chinatti upaiti mātaram upa svasāram upa sagotrām iti |
As it is said: by the Gosava [sacrifice], a brahmin assumes for one year the vow of the bull. As he is leaning down, he sucks up water, cuts grasses [with his teeth], [and sexually] approaches his mother, sister, or a lady of the same family.
Yaśomitra comments on upahā and says that it signifies the sacrificer. Rau is not aware of this passage from the Abhidharmakośabhāṣya but of a part of it cited in Kamalaśīla’s commentary on Tattvasaṃgraha 2786 (cf. Rau 1983: 143). There the text has been corrupted but still the word upahā is intact. Nevertheless, without any argument, Rau proposes to read upahva in place of upahā.
Thus, as far as I can see, all the evidence goes against Rau’s proposal: i) the reading upāvahāya of the edition is grammatically perfect, ii) its meaning fits best in the context, iii) the Vedic passage Vasubandhu quoted comes from an independent source and involves the same root hā- with one of the preverbs present in the above reading, iv) Vasubandhu, Yaśomitra, and Kamalaśīla, all read upahā udakaṃ… in the passage they quoted, v) the combination of upāhvar- that Rau suggested is unattested, and the closest combination of upahvar has a different nuance.
On the other hand, Rau himself has informed us that “upāvajihīte seems to reoccur” in the Śatapathabrāhmaṇa (Kāṇva) 126.96.36.199. I would say it is quite certain though it is emended from three manuscript readings: upāvajjihīte, uyāvajjihīte, and upājjihīte (cf. Rau 1983: 143). First of all, the first two manuscript readings are very close to the emended reading. And more, we can be sure about the emendation by reading the passage in its context. There a person as he is breathing out leans (jihīte) down (ava) closer (upa) to the ground, and we know from the context that the avāna aspect of breathing is implied. Then nearby in the same paragraph, the same verb jihīte appears in the same fashion with two preverbs: the first preverb is the same but the second one ut is the opposite of ava, and this time the udāna aspect of breathing is concerned. Thus, the verb jihīte is confirmed, as are the preverbs.
The threefold Stoma is the recitation of ṚV 9.2 in a specific order. Caland (1919: 157 §135) leaves “…” instead of translating Sudakṣiṇa’s answer: ‘nāyaṃ trivṛd evāhur iti’. Otherwise, I have benefitted from his translation.
Madhyamakahṛdaya 9.61–62: yat kleśadahanāyālaṃ taj jñānaṃ jñānino viduḥ | nātaḥ prakurute pāpaṃ jñānī taddhetvasaṃbhavāt ‖ siddhiyogī na lipyeta karmaṇā pātakena vā | iti bruvāṇaiḥ sanmārgān naṣṭair anye ’pi nāśitāḥ ‖
Interesting is Arrian’s remark that “… this must be another Heraklês” (see the passage cited on the previous page). So, one should not make the mistake of identifying the deity of Śibis with Heraklês on the ground that both hold a staff. As Swanbeck writes, “No writer before Alexander’s time mentions the Indian gods. The Makedonians, when they came into India, in accordance with the invariable practice of the Greeks, considered the gods of the country to be the same as their own. … and whenever, as among the Sibae, they saw the skins, or club, or the like, they assumed that Hérakles had at some time or other dwelt there” (as cited in Schwanbeck & McCrindle 1877: 111–112).
For example, according to Harivaṃśa 2.73, Kṛṣṇa fights and defeats Indra because he refuses to give him the Pārijāta tree. In another episode (cf. Harivaṃśa 2.17–20), Indra is furious that the pastoral community has ceased to celebrate his festival and releases excessive rain. But Kṛṣṇa lifts up the Govardhana mountain to provide shelter to the cows and his people. Indra gives up his pride and comes to praise Kṛṣṇa. It is interesting that the next chapter of the Harivaṃśa narrates a story of Kṛṣṇa’s killing of a demon in the form of a bull. Kṛṣṇa’s lifting of the Govardhana mountain is also mentioned in the Mahābhārata (5.128.45cd) in a half-verse.
It is also noteworthy that in Mahābhārata 12.64–65 Viṣṇu first appears as Indra in front of Māndhātṛ who is desirous to see Nārāyaṇa.
On the other hand, the threefold Stoma (ṚV 9.2) chanted in the gosava addresses Soma as a bull, and thus hints at the possibility that Soma was once equated with the celestial bull/draft-ox.
Kauṇḍinya has taken it as vā ignoring sandhi as well as syntax. This shows that Kauṇḍinya is not familiar with the Vedic syntax of the sūtra text.
See Sakamoto-Gotō 2000 for the meaning of iṣṭāpūrtá. She also discusses the loss of iṣṭāpūrtá in certain situations in the third section of her article.
For Kauṇḍinya, this is a given fact and he does not try to find out whether there is something in the Vedic corpus to substantiate this claim.
The order of stanzas is different in the AVP. So, the stanzas I am citing here are placed there as the third, fifth, and fourteenth. The first two of these stanzas are almost identical in both versions but the last reads quite differently in the second-half.
I translate this passage on the basis of Whitney’s translation, but I have revised it according to my understanding after reading AVP 17.27–29. For the sake of comparison, I cite here Whitney’s traslation (Whitney 1905: 163–165):
“The draft-ox [is] Indra; he looks out from (for?) the cattle; triple ways the mighty one (çakrá) measures out (traverses?); yielding (duh) the past (? bhūtá), the future, existing things (bhúvana), he goes upon (car) all the courses (vratá) of the gods.
Born an Indra among human beings (manuṣyà), he goes about (car) shining brightly, a heated hot-drink (gharmá); he, being one of good offspring, shall not go in mist (? udārá) who, understanding [it], shall not partake of (aç) the draft-ox.
Indra by form, Agni by carrying (váha), Prajāpati, Parameshṭhin, Viraj; in Viçvānara he strode, in Vāiçvānara he strode, in the draft-ox he strode; he made firm, he sustained.”
I follow the suggestion of Narten (1964: 270) who draws attention to other parallel expressions of dāré sṛ- in the Jaiminīya- and Pañcaviṃśa-brāhmaṇas. I am grateful to Werner Knobl for drawing my attention to this solution. The expression ‘not running into the cleft on the path’ conveys the idea of safety. Like here, once more in AVP 5.12.5b we find two rewards of good offspring and safety mentioned together: suprajāstvāya bhadrāya.
This appears to me the earliest statement against beef-eating. This is followed by a statement in the Śatapathabrāhmaṇa (188.8.131.52) which prohibits eating the meat of the dairy cow and the draft-ox.
As I am told, the third volume of Dipak Bhattacharya’s edition of the AVP, containing the 17th and 18th Books, is already in press. Therefore, for the time being I postpone my translation and analysis of this passage. I will eventually present it in a separate article.
Musalīśa means essentially the same as Lakulīśa: the Lord with a club. It is worth noticing that the 11th-century Śaiva exegete Kṣemarāja (cf. Uddyota on Svacchandatantra 10.1134– 1135) names this direct disciple of Lakulīśa as Musalendra, which is not different from Musalīśa in meaning. As Kṣemarāja states, it was a disciple of Lakuleśa called Musalendra who took out six ancillary texts attached to one of the eight Pramāṇa texts, and on the basis of those texts introduced a new ritual-based order, whereas the older Lākula order was centered on gnosis (see Sanderson 2006: 176–177).
This relationship of Lakulin/Lakulīśa/Lakuleśa and Musalīśa/Musalendra as teacher and disciple was known to the tradition apparently already a few centuries before Kṣemarāja. For, the two are presented that way already in the 7th-century Junwani copperplate inscription of Mahāśivagupta Bālārjuna placed at the source of a lineage of some Śaiva teachers whose names end in -soma. Though the name of Lakulīśa’s disciple is spelt as Mugalīśa in this inscription, we have to correct it to Musalīśa in the light of Kṣemarāja’s account and orthographical closeness of ga and sa. And more, Mugalīśa has no sensible meaning, but if it is corrected to Musalīśa, then it becomes equivalent in meaning to Kṣemarāja’s Musalendra. See also Bakker 2000.
“Those twelve nights they declare to be for the vow of Prajāpati; whoso knows the bráhman within them—that verily is the vow of the draft-ox.” (Whitney 166.) (dvā́daśa vā́ etā́ rā́trīr vrátyā āhuḥ prajā́pateḥ | tátrópa bráhma yó véda tád vā́ anaḍúho vratám ‖)
It is, however, surprising that in the body of Pāśupatasūtra Rudra’s identity as Pāśupati is not highlighted, even though the cult practice taught in that text is twice (sūtras 1.1 and 4.10) referred to as Pāśupata, namely, belonging or related to Paśupati. Particularly in the five brahmamantras, other designations of Rudra like Bhava, Śarva, Mahādeva, and Īśāna are mentioned but Paśupati is missing. Rudra is repeatedly mentioned in these mantras and also the sūtras, but Paśupati does not appear even as a description of Rudra. While according to the Maitrāyaṇīsaṃhitā (IV.2.12), Bhava and Śarva are unfriendly and unpeaceful names and should not be uttered, Rudra and Paśupati are friendly and peaceful ones, and so should be uttered.
Lalitavistara, Parivarta 17 (Vaidya pp.182–183): govratamṛgaśvavarāhavānarahastivrataiś ca … śuddhiṃ pratyavagacchanti saṃmūḍhāḥ.
Bhattacharya 1957: 517–518: yathāpīhaikatyaḥ kukkuravratena śuddhiṃ manyate govratena nakulavratena nagnavratena bhasmavratena …
Ingalls 1962: 296. At that early stage of research, he was very clever to “imagine both the cynic and the Pāśupata cults to have derived from sects of men who performed beast-vows” (Id 297). But there he has gone a bit far in his imagination and sought the roots of “the dishonest transfer of hidden forces by the Pāśupata” and the resultant dishonour of him in the black side of Shamanism. Now since we have seen that the roots of the Pāśupata vrata reaches the Vedic Saṃhitās, it is clear that it might in fact even be an Indo-European custom developing in parallel but slightly differently on either side of the India-Europe divide, and there may be no need of involving the black side of shamanism.
See Sanderson 2006Goodall & Isaacson 2007. Similarly the Pāśupatavidhi of the Atharvapariśiṣṭas paraphrases a large number of sūtras from the Pāśupatasūtra; See Bisschop & Griffiths 2003. But like any of chapter of the Atharvapariśiṣṭas it is very difficult to tell which time period this chapter belongs to.
See Bakker 2002 and Bisschop 2006a.
See Pasadika 1989: citation no. 366 for the identification of this citation from the Jñānaprasthānaśāstra one of the Sarvāstivāda Abhidharma texts believed to be composed around 150bce.
Oberlies 2000: 178–179.
Oberlies 2000: 179fn. 18.
Ingalls 1962: 295.
Rau (1983) suggested to correct upāvahāya to upāhvarya in this passage. But that does not seem right to me. First of all the meaning of the verbal combination upāvahā- ‘to lean down close’ fits better in this context because here the sacrificer who imitates the bull is supposed to lean down close to the water-source and drink water directly with his mouth. On the other hand the suggested combination of upāhvar- which is not attested must remain close to the meaning of the attested combination of upahvar- which is connected with the movement of a river. It means ‘to reach or approach the end by an uneven and deviative course curving and turning here and there’ and this meaning does not fit our context. Besides Vasubandhu quotes a passage (bhāṣya on Abhidharmakośa 4.68cd) from some otherwise unknown Brāhmaṇa or Śrautasūtra text as he criticises Brahmanical perversions. That passage agrees with our passage up to this point. There the descriptive noun upahā derived from the same root hā- with the preverb upa is used. This supports the reading of the edition upāvahāya formed from the same root and prefix with one more prefix ava in the middle. Let me cite and translate that passage which Vasubandhu quoted: yathoktam: brāhmaṇo gosaveneṣṭyā saṃvatsaragovratī bhavati | upahā udakaṃ cūṣati tṛṇāni chinatti upaiti mātaram upa svasāram upa sagotrām iti | As it is said: by the Gosava [sacrifice] a brahmin assumes for one year the vow of the bull. As he is leaning down he sucks up water cuts grasses [with his teeth] [and sexually] approaches his mother sister or a lady of the same family. Yaśomitra comments on upahā and says that it signifies the sacrificer. Rau is not aware of this passage from the Abhidharmakośabhāṣya but of a part of it cited in Kamalaśīla’s commentary on Tattvasaṃgraha 2786 (cf. Rau 1983: 143). There the text has been corrupted but still the word upahā is intact. Nevertheless without any argument Rau proposes to read upahva in place of upahā. Thus as far as I can see all the evidence goes against Rau’s proposal: i) the reading upāvahāya of the edition is grammatically perfect ii) its meaning fits best in the context iii) the Vedic passage Vasubandhu quoted comes from an independent source and involves the same root hā- with one of the preverbs present in the above reading iv) Vasubandhu Yaśomitra and Kamalaśīla all read upahā udakaṃ… in the passage they quoted v) the combination of upāhvar- that Rau suggested is unattested and the closest combination of upahvar has a different nuance. On the other hand Rau himself has informed us that “upāvajihīte seems to reoccur” in the Śatapathabrāhmaṇa (Kāṇva) 184.108.40.206. I would say it is quite certain though it is emended from three manuscript readings: upāvajjihīte uyāvajjihīte and upājjihīte (cf. Rau 1983: 143). First of all the first two manuscript readings are very close to the emended reading. And more we can be sure about the emendation by reading the passage in its context. There a person as he is breathing out leans (jihīte) down (ava) closer (upa) to the ground and we know from the context that the avāna aspect of breathing is implied. Then nearby in the same paragraph the same verb jihīte appears in the same fashion with two preverbs: the first preverb is the same but the second one ut is the opposite of ava and this time the udāna aspect of breathing is concerned. Thus the verb jihīte is confirmed as are the preverbs.
Parpola 2002: 365.
Dahlquist 1962: 73–93.
See Oberlies 2000: 183 for further arguments on Indra-Rudra affinity in the Pāśupata sphere.
Cf. Acharya 2005.
Cf. Mylius 1976.
For details see Bakker 2010.
See Ingalls 1962.
Bhattacharya 1957: 517–518: yathāpīhaikatyaḥ kukkuravratena śuddhiṃ manyate govratena nakulavratena nagnavratena bhasmavratena …
Ingalls 1962: 296. At that early stage of research he was very clever to “imagine both the cynic and the Pāśupata cults to have derived from sects of men who performed beast-vows” (Id 297). But there he has gone a bit far in his imagination and sought the roots of “the dishonest transfer of hidden forces by the Pāśupata” and the resultant dishonour of him in the black side of Shamanism. Now since we have seen that the roots of the Pāśupata vrata reaches the Vedic Saṃhitās it is clear that it might in fact even be an Indo-European custom developing in parallel but slightly differently on either side of the India-Europe divide and there may be no need of involving the black side of shamanism.