Version classiqueVersion mobile

Rites hindous

Gérard Colas
Gilles Tarabout

Genèses et transformations

Is the dīkṣā to be performed by the priest?

An analysis of vedic texts

Shingo Einoo


le prêtre peut-il appliquer la dīksā à lui-même ? une analyse des textes védiques

Le rituel du Śautra comprend une cérémonie appelée dīkṣa qui est exécutée par le sacrifiant et son épouse. C’est une cérémonie préparatoire pour rendre le sacrifiant et son épouse rituellement purs en sorte qu’ils soient qualifiés pour participer à un rituel de soma. Il est à souligner que c’est le sacrifiant et sa femme qui subissent la dīkṣā mais jamais les prêtres officiants.

Mais dans les textes rituels post-védiques, il semble que dans certains cas le prêtre officiant lui-même se livre à une cérémonie préliminaire pour se purifier rituellement. Cette cérémonie préliminaire pour le prêtre consiste en un jeûne de plusieurs jours et en un bain, ce qui paraît être très simple comparativement à la dīkṣā védique. En substance, cependant, la procédure et la fonction sont semblables à celles de la dīkṣa védique. Donc, on peut l’interpréter comme une « dīkṣā pour le prêtre ». Dans cet article, je présente des passages de textes védiques tardifs et post-védiques où la « dīkṣā pour le prêtre » est décrite, et je discute du fait que cette cérémonie a été pour la première fois imposée au prêtre dans la description des rites kāmya ou des rites exécutés dans le but de satisfaire certains souhaits inscrits dans les textes védiques tardifs.

Texte intégral

Introductory remarks

The Bṛhatsaṃhitā 45.15 reads as follows:
buddhvā devāvikaṃ śuciẖ purodhās tryahoṣitaḥ snātaḥ/ snānakusumānulepanavastrair abhyarcayet pratimām//
When he notices an unfavorable change of an image of god, the purohita, who is ritually clean by fasting for three days and taking a bath, worships the image by bathing it and offering flowers, fragrant pastes, and clothes.

1Here it seems that the purohita or the family priest of a king, who is acting as a priest in a propitiating rite for averting evil effects resulting from the unfavorable change of an image of god, has to perform a certain kind of preparatory actions consisting of three days’ fasting and bathing in order to make oneself ritually pure. G. Colas collects many examples of thedīkṣā to be undergone by the temple-priests from the Vaikhānasa medieval corpus and discusses them (see Colas 1999: 41-44). The adhivāsana which is the first stage of the dīkṣā described in the Tantrāloka appears to be nothing but the dīkṣā to be performed by the guru himself to make himself Śiva in order to perform the following dīkṣā for a disciple (see Takashima 1992: 48).

  • 1 The procedure of the dīkṣā is described, for example, by Caland & Henry 1906: 11-22. The Bṛhatsaṃhi (...)

2In the Śrauta ritual, there is a preparatory action which makes someone ritually clean. It is called dīkṣā. As regards the performer of the dikṣā, Jan Gonda says in his Change and Continuity...: “An interesting feature is that most forms of the Indian dīkṣā discussed in this chapter are to be undertaken, not by a priest, but by a private person who in so doing has his own objects and interests in view, [...]. A Vedic priest and many officiants in non-Vedic cults are not ordained or consecrated” (Gonda 1965: 316). In the Bṛhatsaṃhitā the word dīkṣā is not used, and the ritual procedure of the dīkṣā laid down for a yajamāna (sacrificer) in the soma sacrifice is much more complicated than in the simple description of the Bṛhatsaṃhitā.1

  • 2 For these regulations, see, e.g. Hillebrandt [1879] 1977: ix, and Rustagi 1981: 27-40.
  • 3 For the equation of the upavasatha with the dikṣā, see Kaelber 1989: 51-52.

3The Śrauta ritual, further, has another kind of ritual regulation to be observed by a yajamāna and his wife. The yajamāna and his wife, e.g. at the time of the new and full moon sacrifice, spend the night before the main day of this sacrifice by following certain regulations.2 This ritual act, namely the upavasatha, is not called dīkṣā and it is much simpler than the dīkṣā of the soma sacrifice. But, at least in that it is undertaken by the yajamāna and his wife in order to make themselves fit to perform the sacrifice, this ritual regulation can be said similar to the dīkṣā. BaudhŚS 24.21 [206,5], after describing the upavasatha, summarizes the point by saying: “He should accomplish the outer appearance of the dīkṣā.” From this we can surmise that the performance of the concrete actions which constitute the preparation of the upavasatha is equated with the performance of the dīkṣā.3

4So the point is that, while in the Śrauta ritual the priest is not obliged to undertake any ritual observances before the perfomance of a sacrifice, there are, at least in the Bṛhatsaṃhitā and some medieval religious texts, passages which suggest that the officiating priest is demanded to undertake certain preliminary actions to make himself ritually clean which can be interpreted as a “dīkṣā for the priest” (I will use hereafter this expression). In the following I will collect passages of the “dīkṣā for the officiating priest” mainly from the late Vedic texts and post-Vedic Vidhāna texts and try to make clear that this minute change in the development of Hindu ritual occurred in the late Vedic texts and post-Vedic Vidhāna texts in a special context, namely in the description of the kāmya rites.


5The Kauśikasūtra belongs to the tradition of the Atharvaveda. KauśS 7.29- 8.1 reads as follows:

[7.29.] “Being pure, one should perform ritual.”
[8.1.] “In the nocturnal ritual which contains the former offerings (purastāddhoma), one sweeps the entrance of the hut, while one carries the sacrificial cord over the left shoulder and under the right armpit. After having eaten food before one’s fast, taken a bath, and put on one’s new clothes, one performs ritual.”

  • 4 For the classification of ritual rites described in the first half of the Kauśikasūtra, see Bloomfi (...)
  • 5 See also Dārilabhāṣya on KauśS 8.1: kartṛkārayitror ayam upadeśaḥ [...] prayogārthaṃ ca kṛcchraṃ pr (...)

6These sūtras appear in the description of the general rules for the performance of main ritual rites described in the first half of the Kauśikasūtra. These rites are classified into six groups: bhaiṣajyāni or remedial rites, āyuṣyāni or rites for promoting long life, sāṃmanasyāni or rites for concord, pauṣṭikāni or rites for promoting welfare, strīkarmāṇi or rites performed for women, and rājakarmāṇi or rites performed for a king.4 This classification shows that the rites of the Kauśikasūtra are not the nitya rites or regular ones, rather the so-called kāmya rites or rites to be performed with a view to obtaining certain desires. According to KauśS 7.29-8.1, therefore, the officiating priest of the kāmya rites is required to be ritually clean. As regards to KauśS 7.29 V. Henry admits that in the case of magical rites not only the officiating priest but also his client must submit themselves to certain preliminary lustrations (Henry [1904] 1988: 44).5

7This prescription of the Kauśikasūtra reminds us of a prescription given in BaudhhŚS 26.5 [276,7-8]. Caland (1908] 1968: I) paraphrases it as follows: “2. Vorbereitung zur Isti. Im Baudhāyanasūtra wird gelehrt, dass er (d.h. wohl sowohl Yajamāna wie Adhvaryu und die anderen Priester) frei von Unrat die Iṣṭi unternehmen soll, dass er ein Jahr lang oder zwölf Tage nur von Milch leben soll oder solange er es wunchen mag.” According to his interpretation the persons who are required to perform these preliminary consecrations are not only the yajamāna but also the officiating priests of the kāmyeṣṭis.

  • 6 See also KhādGS 4.1.1-6.

8For the so-called dīkṣā for the priest before performing the kāmya rites see also the Gobhila Gṛhyasūtra = GobhGS 4.5.9-13. “9. When undertaking ceremonies for the obtainment of special wishes, let him fast during three (days and) nights. 10. Or (let him omit) three meals. 11. At such ceremonies, however, as are repeated regularly, (let him do so only) before their first performance. 12. He should (simply) fast, however, before such ceremonies as are performed on sacrificial days (i.e. on the first day of the fortnight). 13. (At a ceremony) which ought to be performed immediately (after the occurrence by which it has been caused), the consecration follows after (the ceremony itself).” (Transl. Oldenberg 1892: 115-116.)6

  • 7 See also Gonda 1965: 323-324; Henry [1904] 1988: 223-224; Shende 1985: 154-155.

9In the classification of the rites to be performed by using the mantras of the Atharvaveda, M. Bloomfield adds, as the seventh one, the abhicārika or magical and imprecatory rites. KauśS 47.12-22 prescribes preliminary actions performed at the time of magical rites. Even though these actions are not called dīkṣā in the text itself, W. Caland, in his translation, titles this section “Die Weihe (dīkṣā) dessen, der die Behexung halten will.” J. Gonda (1980: 464) paraphrases this difficult passage as follows: “Consecration (dīkṣā), a well-known element of the Śrauta ritual, is the subject of the passage KauśS. 47.12- 22 teaching how to prepare oneself and one’s requisites, a girdle and a staff, for definite acts of ‘sorcery’: one has to cut a staff, to put a special girdle in sour milk and honey, to girdle oneself with it, to besmear the staff with the residue of oblations, to mutter mantras over it, etc.” He also uses the word dīkṣā for these actions. For Caland and Gonda it was natural to think about the dīkṣā when they read this passage, because these actions are preparatory or preliminary for other ritual rites which follow, even if the person who undertakes them is the priest, not the yajamāna.7

10Thus we can certainly assume that in the tradition of the Atharvaveda the officiating priest of the kāmya rites, including the imprecation (abhicāra), is demanded to perform certain preliminary actions to make himself ritually fit for the performance like the yajamāna of the Śrauta ritual.

11The Atharvavedapariśiṣṭa, composed between the second century B. C. and the fifth century A. D., also belongs to the tradition of the Atharvaveda. It is regarded later than the Kauśikasūtra (see Modak 1993: 471). As regards the so-called dīkṣā for the priest, the Atharvavedapariśiṣṭa gives an interesting statement:

[11.1.3] “The priest and the yajamāna, having their hair cut and their beard
shaved, let their hair on the body be shaved and their nails cut.”

12AVPŚ 13.1.3 states almost the same thing. It is interesting to know that the priest undergoes the same preparatory procedure as the yajamāna.

13In the description of the small one hundred thousands burnt offerings (laghulakṣahoma) AVPŚ 30.1.5cd-2.2ab describes the conditions necessary for becoming an officiating priest:

“Those wise Brahmins who have finished their studies of the Vedas (vedasaṃpannāḥ) and who have collected the knowledge of the Atharvaveda ritual
(brahmakarmasamādhayaḥ), [...] should recite the gāyatrī-verse (RV 5.82.1)
ten thousand times after fasting one day. They, wearing white clothes,
having bathed, being decorated with garlands and perfumes, having taken
no food, being restrained and satisfied, and having subdued the senses,
sit down on a seat made of kuśa grass and perform the excellent fire-offering.”

  • 8 In the AVPŚ we find similar descriptions of the so-called dīkśā to be performed by the priest, for (...)

14Here a number of acts are imposed on the Brahmins who are to act as officiating priests.8

15The Atharvavedapariśiṣṭa is a ritual text for the domestic chapelain (purohita) and the purohita stands in a very close relation to the king. It is understandable that the purohita performs the ‘dīkṣā’, because he represents the king himself; he is, in a sense, the king, being a yajamāna, on behalf of whom the purohita, at the same time, acts as an officiating priest. The purohita plays, thus, the double role of the officiating priest (ṛtvij) and yajamāna, therefore he must perform the ‘dīkṣā’, and the ritual acts to be performed by him for the sake of the king are, again, mostly kāmya rites. So, it is necessary for him to do so for this reason, too.


16I now turn to a group of texts which are generally called Vidhānas. There are three published texts: the Ṛgvidhāna, the Sāmavidhānabrāhmaṇa, and the Śuklayajurvidhāna. M. S. Bhat (1987: 18) sums up their characteristics:

“All these works relate in the strictest manner to Vedic Saṃhitās to which they belong and from which they draw Mantra-material in the form of Pratīkas. The Pratīkas, then, along with the rules (Vidhāna) and the desired results (Phala) constitute another characteristic feature of these works. The rules (Vidhāna) formulate Japa or Homa or both. Finally the sum and substance of these works is to lay down in respect of hymn or verse, the Kāmya ceremonies i.e. ceremonies which are intended to secure the attainment of various benefits and which are mostly of a magical nature.”

17Following the characteristics of the Vidhāna literature, we can assume with J. Gonda that the Kauśikasūtra itself can be regarded as a kind of Vidhāna:

  • 9 See Gonda 1977: 612. See also Bhat 1987: 16.

“In the several sections [...] the author has arranged his topics almost entirely in accordance with the order in which they occur in the Śaunakīya-Saṃhitā. As a whole it is comparable to the Vidhāna works of the Ṛg- and Sāmaveda.”9

18Just before the passage which sums up these characteristics of the Vidhāna texts, M. S. Bhat mentions the first characteristics as follows:

“All these works begin with the description of various penances, expiatory practices, etc. and the individual Vidhāna texts differ only in matters of detail or scope. Thus the description of the preparatory restraints or austerities in the Ṛg-vidh [i.e. Ṛgvidhāna] is somewhat more detailed than the Svidh [i.e. Sāmavidhānabrāhmaṇa] which contains few of them, but the former work has a number of expiatory practices which are totally absent in the latter. The Yvidh [i.e. Sukla-yajurvidhāna] stands on a different footing altogether for it merely alludes to penances without so much as elucidating them.” [Bhat 1987: 18]

19And in the footnote 48 he quotes a passage from the Yajurvidhāna, p. 1-2:

“When one performs a ritual rite, one (first) purifies one’s body by the penances like the tulāpuruṣa or cāndrāyaṇa, and for the sake of each rite one should (then) perform the expiations which form the part of the rite in question for twelve nights or six nights or three nights or one night.”

  • 10 Compare the prescription given in GobhGS 4.5.9-13 mentioned above.
  • 11 See Bodhāyana Gṛhasūtra 2.9.25; Bhāradvāja Gṛyasūtra 3.12 [79,1 -3]; Āpastamba Dharmasūtra (...)

20In my interpretation of this passage I add two adverbs, i.e. ‘first’ and ‘then’. The preparatory rites to which the priest is subjected have two stages. The first stage, which is, in my opinion, ordained to be observed when a person first intends to become an officiating priest of the rites prescribed in the Yajurvidhāna, makes a person eligible to act as a priest; and the second one is repeated every time when he concretely performs a certain rite.10 The first stage of preparatory rites seems to correspond to the preliminary rites to be undertaken by a householder who performs for the first time the vaiśvadeva ceremony which is prescribed in the tradition of the Taittirīya school.11

21The Sāmavidhānabrāhmaṇa describes three kṛcchras or penances. After having described them, the Sāmavidhānabrāhmaṇa 1.2.5 (30,8-9) states as follows:

  • 12 Note the similarity of the prescription with that of Dārilabhāṣya on KauśS 7.29 cited in note 5.

“One becomes pure, purified and qualified to perform a ritual rite after having performed (a penance) at first.”12

22According to this text the performer must first undertake a penance to become suitable to perform a rite.

23The opening part of the Ṛgvidhāna 1.13-65 deals with the points under consideration in greater detail. As regards these stanzas M. S. Bhat gives an interesting opinion: “The first passage (Sts. 27-56), which enumerates the various penances–mostly from the Manusmṛti–and teaches the manner of performing them, interrupts unnecessarily the continuity of the text, and has nothing whatsoever to do with the matter in hand except that one should purify before performing any rites.” (Bhat 1987: 150.) My analysis of these passages is slightly different as I show it in the following synopsis.

  • 13 Prājāpatya is a penance of 8 days’ fast. See Gampert 1939: 47-48 and kane 1973: 145-146.
  • 14 Cf. Taittirīya Saṃhitā See also Kaelber (1989: 53).

[1.13] “The preparatory restraint is efficacious at the beginning of rites (and) a twice born person should perform the prājāpatya13 at the commencement of self-reading (svādhyāya)” [In Bhat 1987: 283]
[1.14-26] “A preliminary act for twelve days in order to become eligible for the performance of the ritual practices prescribed in the Ṛgvidhāna, with the sthālīpāka or burnt offerings to various deities on the twelfth day.”
[1.27-45] Descriptions of various penances: a clearly later interpolation, almost identical with Manusmṛti 11.211-221.
[1.46-50] “Another interpolation (?). A preliminary rite to purify the constituent parts of the body like skin, blood, etc. by drinking only yāvaka or barley gruel for seven nights.”14
[1.51-53] (The preparatory restraint alluded in 1. 13ab?)
[51] “At the commencement of all rites, one should fast for three days or take food at night (nakta) for three days: then one should start the rite.” [Bhat 1987: 291]
[52] “At the commencement of a [religious] practice, a three-days penance [is laid down] for a regular ritual practician or one should fast later on, after performing a Sāṃnipātika [-rite].” [Bhat 1987: 291]
[53] (For the case of āyuṣya rites.)
[1.54-65] The method of the svādhyāya or self-reading of the Ṛgveda.

  • 15 Note that these prescriptions are very similar to those of GobhGS 4.5.9-13, as suggested by J. Gond (...)

24In my supposition it is necessary to confirm that at least Ṛgvidhāna 1.51-52 is not a later addition. If we can assign it to the original or an older version of this text, we can say that the Ṛgvidhāna also lays down a purifying rite for the priest before performing a certain ritual act.15 This purifying rite is not complicated; it is just a fast of three days. In my opinion, then, the case of the Yajurvidhāna becomes very similar to that of the Ṛgvidhāna: the Yajurvidhāna imposes penances for twelve nights or six nights or three nights or one night and the Ṛgvidhāna (1.52ab) prescribes a fast of three days or taking food only at night for three days. Anyway, it is clear that the Ṛgvidhāna lays down several sets of the preliminary acts to be undertaken by the officiating priest before performing the rites which are prescribed in the following parts.

25In the form of the general rule Bodhāyana Gṛhya Paribhāṣā 1.3.7 prescribes also the so-called dīkṣā for the officiating priest as follows:

“One who will perform the Vedic ritual rites, at first bathes at the ford in a river, comes up from there, wears new clothes, sips water,
purifies oneself with twenty-one bunches of darbha grass and purifies the person for whom he performs the rite.”

26In the soma sacrifice this purification with twenty-one bunches of darbha grass is performed only for the sake of the yajamāna (see Caland & Henry 1906: 13), but this text prescribes the same purification also for the priest.


27In the description of the dhruvāśvakalpa, a Vedic precedent for the post-Vedic pacificatory rite of a horse (aśvaśānti) (see Einoo 1999: 64), the MānGS 2.6.3 describes the necessary conditions of the priest as follows:

“The officiating priest should have no (physical or mental) defect (s), and (should be) cleansed and pure (in mind) and he should wear a new garment.”
[transl. Dresden 1941: 129]

28Here it is explicitly said that the priest must fulfill certain conditions in order to act as a priest.

29There are further examples which seem to suggest that the performer must undertake some preliminary acts before the performance of certain ritual acts. I will mention here three examples:

Aitareya Āraṇyaka 3.2.4 [134,4-6]:
“If he sees any of these, he should fast and cook a pot of milk, and offer it, reciting a verse of the Rātrī hymn (RV, X, 127, 16) to each oblation, and having fed the Brahmins with other food, himself eat the oblation.”
[transl. Keith [1909] 1969: 254]
Sāṅkhāyana Gṛhyasūtra 5.10.1-2:
“1. If the bees make honey in his house, 2. Let him fast and sacrifice a hundred and eight pieces of Udumbara wood, which are besmeared with curds, honey, and ghee, with the two (verses), ‘No (harm) to us in our offspring’ (Rīg-Veda I, 114, 8.9).” [transl. Oldenberg 1886: 139]
Apastamba Gṛhyasūtra 7.18.1,3:
“1. When a boy is attacked by the dog-demon (i.e. epilepsy), (the father or another performer of the ceremony), having devoted himself to austerities (such as fasting), covers him with a net. [...] 3. Over a boy who suffers from the ‘Śaṅkha’ disease, (the father, & c.) having devoted himself to austeries, [...]” [transl. Oldenberg 1892: 286-287]

  • 16 For fasting as tapas, see Kaelber (1989: 57-58).
  • 17 For further examples, see Gonda (1980: 254, 321).

30As Oldenberg interprets, if the word tapas in effect means fasting,16 the above three examples impose the fasting upon the performer as a preliminary act.17


31Thus, from all the examples given above, we can say that there are a number of cases in which the officiating priest or the performer of certain ritual practices should subject himself to certain preliminary acts which make him qualified to perform them. On the whole, these ritual practices can be classified into the kāmya rites in the widest sense of the word. The kāmya rites, then, include also such rites as are usually called imprecation (abhicāra), pacification (śānti), expiation (prāyaścitta), etc. But in these cases it is, in fact, not always certain who is the performer. In the course of my discussion I use both the words: the officiating priest and performer. In the case of the kāmyeṣṭis for which BaudhŚS 26.5 gives the general rules, it is certain that these kāmyeṣṭis involve the adhvaryu and hotṛ as the officiating priests in so far as they are performed according to the procedure of the New and Full Moon sacrifice, darśapūrṇamāsa. For this reason, W. Caland supposed as the performers of the preparatory acts “the yajamāna and the adhvaryu and the other priests as well”.

  • 18 For the relevant passages in the Gṛhyasūtras, see Tsuji (1968: 178).

32But for other cases which are given in the non-Śrauta ritual texts, I am not sure whether there is a difference between the officiating priest and the client. KauśS 7.29 prescribes: “śucinā karmaprayogaḥ / Being pure, one should perform ritual.” Who is the person who should be śuci (pure) for the performance of the rites? Is he the officiating priest who acts for a client or is he the very person who performs the various rites for the sake of himself? In the Gṛhyasūtras there are a group of rites for pacifying the evil omen, the so-called adbhutaśāntis. The last two examples given above belong to this group.18 When it is prescribed that one should perform a certain rite according to the procedure of the simple domestic rites described in the Gṛhyasūtras (pākayajña) for pacifying the evil omen (adbhutas), it is natural to hire an officiating priest. But when a recitation of certain verses or a simple butter-offering to the accompaniment of certain verses is ordained as the adbhutaśānti, the performer can be the very person who is inflicted by an evil omen. In these cases the performer and the client are the same person.

33This would also apply to a number of simple rites prescribed in the Kauśikasūtra and the Vidhāna texts. Or in other words, apart from the complicated Śrauta rituals which clearly demand the participation of the priests and some Gṛhya rites seemingly to be performed by the priest, there are a number of simple rites which are performed by the very person himself without the help of any priest.

  • 19 See Hillebrandt ( [1897] 1981: 154) and Keith ( [1925] 1970: 349).

34The dīkṣā and the upavasatha are prescribed for the yajamāna and his wife, namely for the persons who are the beneficiaries of the desirable effects resulting from the performance of a ritual rite. In other words the beneficiary of a rite must subject himself to certain kinds of preparatory purification. This is the principal rule of the preliminary consecration. According to this rule the sattrins, participants of the Sattra (a soma sacrifice which lasts more than twelve days and in which all the participants are the sacrificers), must consecrate themselves by a dīkṣā.19 The only exception may be the performer of the agnihotra (see Bodewitz 1976: 124). Now, if the performer himself is the beneficiary, then the performer should consecrate himself in any way. As I have shown by giving several examples, in the course of time from the period of the Gṛhyasūtras the simple forms of the rituals increased which were performed by a person for his own sake. For this reason we find in the later Vedic and post-Vedic ritual texts passages that seem to suggest that the so-called dīkṣā was to be performed by the performer.

35In this connection the statement of Baudhāyana Dharmasūtra is relevant:

“He who sprinkles himself (with water) at the beginning of any sacred rite, –before the time of the twilight devotions, –while reciting that same collection of purificatory (texts) (cf. BaudhDhS, becomes pure.” [transl. Bühler 1882: 246]

36This text prescribes that one should be prayata before beginning any acts. As J. Gonda pointed out in his article “Prayata” the word prayata means in the ritual sphere “(ritually) pure”. According to him, this word “is one of those words which without occuring in the Veda express an element of the religious life of the Indians of the epic, classical and post-classical periods” (Gonda 1963: 51). The importance attached to the word prayata in post-Vedic India also seems to have some relation with the change of the concept of the dīkṣā in the post-Vedic rituals.

37s. e.



Sanskrit Texts

AVPŚ: Atharvaveda Pariśiṣṭa
The Parśiiṣṭas of the Atharvaveda, Ed. by G. M. Bolling & J. von Negelein. Parts I & II. Leipzig, Otto Harrassowitz, 1909 and 1910.

BaudhŚS: Baudhāyana Śrautasūtra
The Baudhāyana Śrauta Sūtra belonging to the Taittirīya Saṃhitā. Calcutta, The Asiatic Society of Bengal, 3 vols., 1904-1924.

GobhGS: Gobhila Gṛhyasūtra
Gobhilagṛhyasūtram with Bhaṭṭanārayaṇa s Commentary. Critically edited from original Manuscripts with Notes and Indices by Chintamani Bhattacharya (1936), Metropolitan Printing and Publishing House. (Repr. New Delhi: Munshiram Manoharlal, 1982.)

KauśS: Kauśikasūtra
The Kauśika Sutra of Atharva Veda. With extracts from the Commentaries of Dārila and Keśava. Ed. by M. Bloomfield. (Repr. Delhi: Motilal Banarsidass, 1972.)

MānGS: Mānava Gṛhyasūtra
Mānavagṛhyasūtra of the Maitrāyaṇīyā Śākhā. With the Commentary ofAṣṭavakra, Edited with an Introduction, Indexes, etc. by Ramakrishna Harshaji Sastri. New Delhi, Panini, 1980.

Ṛgvidhānam, [Edidit cum Praefatione]. Ed. by Rudolf Meyer. Berlin, Ferd. Dümmlers Verlagsbuchhandlung, 1878. (See also Bhat 1987.)

Sāmavidhāna Brāhmaṇa with Vedārthaprakāśa of Sāy ana and Padārthamātrāvṛtti of Bharatasvāmin. Critically edited by B. R. Sharma. Tirupati, Kendriya Sanskrit Vidyapeetha, 1964.

Suklayajurvidhanasutra of Katyana with the Commentary entitled Yajurmanjari, Ed. by Annasastri Vare. Bombay, 1943.

Published Studies

Bhat, M. S. (1987), Vedic Tantrism. Delhi, Motilal Banarsidass.

Bloomfield, M. (1899), The Atharvaveda and the Gopatha Brāhmaṇa. Strassburg, Verlag von Karl J. Trübner.

Bodewitz, H. W. (1976), The Daily Evening and Morning Offering (Agnihotra) According to the Brāhmaṇas. Leiden, E. J. Brill.

Bühler, G. (1882), Sacred Laws ofthe Aryas, Part II: Vāsishtha and Baudhāyana. Oxford, Oxford University Press.

Caland, W. ( [1900] 1967), Altindisches Zauberritual: Probe einer Ubersetzung der wichtigsten Theile des Kauśika Sutra. Wiesbaden, Dr. Martin Sändig oHG.

Caland, W. ( [1908] 1968), Altindische Zauberei, Darstellung der altindischen “Wunschopfer”. Wiesbaden, Dr. Martin Sändig oHG.

Caland, W & Henry, V (1906), L’agniṣṭoma. Description complète de la forme de soma dans le culte védique, Tome premier. Paris, E. Leroux.

Colas, G. (1999), "The reworking of ‘Vedic’ paradigms in Medieval liturgies”, in J. Assayag, ed., The Resources of History, Tradition, Narration and Nation in South Asia. Paris/Pondichéry, École francaise d’Extrême-Orient, Institut francais de Pondichéry, pp. 41-50.

Dresden, M. J. (1941), Mānavagṛhyasūtra. Groningen/Batavia, J. B. Wolters’Uitgevers-Maatschappij.

Einoo, Sh. (1993), “Changes in Hindu ritual: with a focus on the morning service”, in Y. Nagano & Y. Ikari, eds., From Vedic Altar to Village Shrine. Osaka, National Museum of Ethnology, pp. 197-237.

Einoo, Sh. (1999), “The autumn Goddess festival [as] described in the Purāṇas”, in M. Tanaka & M. Tachikawa, eds., Living with Śakti: Gender, Sexuality and Religion in South Asia. Osaka, National Museum of Ethnology, pp. 33-70.

Gampert, W. (1939), Die Sühnezeremonien in der altindischen Rechtsliteratur. Prag, Orientalisches Institut.

Gonda, J. (1963), “Prayata”, Bhāratīyā Vidyā, 20-21: Munshi Indological Felicitation Volume, pp. 45-51.

Gonda, J. (1965), Change and Continuity in Indian Religion. London/The Hague/Paris, Mouton & Co.

Gonda, J. (1977), The Ritual Sūtras. Wiesbaden, Otto Harrassowitz.

Gonda, J. (1980), Vedic Ritual: The Non-solemn Rites. Leiden/Köln, E. J. Brill.

Henry, V. ( [1904] 1988), La Magie dans l’Inde antique. Paris, Librairie d’Amérique et d’Orient.

Hillebrandt, A. (1879 [1977]), Das altindische Neu- und Vollmondsopfer in seiner einfachsten Form. Graz, Akademische Druck- u. Verlagsanstalt.

Hillebrandt, A. (1897 [1981]), Ritual-Litteratur, Vedische Opfer und Zauber. Graz, Akademische Druck- u. Verlagsanstalt.

Kaelber, W. O. (1989), Tapta Mārga: Asceticism and Initiation in Vedic India. Albany, State University of New York Press.

Kane, P. V ( [1941] 1973), History of Dharmaśāstra: Ancient and Mediaeval Religious and Civil Law, Vol. IV Poona, Bhandarkar Oriental Resaearch.

Keith, A. B. (1909 [1969]), The Aitareya Āraṇyaka. Oxford, Clarendon Press.

Keith, A. B. (1925 [1970]), The Religion and Philosophy of the Veda and Upanishads, The second half. Delhi, Motilal Banarsidass.

Modak, B. R. (1993), The Ancillary Literature of the Atharva-Veda. New Delhi, Rashriya Veda Vidya Pratishthan.

Oldenberg, H. (1886-), The Gṛihya-Sūtras, Rules of Vedic Domestic Ceremonies. Oxford, OUP. Part I, 1886; Part II, 1892.

Rustagi, U. (1981), Darśapūrṇamāsa (A Comparative Ritualistic Study). Delhi/Varanasi, Bharatiya Vidya Prakashan.

Shende, N. J. (1985), The Religion and Philosophy of the Atharvaveda. Poona, Bhandarkar Oriental Research Institute.

Takashima, J. (1992), “Dīkṣā in the Tantrāloka”, The Memoirs of the Institute of Oriental Culture. Tokyo, University of Tokyo, pp. 45-84.

Tsuji, N. (1968), “On the formation of the Adbhuta-Brāhmaṇa”, Annals of the Bhandarkar Oriental Research Institute, 48-49, pp. 173-178.


1 The procedure of the dīkṣā is described, for example, by Caland & Henry 1906: 11-22. The Bṛhatsaṃhitā is a technical text on divination written by Varahamihira in the sixth century A. D.

2 For these regulations, see, e.g. Hillebrandt [1879] 1977: ix, and Rustagi 1981: 27-40.

3 For the equation of the upavasatha with the dikṣā, see Kaelber 1989: 51-52.

4 For the classification of ritual rites described in the first half of the Kauśikasūtra, see Bloomfield (1899: 57-76).

5 See also Dārilabhāṣya on KauśS 8.1: kartṛkārayitror ayam upadeśaḥ [...] prayogārthaṃ ca kṛcchraṃ prathamaṃ caritvā śuciḥ pūtaḥ karmaṇyo bhavati / ‘This is a prescription for the client and the officiating priest as well [...] for the performance of ritual one becomes pure, purified, and suitable for the rite after having performed the penance at first.’

6 See also KhādGS 4.1.1-6.

7 See also Gonda 1965: 323-324; Henry [1904] 1988: 223-224; Shende 1985: 154-155.

8 In the AVPŚ we find similar descriptions of the so-called dīkśā to be performed by the priest, for example, in the following places: AVPŚ 5.3.3, 8.1.3, 30b. 1.2, 66.1.5-2.1, and 70.1.11-2.3. See Modak 1993: 449.

9 See Gonda 1977: 612. See also Bhat 1987: 16.

10 Compare the prescription given in GobhGS 4.5.9-13 mentioned above.

11 See Bodhāyana Gṛhasūtra 2.9.25; Bhāradvāja Gṛyasūtra 3.12 [79,1 -3]; Āpastamba Dharmasūtra; Hiraṇyakeśi Dharmasūtra 2.1.43-45. See Einoo 1993: 222.

12 Note the similarity of the prescription with that of Dārilabhāṣya on KauśS 7.29 cited in note 5.

13 Prājāpatya is a penance of 8 days’ fast. See Gampert 1939: 47-48 and kane 1973: 145-146.

14 Cf. Taittirīya Saṃhitā See also Kaelber (1989: 53).

15 Note that these prescriptions are very similar to those of GobhGS 4.5.9-13, as suggested by J. Gonda in his translation of this passage.

16 For fasting as tapas, see Kaelber (1989: 57-58).

17 For further examples, see Gonda (1980: 254, 321).

18 For the relevant passages in the Gṛhyasūtras, see Tsuji (1968: 178).

19 See Hillebrandt ( [1897] 1981: 154) and Keith ( [1925] 1970: 349).


Professor at the Institute of Oriental Culture, University of Tokyo Hongo 7-3-1, Bunkyoku, Tokyo 113-0033, Japan, his research field includes the study of Vedic and post-Vedic rituals.
– 1988 Die Cāturmāsya oder die altindischen Tertialopfer : Dargestellt nach den Vorschriften der Brāhmaṇas und Srautasūtras, Tokyo, Institute for the Study of Languages and Cultures of Asia and Africa, Tokyo University of Foreign Studies (« Monumenta Serindica » 18).
– 2002 « Two ritual topics in the yājnavalkya smṛti : the tīrthas in the hand and prāṇāyāma », Journal of the Japanese Association for South Asian Studies, 14, pp. 20-39.
– 2004 « Notes on the vṛṣotsarga », in A. Griffiths & J.E.M. Houben, eds., The Vedas : Texts, Languages & Ritual. [Proceedings of the Third International Vedic Workshop, Leiden 2002], Groningen, Egbert Forsten (« Groningen Oriental Studies » XX), pp. 35-48.

© Éditions de l’École des hautes études en sciences sociales, 2006

Conditions d’utilisation :

Cette publication numérique est issue d’un traitement automatique par reconnaissance optique de caractères.
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search