STUDIEN ZUR INDOLOGIE UND IRANISTIK herausgegeben von Georg Buddruss, Oskar von Hiniiber Hanns-Peter Schmidt, Albrecht Wezler und Michael Witzel Heft 13/14 Dr. Inge Wezler Verlag fir Orientalistische Fachpublikationen Reinbek 1987. 12 24Die STUDIEN ZUR INDOLOGIE UND IRANISTIK (StI) erscheinen jahrlich in minde- stens einem Heft. Sie werden herausgegeben von Prof. Dr. Georg Buddruss (Neuindisch), Seminar fiir Indologie, Postfach 3980, D-6500 Mainz; Prof. Dr. Oskar von Hintiber (Mit- telindisch), Orientalisches Seminar, Abt. Indologie, Universitat Freiburg, Humboldstr. 5, D-7800 Freiburg; Prof. Dr. Hans-Peter Schmidt (Iranistik), University of California, Near Eastern Languages and Cultures, Los Angeles, Cal. 90024, USA; Prof. Dr. Albrecht Wezler, Bernhard-Ihnen-Str. 18, D-2057 Reinbek; Prof. Dr. Michael Witzel, Havard Uni- versity, Dept. of Sanskrit and Indian Studies, Grays Hall 1, Cambridge, Mass. 02138, USA. Die StI sind zu beziehen durch: Dr. Inge Wezler Verlag fiir Orientalistische Fachpublikationen Bernhard-Ihnen-StraBe 18 D-2057 Reinbek Telefon 040 / 7226731 ISBN 0341-4191 Alle Rechte vorbehalten Ohne ausdriickliche Genehmigung des Verlages ist es nicht gestattet, Foto Met O48 sinvelne Teile daraus nachzudrucken oder auf lotomechanischem Wege (Fotokopie, Mikrokopie usw.) zu vervielfailtigen. © by Dr. Inge Wezler Verlag fiir Orientalistische Fachpublikationen, Reinbek Herstellung: Proff GmbH, Starnberg, Printed in Germany363 Michael Witzel in The Case of the Shattered Head The dialogue of Yajfavalkya and a number of opponents in the Brhaddran- yaka Upanigad (Kapva version 3.1-9) is one of the more famous pieces of Upanisad literature. It is here that he, exceptionally, is questioned two times by a woman, Garg? (3.6 and 3.8). She asks Yajfiavalkya (BAU 3.6.1): "As ‘all this world’ (the universe) is woven into the waters into what ea then are the waters woven?" (yad idem sarvam otan ca protan ca, kasmin nu khalv dpa otds ca protds 2a?) Yajfiavalkya answers: ‘into the winds, the Antariksa-world, the Gandhar- va-, sun-, moon-, star-, gods-, Indra~, Prajapati-world,' and finally: ‘the Brahma-world.' When she insists and asks into which the Brahma worlds are woven, he warns her: "'Gargi, do not overask ('ask beyond’) so that your head does not burst (‘fly apart')! Really, you ask beyond the deity which is not to be asked beyond. Gargi, do not overask.' Then Gargi Vacaknavi fell silent." (Gargt matiprakgth! ma te mirdhd vyapaptat! anatipragnyam vai devatdm atippechasi. Gargi, matiprdkgir! iti. - tato Gargt Vdeaknavy uparard~ ma This paper was first presented as a lecture at the Centre des Etudes Jaina et Bouddhists, Paris, in December 1983. I take this opportunity to thank Prof. C. Caillat for her kind invitation to this and to other lectures at Paris. - Part of the legend treated in this paper has recently been discussed by I. FiSer, Acta Orientalia, 1984, p. 56-87: "Yajfiavalkya in the Sruti tradition of the Veda." Most probably the waters surrounding the earth and those flowing down from the night sky: Sarasvati/Sindhu/Samudra, see Bull. Etud.Ind., 2, 1984, p. 213 saq.364 Michael Witzel This passage usually is mistranslated: mirdha vi.pat does not mean "the head flies off" but, obviously, vi.pat means "to fly apart, to burst".? What is intended can be seen is such passages as JB 2.393 § 164 or TS 7.6.5.2 where a leather pipe used to store water is described as bursting.3 Secondly, the verb form vyapaptat is an aorist which cannot be used in conjunction with the prohibitive particle md. Even in late Vedic, ma always is used with the injunctive. One would, therefore expect md... 4 *vipaptad. We have to return to this problem later. At first sight, one may be lead to understand the passage quoted above, in terms of Western languages, as "to break one's head", an expression denoting "thinking deeply" about a difficult problem.> It remains unclear, what this "flying apart" or "bursting" of one's head means, and how it could occur. I will here try to find out the exact meaning of this expres- sion and also its wider background. One thing seems certain, however: The bursting of one's head seems to be the effect of or, perhaps, the punishment of asking too many questions, the one directed ‘beyond the brahman', towards the true nature of brahnan. However, this actually is what the Upanigads are about, and yet Yajfavalkya warns Gargi not to pursue her questions.° 5+ I. Fi¥er, AO 45, 1984, p. 79, correctly translates "burst asunder" and criticises the older translations; with regard to the idiom, however, he simply states: "often used on similar occasions in the Upanisads". JB 2.393 § 164 (cf. PB transl. p. 95) yo vat pirga: duapati, vi vd vai tat patati. pra vd Siryate; TS 7.5.6.2 ydthd dytir upanaddho vipdtaty, evdn ... See K. Hoffmann, Inj. p. 98. Only in some late Up.s, in the Epic and in similar texts there is confusion with ma; see ann. 81; cf. RV 10.95.15 md prd paptah "do not throw yourself down (into a cleft)", in ann. 95. This indeed was our first idea, when A. Wezler and I talked about this, in 1970 or 1971. Subsequently, A. Wezler gave a talk (so far unpublished) on the history of discussion, at Tubingen University, of which he sent me a copy some time later which I recently, after ay move to America, ‘rediscovered’. The present article, however, Was already written in outline in Dec. 1983 when I presented it as @ lecture at Paris, at the Centre de Philologie Bouddh.-Jaina. This paper has now been altered and added to slightly. ~ It is difficult for me to estimate now how far I then had been influenced by A. Kez- ler's lecture. We both have come to the result that atiprch means "to overask (in a discussion)". This is a technical term meaning "to ask beyond the limits of one's own) knowledge". hae Upanisad, ZDMG 136, p. 79 sqq. understands it as: "bewirkende cht".The Case of the Shattered Head 365 2, If one compares this situation with others in the rest of the present Upanisad dialogue, it is notable that the questions are placed in the context of an open challenge. In the present story, Janaka has begun a sacrifice and has invited nany priests, among whom he wanted to find out the most learned one. Con- sequently, he set out a prize of one thousand cows with gold pieces bound to their horns. Of all the Kuru-Paficdla Brahmins who had come from far away to Videha in order to take part in the sacrifice or the contest Yajfa- valkya was so sure of himself that he dared to appropriate the prize imme- diately, even without entering into a dialogue. It is at this moment that the other Brahmins angrily challenge him. Interestingly, the challengers seen to be the best among the various groups of Brahmins, (and both Yaj valkya's and their personalities require further study).? In order to understand better and to define the conditions of the threat levelled at Yajfiavalkya, a closer investigation must be made of the other Upanisadic passages which deal with the topic of the head of a discussant that is threatened to fly apart.® As in the BAU passage, insufficient knowledge is the cause of the threat. TB 3.10.9.5 (< *KathB): A student of Atyatiha Arunpi was sent to Plaksa Dayyampati to question him on the Savitra fire. When Plaksa finally declares that the Savitra is based on the prdna, the student says: "My teacher told me I should not ask beyond the prdza" Plaksa D. said: "If you had asked beyond the Prana, your head would have burst apart. And I shall (always) be better than your teacher, who has challenged me to discuss on the Savitra (fire)." See below, ann. 77; and cf. L. Renou, Les relations du Satapathabrah- mana avec le Brhaddranyakopanisad et la personalité de Yajfavalkya; Fier regards the whole episode as artificial but overlooks that such contests took place quite commonly both in the early Vedic period (RV, AV) as well as in the later Vedic literature (SB etc.); see for example the story about Uddalaka Aruni among the Northerners at SB 11.4.1.9, - where there cannot be any question of "legend forming". 8. For the art of discussion in general at the time of the Upanigads, see Ruben, Uber die Debatten in den alten Upanisads, ZDMG 83, 1928 p. 238-255.366 Michael Witzel (.... ma sma pranam dttprecha, ttt mdedryo ‘'bravid tti hovdea brakma- cané. ed hovica plaksé ddyydmpatih: ydd vai, brahmacdrin, prandm atya- prakgyah (*atydprakgyah) mirdhd te vydpatigyat (*vydpatigyat). ahén utdedryde Sréyin bhavigydnt, yo mi edvitré sandvadigtéti.)? Here, the discussion is clearly described as a challenge, even though it is acted out by a Veda teacher and the pupil of another one. An obvious challenge is also met with at $B 11.4.1.9: Udda@laka Aruni, driving around’ the land of the Northerners, was challenged by their Brahmins. He offered a gold coin which was the prize 'formerly', in order to "call out the timid for discussion", Udda@laka, after a secret discussion with one of the North- ern Brahmins, Svaiddyana, had to concede insufficient knowledge and gave him the coin. Svaidayana however, concealed it, apparently as to keep his gain secret, and told his friends, on his return from the discussion: "Even if one was a Brahmin, a relative of a Brahmin, the head of the one who challenges him (Uddalaka) to a discussion (upa.valh) would fly apart! (ydthd. brahnd brahmaputrd mirdhkeya vtpated, yd enam upavdlhétt). Uddalaka then came back to Svaiddyana, with firewood in his hands, and asked him for an explanation."!! 9. Correction of the text acc. to Dumont (Proc. Am. Phil. Soc. 95, No. 6, 1951, p. 639). See below for a further treatment of these incorrect forms; the present case is perhaps due to faulty Taitt. transmission of an original Katha text, a part of the lost KathB.; Katha texts generally are better transmitted than late Taitt. texts (TB/TA), see V. Schroeder/Bihler, SB Wien 1898: Die Tiibinger Katha-Hss., appendix. lo. Note that one "drives" to a yajiia or discussion by chariot, as 2 challenge - certainly a reminiscence of an older time when one vould interfer, during a race or ritual by driving close/into others! "terri- tory"; cf. also the German saying "jemanden an den Karren fahren"); on dhav see Bodewitz in IIJ 16, and cf. below § 6.d. on Sudakgina Ksaimi). 11, Note the similarity of the formulation with that of Gargi and Yajfia~ valkya, see below § 2 (b): BAU 3.8.1. - $B 11.4.1 still speaks only about one gold coin that "formerly" was the prize, but normally in the Brahmanas and Upanisads, one gets about 100-1000 cows, sometimes also Seo horses, a mule driven chariot, and exceptionally, a wife, as the prize, see $B 11.3.1.4 (Janaka and Yajfiavalkya: 100 cows); ChU 4.2.1 (600 cows, a golden chain, a mule chariot, and successively more? including a wife, the daughter of Janagruti; in discussion with Raikva); $B 11.4.3.20 (1000 cows: J. and Y.); BAU 4.3.33 (1000 cows? J. and Y.); 4.1 (ooo cows and a large bull, repeatedly, J. and Y.); JB 1.25 (1ooo cows and S00 horses: J. and several Brahmins). - cf- ann. 12, 16, and Renou, Les rélations (ann. 7); Hauschild, Mélanges Renou, p. 354 sq.The Case of the Shattered Head 367 Another version of this story occurs at GB 1.3.6 sqq. (cf. SB 11.4.1.1- . 4s in SB, Svaidayana pretends that Uddalaka was better than him only during a friendly disputation, under four eyes only (GB 1.3.8). (tak ha paprdccha: kim ega gautamasya putra iti, ega brahnd brahma~ putra ité hovdca. yad enam kaSeid upavadetota mim@iseta ha va mindhd vd asya vipatet. prand vainam jahyur iti ...)!2 ChU 1.10.9: US4sti Cakrayana, who was so poor he had to eat uechista food, went to a Raja's sacrifice and challenged the udgdtars; »prastotar, if you will sing the prastdva (introductory ‘praise’ of a sdnan) without knowing the divinity connected with the prastdva, your head will burst apart!" (and likewise to two more priests). When the king agreed to pay him as much as the others, they repeat his threat (1.11.4, etc.) and ask him for teaching on the matter, USasti always ends in saying: "If you had taken up the prastdva (etc.) without knowing it, your head would (have) burst asunder - a f te r you had been told so by me. (prastotar, yd devatd prastdvam anvayattd, tam ced advidvan prasto- gyast, mindhd te vipatigyatéti (1.10.8) ... tdn ced avidvin pratya~ havigyo, mirdhd te vyapatigyat tathoktasya mayeti, tathoktasya maye- #8.) (ait. 33 A discussion resulting in the threat of a shattered head apparently has to be announced formally; this is stressed here by doubling the state- ment. ChU 1.8.6: Three Brahmins hold a discussion on the udgétha. When one says that the edman is heaven, the other says: "Your edman is without foundation, Dalbhya. If someone would say now 'your head will burst apart,' your head would burst apart." - [The other one]: "Then let me ask it of you" ... [The same is repeated] (apratigthitam vai kila te, Ddlbhya, sdma, yas tv etarht bruydn: mirdhd te vipatigyatiti, mirdhd te vipated itt. hantdham etad bhaga~ vato veddntti) Here a friendly discussion, held without any threat, can be turned into a challenge immediately if an outsider (or, of course, one of the 12, See transl. gold coin is given as a prize even in the very late GB 1.3.6 (where the tale is excerpted from $B but also added to). Geldner-Pischel, Ved. Stud. IT, 185; note that only one 13. The passage looks rather like a satyakriyd, sce below ann.s 75 and 23.368 Michael Witzel participants who happens to know more) announces the insufficient knowledge of the other(s), saying "Your head will burst apart!" $B 11.5.3.13: Sauceya Pracinayogya came to Uddalaki Aruni for a dispu- tation on spiritual matters. They discuss the Agnihotra ... "Verily, I did not know that" said (Sauceya). Sauceya, thus instructed, said: "Here are the logs for fuel; I will become your pupil, sir." (Udddlaka) replied: "If you had not said so, your head would have burst apart." (ydd evdm ndvikgyo, mirdnd te vydpatigyat.) (He then taught Sauceya.) The GB version of this story (GB 1.3.14) differs slightly at the end: Sauceya Pracinayogya came to Udd&laka Aruni for a discussion on the Agni- hotra (see the transl. of Eggeling $B 11.5.3, Bodewitz, Agnihotra p. 153 sqq-). (Udd.) said: "Sir, as you have said, I approach you (sir) (for in- struction)." (Sauceya) said: "If you had not said so, your head would have burst apart. But now, I will tell it to you so that your head will not burst apart." (evam evaitad: bho bhagavan, yathd bhavin dhopaydmi tv eva bhavantan ity. evam cen ndvakgyo, miirdhd te vyapatisyad iti. hanta tu te tad vakgydmt, yatha te na vipatigyatiti.) As in the other cases mentioned so far, the discussants who do not know the whole truth, have to give up. They have to state this clearly, and they even can approach the winner in the debate as pupils, with fuel wood in their hands. (It is well known that it is one of the many duties of pupils to collect fire wood for the teacher.) The situation is similar at ChU 5.11-17: Five Brahmins discuss, in a friendly way, about the dtman and brahman and then go to Uddalaka Aruni to learn more. He fears not to know enough and joins them in going to Agvapati 4 Kaikeya, apparently a Ksatriya, to learn more.'4 They approach Agvapati as pupils, with fire-wood in their hands, but he accepts them without a 14. Notably it is Uddalaka who knows only the ‘lowliest' part, the feet; for this concept, cf. that of the Sidra iginatingfrom the feet of the Puruga RV 10.90, cf. DN 27.3. Note that Udd. feared not to be able to answer the brahmins at the outset, and had sent them on to Asvapati- This looks like a rhetorical device on part of the Samavedins of ChU to put the famous $B teacher Uddalaka in a lower position; cf. below § 7, on Vidagdha Sakala and Yajfavalkya, and cf. ann. 19.The Case of the Shattered Head 369 special initiation and asks one after the other about the déman. Their answers are deemed insufficient. One of them understands heaven as the dtman vaiévdnara. Agvapati answers: (5.12.2) "But this one (which you explained) is only the head of the dtman', he said. 'Your head would have flown apart if you had not come to me (for my explanation, - which follows at the end, ChU 5.18 sqq.)!" (mindhd tv ega dtmana iti hovdea. mirdhd te vyapatigyad, yan mam ndgamigya itt) It is interesting to note that the other four Brahmins, who are said to know only the sight, the breath, the body, the feet of the atman, are threatened individual], "This is only the sight (breath, body, feet) of the dtman. You would have become blind / your breath would have left you / your body would have dissolved / your bladder would have split / your feet would have withered - if you had not come to me". (... andho 'bhavigo / pr&yas ta udakramigyad / samdehas te vyastryad / bastis te vyabheteyad / padau te vydmldsyetam, yan map ndgamigya itt) This is, if I have seen correctly, the only case where insufficient knowledge has such diverse effects. Usually it is the head that flies apart 15 This kind of a discussion cannot be avoided if demanded by one of the people present at any occasio JUB 3.2.1 (cf. ChU 4.3.5-6): A brahmin (ChU: a brahmacdrin) begged food of Saunaka Kapeya and Abhipratdrin Kakaseni who were being served food. They did not pay attention to him, thinking "who is he?" He (the brah- min) sung (a verse) to them, "One god swallowed up four magnanimous ones; who is he that world's keeper? Him, o Kapeya, some do not know; (him), 0 Abhipratarin, settled down in many places." - Abh. said: "Stepping forward, answer this man. By y ou this man must be answered." 15. Cf. however, also the case of the Caraka and Yajfiavalkya at SB 3.8.2. 24-25: The Caraka threatens Y. that his breath would leave him if he did not know a particular detail of the ritual, see StII 8/9, p. 172, ann. 112.370 Michael Witzel (... tau ho 'pagagau: mahdtmanaé caturo deva ekah - kas sa - jagdro, bhuvanasya gopah; tam, kdpeya, na vigdnanty eke; 'bhipratdrin, bahuihé nivigtan iti. - sa hovdedbhipratariman vava prapadya pratibrihiti. tvayd vd ayam pratyucya itt .) Also, one cannot just pretend to know the answer; in that case, the competitors will not allow one to get away with it, see SB 14.6.1 = BAU 3.1, where Yajfavalkya simply orders the driving away of the cows, the prize set out by Janaka (cf. SBM 11.6.3, JB 2.76). The threat to Yajfiaval- kya is repeated at BAU 3.7.1. If one does not know the answer, the only honorable way out of the predicament is to become the pupil of the winner in the discussion, see above: TB 3.10.9.5; and cf. $B 11.5.3.13 (Uddalaka as pupil of Svaidayana), ChU 5.11.7 (five Brahmins, including Uddalake as pupils of Agvapati Kaikeya). After the extensive presentation of the examples dealing with the shattered head of a discussant, the evidence can be summed up: (a) The general context can now be described more precisely: two or more persons, almost exclusively men and mostly brahmins, challenge each other to answer certain questions of a ritual or spiritual nature; or, one man is challenged by a group of others. This may occur in a private or in public situation; often, it takes place in ritual context (see 16 below). (b) The discussion can be initiated in the form of an open competition. In this case, each of the participants is automatically challenged right from the beginning. Or it can develop between ritualists in the course of a friendly exchange of views, while “talking shop". In this case, 7 the challenge has to be stated expressively.! 16. For the terminology, cf. already F. B. J. Kuiper, Old Indian verbal contest, IIJ 4 (1960), reprinted in his book, Ancient Indian Cosmo- gony, Delhi 1983, pp. 151-215: vivdc, a-vivdkya day, upa.vath.; at SB 11.4.1.9 Uddalaka Arupi is challenged by the Northerners, for @ piece of gold and this is called a "former" custom, see ann. 11; for sam,vad. see above ann. 12; cf. above, upa.vad and cf. below, samuida, anu.gthd, ati.kr, see ann. 72; these and other terms of the brahmodya and of the general rules for a discussion have to be treated separate- ly, in more detail. 17. For other cases, occurring in sattras, see H. Falk, Bruderschaft und Wirfelspiel. Untersuchungen zur Entwicklungsgeschichte des vedischen Opfers, Freiburg 1986, p. 35 sqq.(c) (a) The Case of the Shattered Head 371 The actual challenge can be voiced quite forcefully; Garet, for ex- anple, does so in her second questioning of Yajfavalkya, at BAUK 3.8.1 sqq-: "I will ask this Yajfiavalkya two questions. If he can answer them, nobody of you will overcome him in a disputation. If he cannot answer them, his head will burst." - 3.8.2: "I will approach you (Yajfia- valkya) with two questions, like the son of a military leader (ugra) from Kagi or Videha approaches, having strung his unstrung bow and two arrows in his hand (ready) to pierce his rivals".!® One cannot avoid such a challenge, cf. also above, JUB 3.2.1. The challenge may be especially instigated by a prize, mostly gold, cf. for example, the tale of Uddalaki among the Madras, SB 11.4.1.1: "for in the time of our forefathers a prize used to be offered by chosen priests when driving about, for the sake of calling out the timid to a disputation." By the time of the disputation of Ya fiavalkya with his opponents, a famous king like Janaka seems to have increased this reward ten thousand times: According to BAU 3.1.1 at least, he presents a thousand cows, each with ten pddas of gold bound to their horns. Is this increase in value of the prize due to the relative wealth of Videha (which, according to $B 1.4.1.16, "now" is fit for agriculture, having been "sweetened" by the Brahmins with the help of sacrifices)? There may be, however, an element of exaggeration, as FiSer generally supposes for BAU, for example, when Janaka gives away himself and his : 1 whole country to Yajfavalkya, see BAU 4.4.23.'9 19. Figer, AO 45, p- 79 sq. regards this passage as an example of the Yajflavalkya legend being "already in full bloom" ... "The story con- tinues in this unusually pompous style". The truth of the matter is that Gargi quite appropriately expresses her challenge in warrior-like terms, which only fits the challenge to a discussion during which one may lose "face", - and more, — Cf. also such treatments of Yajfiavalkya as that of K. Werner, On the philosophy of Yajfiavalkya, Bhar. Vidya 11, 166-177. Or is it just édkhd propaganda, see ann. 88 (cf. ann. 14), and in connection with Sakalya and Yajnavalkya, § 7; - cf. already Deussen, transl. 60 Up.s, p- 427, on Janaka; he is a legendary king already at BAU 2.11 and KU 4.1; - cf. also Mahajamaka (and Mahakosala) in the Pali sources see Franke, K1.Schr. p. 379 (Janaka as a famous king of old Mithila). Nothing is known about his ancestors or position in history. One may, perhaps, compare Janagruti of ChU 4.1, the grandson of a janagruta. The short form of this name would be Janaka. Note that Janagruti. is described as "possessing Sraddaa" and as distributing wealth etc., - which would make him a worthy continuant of Janaka's line. Cf. ann. 48.372 (e) (f) (g) (h) Gi) Michael Witzel In the course of the discussion, participants who do not know the whole truth have to state this clearly, they must cease questioning, (see above, especially BAU 3.1 sqq-), and thus declare defeat, or they must become the pupil of the winner; cf. also 11.4.1.9 (Uddalaka and Svaida- yana) $B 11.5.3.13, cf. ChU 1.1o.11. The bursting of one's head is threatened as the effect of asking one question too many, that is to ask beyond the limits of one's oxn know ledge. It is not "wrong" to ask ultimate questions as such3 but to do this with insufficient knowledge results in the formal pronouncement of the threat by the opponent. It can and, usually is, repeated to warn the challenger, see above and cf. also: Pragna Up. 3.2: (Pippalada speaks): "You ask questions that lead too far (atipragna). You are the best Brahmin, therefore I will tell you ..." (tasmat sa hovicdtiprasndn prechast; brahmigtho ‘siti; tasmit te ‘han bravimi). If the opponent cannot answer, he may be warned again, and usually "keeps his peace" then, in order to avoid the consequence of losing his head (see above, especially BAU 3.1 sqq.). One is, however, not always required to announce one's secret solution of a problem in public; one may choose to reveal it to the opponent in private, as Ydjfiavalkya does at BAU 3.2.13 (when they talk about karma), or as Svaidayana does, when challenged by Uddalaka, at SB 11.4.1.3; 9. This merely underlines the fact that ritual and spiritual knowledge is secret, and normally, is to be taught only to special students, one's son, or to friends. On the other hand, one cannot just claim to be better than the rest. The Brahmins do not take this even from Yajfiavalkya (BAU 3.7.1 = SBM 14.6.7.4). He is threatened that his head would split if he should not know a particular secret, but would, in spite of this, actually assert, ‘I know it, I know it!' ("as everyone would say":) (tac cet tvap, Yajnavalkya, siltram avidvaris tap cdntaryaminan brahma- gavir udajase, mirdhd te vipatigyatiti; veda vad ahan, Gautama, tat sutran tam cdntarydminam iti. yo vd idan kascid (kds ca $B) brid veda vedett. yathd vettha, tathd bruhiti!) cf. also 1.11.4-The Case of the Shattered Head 373 Mere brazen assertion does not suffice; one must be able to prove one's knowledge. It will habe become obvious by now that the texts speak about a set of general rules of discussion, rules of challenge and defeat. One can enter stich a contest whether one indeed does know the ultimate answer or not, but one has to stop short of asking the "last question": — in other words, one is not allowed to ask (prech) beyond (ati) the limit of one's own knowledge. Of course, the one entering such a discussion is not in a position to foresee whether he might be the winner or the loser. The opponent, after all, may turn out - even if he is a Yajfiavalkya - not to be as learned as one rumours him to be ... In such a case, the stronger, victorious opponent will soon notice that the challenger cannot answer the “last question" as well as he himself would be able to do if asked for, ~ and therefore, he warns the opponent: "Do not overask, so that your head might not, burst 107° Conceding defeat in a discussion has, of course, the social effect of clearly stated and admitted superiority, of gaining and losing ‘face! among one's fellow brahmins and in the tribe at large. This is a 21 point to which we will have to return later. n the other hand, however, one gets the impression, incidentally, of an empty threa’ Yajfiavalkya (and others, like Svaidayana at $B 11.4.1) apparently do not wish to be (openly) asked about their own specific secret insights or speculations. When pressed for an answer, they just take the challenger apart and instruct him in secret.“~ 2. For Uddalaka as a loser see SB 11.4.1. 21. CE. below, RV passages on brahmodya, on wrong accusation and on poetry contests, cf. ann. 1o1; mote the expression "to make someone one's leader" in poetry. 22, Similarly Uddalaka at 11.4.1, or BAU 3.2.13: Yajfiavalkya and his opponent Artabhaga talk about karman.374 Michael Witzel Normally only well known - though technically complicated ~ questions are allowed (and even the section treated here at length, BAU 3.1-9, is full of them). Yet, there are a few passages of another nature, i.c. those that do not involve a brahmodya or ritual discussion. Yet still they indi- cate that danger to one not knowing the proper answer. In such cases, how- ever, the questioning concerns the proper procedure or ritual and its se- cret, esoteric meaning known only by brahmins initiated into the sacred lore and having undergone a long period of training in the Srauta ritual and the special consecration (dékea). One example has already been given above: lack of knowledge about the devatds connected with certain ways of singing the Samans results in having one's head split, in death (ChU 3.10.9-11). In the same way, $B 3.6.1.23 clearly threatens this outcome, if one - not knowing the consequences - eats in the Sadas shed of the Soma (Agnigstoma) ritual; the Sadas is reserved for the gods: "“(Therefore) the cart-shed belongs exclusively to the gods; hence neither food nor drink is taken therein; because it belongs exclusive- ly to the gods; and were anyone either to eat or drink therein, his head would, verily, fly apart." (tasmdd tdtra ndénanti nd bhakgayantt ntgkevalyan hy etdd devindm. sf yO ha tdtra6ntydd vd bhaksayed va mivdhd hdsyd vtpated) A similar case, though in a mythological setting is found at GB 2.1.2. This section deals with the prdSitra offering which was to be eaten by various gods. They all suffered. Bhaga lost his eyes, Pagan his teeth, etc. The gods transferred it to Afgiras. (This section dealing with Ahgiras is only found in GB, certainly because of the Atharva-Aigiras origin of the AV texts; cf. introd., ed. Gaastra, p. 34, for parallels in AB, KB, TS, $8.) Aigiras ate of it: “His head burst apart. The sacrifice fashioned him." (tasya Siro vyapatat. tam yajria evakalpayat.) During the performance of a ritual one can even put one's ritual know- ledge to a test: BAU(K) 1.3.24 = $B 14.4.1.26 24: “Therefore Brahmadatta, son of Cikitdna, said, when drinking the King (Soma), ‘The King (Soma) shall split the head of this one (= me), if Ayasya Migirasa sang (the udgitha) with other means; he (too) sang the (udgitha) only with speech and breath.'"The Case of the Shattered Head 375 (taddhdpi Brahmadattas Caikitdneyo rajanam bhaksayann uvacdyam tyasya raja mirdhanam vipdtayatdd , yad ito ‘ydsya Mngtraeo 'nyenodagdyad i. vded ca hy eva sa prdnena codagdyad itt.)?3 This is the only passage so far that actually names the one who exe- cutes the threat: it is a god, King Soma, who is, next to Agni, one of the two gods that are visibly present on the offering ground. The conditions for the ominous threat involving the splitting head can now be redefined: - insufficient knowledge on part of one of the contestants and lack of admission of this fact ~ perpetration of ‘forbidden! actions (based on lack of knowledge) ~ asking of ‘forbidden questions! (i.e. those beyond one's knowledge: anatiprasnya). An important factor is this one: if someone challenges the other con~ testant with the announcement that his head would split, should he act in such and such a way, it will actually happen. This looks like one of the well-known satyakriyd's (saccakiriyd), the truth (satya) that is unknown to the other party being the secret, esoteric knowledge about facts of ritual or mythology, ete. "4 3. Obviously, the apparently empty threat was quite effective with an Upa— nisad time audience, for usually nobody dares to question beyond the point which had just been declared not to be asked beyond (mitiprakgih) or to be anatipraSnya. The one, and it is only one person, who actually dared to do so, Vidagdha Sakalya, promptly died shortly after he lost in such a discussion - a case to which we will have to return. 23. ‘This one’ = 'me', here expressed by tya-, but note, Zimmer TJ 28, 1985, p. 192 sqq. — The passage again looks like a saccakirtya. 24. Note KB 27.1, the loth day of Dagaratra (ends the 12th month of a one year Sattra): "The loth day is that which is above in the sky; it is hot to be explained, as nobody knows this (the heaven). ‘Let me not explain ignorantly.' (ned avidvan vibravanitt).376 Michael Witzel How can this be understood? In a period, where the ultimate question is put in so many ways, albeit by solitary speculation and subsequent se- cret teaching to selected pupils, at a time when Brahmins usually got, for answering questions about the brahman, at least a gold piece, and when Yajfiavalkya receives 1000 cows with gold pieces bound to their horns, or is reported as having received even the whole of the Videha land fron King Janaka, - in this period merely as king a spiritual question beyond one's own knowledge remains not only forbidden but is inherently dangerous. Now, we are not told exactly how Vidagdha Sakalya died; so we must look for more incidental information concerning the threat of the shattered head viz. of splitting the head. One such circumstance could be: The "splitting headache" which not only men, as we all know, but even the gods can contract in a certain stress situation: VadhB 4.75 tells of a particular Rsi who performed a sacrifice behind the Garhapatya fire.”> "His (magical power of accumulated) heat went to Indra who was in the third heaven. Addressing (Agni) Jatavedas, he said: 'Jatavedas, ny head aches somehow (iva, to a certain, i.e. here:, to a great extent).'" (tasya hendram trtiyasydm divi safntam7 tapo jagama. sa jatavedasan Gmantryandna uvdea: Jatavedah, Siro me rujatéveti ... - no hi te tad evaitad rujattveti .)2© Here, it is the magical power of a Rsi which - similar to the well-known Mahabharata legends - induces even a god to feel a headache. Can it be that a similar power, that of an unspoken, unknown truth which is known, however, to someone else, e.g., to one's adversary in dispute or in sorcery, induces such a ‘splitting headache'??7 In sorcery, such an occur 25. vache 6.75: p. 190 sqq.; cf. Oertel, Transact. Conn. Acad. Sc. Wy, Pp. 180. 26. Cf. the similar expression at AB 6.24. virujya; hrdayo me tapati. : vagrena ... véeah kitena «+ 27. That inspiration (by Agni) could be described in similar terms 35 evident at RV 6.9.6: vt me kdrnd patayato, vt cdkgur, vidan Jystir hfdaya dhitam yat, vt me mdnaa carati dindadhih. kim svid vaksyatl, ktm u nit manigye.The Case of the Shattered Head 377 rence is not unheard of: a particular rite can produce — strange as it may sound to us ~ any number of symptoms - if only the person who is to be affected actually knows of the spell.2° In Africa, for example, spells of this kind have been reported to result in the death of a person within hours: this is due to heart failure, which of course can only occur after the threatened person had got to know of the spell.29 The stereotype warning of the Brahmana and Upanigad texts of this effect could point to a similar belief: If one dares to ask one's head will split apart - at first, t ‘too far’, perhaps, by severe headache when ying to think of an answer, and subsequently, after the inevitable death caused by heart failure, - it will split in reality, i.e., when the body is burnt during cremation: Then the bones of the head will crack along the fissure lines of the fontanels.°° 4. Such an explanation would seem rationalistic; it takes into account what we know about sorcery and the belief in magic, yet it does not - I am afraid - solve our problem. The one case where actual death is reported i.e. the one of Vidagdha Sakalya, remains mysterious. While the mere threat of death is usually sufficient in the Upanisads to deter anyone from asking too far, here the threat alone does not suf- fice, Sakalya, though properly warned by Yajfiavalkya, asks the question "beyond the deity', the one "not to be overasked" (anatipragnyam), and Promptly dies. The passage is worth a closer look. It occurs in 4 versions out of (theoretically) 5, including the so far unedited, accented version of BAUK. Actually available in print are: SBn 11.6,3.11, JR 2.77 (cf. Oer- tel, JAQS 15, pp. 238 sqq-, Lokesh Chandra, ed. JB IT 1-80, Nagpur 1950), SBM 14.6.9.28, BAUK 3.9.26: 28. If a sorcerer is really clever, he can 'send it back', see Bloomfield, The Atharvaveda and the Gopatha Brahmana, p. 68: kytydpratiharana, 29. Observe that this concept is actually known to modern "sorcerers", the Atharvavedins of Orissa: They told me (Summer of 1983), quite acciden- tally, that "nowadays our magic does not work: people do not believe in it any more." 30. Note the medieval and modern custom of smashing Yogis' heads before burial in the ground; they are not cremated, see Sprockhoff, Sannyasa, 62, cf. ann. 107.378 Michael Witzel $B 11 (Y8jni.) ed hovdea: anatiprasényam ma devdtdm dtyaprakgth JB (vajfi.) sa hovdednatiprasnydm vai ma devatdn atyaprakgth $B 14 (Yajfi.) tén tvaupanigadém purugan prechami. tam cén me nd vivakgydet, mirdhd te vtpatisyatitt. BAUK (Yajfi.) tam tvaupantgadam purugam prechdmt. tam cen me na vivakgyasi, mindhd te vipatigyatiti. $B 11 purdtithydi marigyast. nd té 'sthint cand gphdn prapsyanti: JB puraitdvatithyai martdst. na te 6arirdni cana grhan prapsyantiti. $B 14, BAUK - i SB 11 JB - $B 14 tan ha Sakalyd nd mene. tdsya ha mirdhd vtpapdta. BRUK tay ha na mene Sékalyah. tasya ha mirdhd vipapaté- SB IL sd ha tdthaiva mamara, JB tad dha tathaivdsa, sa ha tathaiva mamdra. SB 14, BAUK - - $B 11 tdeya hdpy anydm ményaminah parimogind 'sthiny dpajahrus. JB tasya hépahérino 'mantarena Sarindny apajahrur, $B 14 tdsya hdpy anydn manyamindh parimogind 'ethiny dpajahruh. BAUK — ~pd hdoya parimogino 'sthiny apagahrur, $B 11 tdsmin ndopavadi sydd. JB anyan manyamdndh. tasméd u ha nopavaded. $B 14 (dtha ha Yajravalkya uvaca: BAUK anyan manyamdnah. (atha hovaca: $B 11 utd hy évayvtt pdro bhavati. JB apt hy evamvit paro bhavateti3! $B 14 brdhmand bhagavanto, yb vah kandyeta, ea md prechatu BAUK brdhmana bhagavanto, yo vah kdmayeta, sa ma prechatu ... 31, Note the curious i¢i at the end of the JB version. Is this an indi- are that the passage had been taken over from another text, proba~ y SB?The Case of the Shattered Head 379 Translation: $B11/JB He (Yajfiavalkya) said: "You have asked me beyond the deity which is not to be questioned beyond." $B14/BAUK Yajfiavalkya said: "I ask you about the secret puruga. If you will not explain it to me, your head will burst apart. $B11/JB You will die before the day so-and-so. - Not even your bones will reach your home." $B 14, BAUK - SBit Sakalya did not think so (did not know it.) His head burst apart. BB - $B14/BAUK Sakalya did not think so (did not know it). His head burst apart. Bu - He died just that way. IB That was just so. He died just that way. $B14,BAt a $B11 And robbers took away the bones, thinking them to be something else. IB And robbers immediately / afterwards took away his bones, think- ing them to be something else. $B14/BAUK And robbers took away the bones, thinking them to be something else. $B11/JB_ Therefore one should not be someone who speaks badly about (another person, one should not insult). SBig Then YAjiiavalkya said: BAUK Then he said: $B141JB For someone who knows thus (about the puruga), becomes superior." SB ig “Brahmins, Sirs, who of you should wish, he may question me ..." BAUK "Brahmins, Sirs, who of you should wish, he may questions me ..."380 Michael Witzel The Upanisad (BAUK, $B 14) version is clearly later;** its end (3.9, 28) consists of memorial verses! The older Brahmana version (SB 11 / JB) is interesting as it only speaks of dying very generally; this is to take place on a certain day. Both versions, however, agree in reporting an actu- al death, - having taken place, perhaps, as robbers are mentioned, sone- where in the wildernis (aranya).33 They happen to come across Sékalya's bones which obviously were already picked clean by animals. Why the robbers carry them off remains mysterious. - Why should anyone carry off the bones of a dead person which one (accidentally?) comes across? - Better perhaps is the explanation of $aikara: robbers carried away even the bones of Sake- lya while they were taken home by his pupils for the funeral rites. He argues: “Why? ‘mistaking them for something else’, that is: ‘treasure in them'. An earlier story is referred to here: In the Astaddhyayi (= $B book 11)..." etc.; Saikara then quotes from the variant of the legend found at SB 11.6.3. Already in the older $B version Yajflavalkya threatens Sakalya to die before a certain day: he probably predicts the death to take place before 35 the next full moon or new moon.5° During this period, there was enough time for the curse to take effect psychosomatically, while Sakalya was on his way, returning to his home; it could have caused heart failure. I suspect, however, that all of this has other backgrounds. 32. Cf. already Deussen, 60 Up.s, and now, Figer AO 45, p. 74 sqq-, p» 73+ 33. On robbers, typically found in the wilderness (aranya), see W. Rau, Staat und Ges. p. 98: dvyddhin, kuluica, taskara, paripanthin, part mogin, prakrnta, malimlu, mugnat, selaga; a troop of robbers: avy’ dhint (send); cf. also SB 13.2.4.2: "pkgtkah, purugavyaghrdh, pari mogtna, dvyddhtnyas, tdskarah" “ogres, man-tigers, thieves, murderers, robbers", dto. at TB 3.9.1.3.4; - for robbers in RV see e.g. 10-85.2+ 34. The words used in the four versions of the text, as given above, do not mean, incidentally, "wild, carniverous animals" or something the like, but "robbers, highwaymen"; see preceding note. 35. Note that the word dtithi "such and such day" is used. This could mean that even the redactors of $B (JB) no longer knew the actual words of Yajfiavalkya or that they regarded the actual wording of @ legend as immaterial; this is more likely, cf. the Caraka quotations (StIZ 8/9, p. 172-177), the doubling of passages in BAU 3, etc. - For on ttithi-, etavathitht see Wackernagel-Debrunner, Ai.Gr. II 2, § 535 b, § 211 (etdvatitha- is "unbelegt"), Lokesh Chandra, ed. JB IT, 1-80, P. 103 ann. 153 cf. also Eggeling on the $B passage.The Case of the Shattered Head 381 5. As often, it is the early Buddhist texts which provide more detailed, and very useful information. The Pali texts, which have been composed only shortly after the end of the late Vedic period, frequently describe in lively and graphic detail what is only alluded to in the Vedic texts, which vere, after all, composed by Brahmins for Brahmins: one did not have to explain ritual matters of everyday occurrence or of common knowledge to one's fellow Brahmins or to brahnacdrin students when discussing, explain- ing or teaching the ritual, undertaken, apparently, in ‘classroom! on the offering sessions round itself. - There are a number of instances where "the splitting of the head" is decribed in more detail in the Buddhist canon than in the Brahmanical texts: The young Ambattha pretends, among his peers, to be of high birth, vhile his ancestors were born of a slave girl of the Sakka king Okkaka The Buddha explains this and questions him about his ancestry (DN 3.1.20): "‘Anbattha, now this sakadhanmika question is put before you (‘reaches you'); it has to be answered without passion (akdma), If you will not answer it, or if you will answer something else, (as if having been asked) with another (question), or if you will (remain) silent, or if you will go away, your head will burst asunder into seven pieces at once.! (ayam kho pana te, Ambattha, sahadhanmiko partho dgacchati, akami vydkdtabbo. sace na vydkariseast, arriena vd arian paticarissast, tuyht vd bhaviseast, pakkamissasi vd, ethh'eva te sattadhd muddhd phalissatt) Ambaggha is asked two times, but keeps quiet. The Buddha then warns him: 'If one, having been asked the sakadhammika question by the Tathaga— ta three times, does not answer, then one's head will burst into seven pieces at once.! (Yo kho, Ambattha, Tathdgatena ydva tatiyakam sahadhammikam pavtham puttho na vydkarott, etth'ev'assa sattadha muddhd phalissatitt.) "At that moment, the Yaksa Vajrapani, having taken a large metal ham- ner, was (appeared) above Ambattha in the air, (with the intention to kill him if he did not answer, thinking: sahadhammtkam parthan puttho na vydkarissati, etth ‘'ev'assa sattadhd muddhap phdlessdmiti) ... Only the Buddha and the young Ambattha saw the Yaksa Vajrapani."382 Michael Witzel (tena kho pana samayena Vajirapdnt yakkho mahantam ayokiitam dddya, ... Ambatthassa ménavassa upari vehdsatthito hott ... Tam kho pana Vajira~ panin yakkham Bhagavad c'eva passati Ambattho ca maxavo. ) Out of fear, Ambattha then gives in publicly and agrees with the Bud- aha's description of his ancestry .3° In a similar incident (DN 5.21), the Brahmin Kitadanta does not agree with the Buddha's explanation of ‘true sacrifice’ (i.e. that made without offerings of animals), but he does not venture to present a counter-argu- ment and remains silent. The other Brahmins ask him why he does not agree with the Buddha. He answers: "It is not, Sirs, that I do not accept as well-spoken the well-spoken words of the Samana Gotama; - the head of the one who does not accept the well-spoken words of the Samana Gotama as well-spoken would burst. But, Sirs, this is what (troubles) me ..." (... muddhd pi tassa vipateyya yo samanassa Gotamassa subhasitan subhdsitato ndbbhanumodeyya. ) Kutadanta has to agree, as he cannot state another truth; conse- quently, if he would openly challenge the Buddha, his head would burst. In a later text, the Mahavastu, (ed. Senart, vol. 3, p. 114.12) it is sufficient to let a person's head burst into seven pieces if the Buddha stands up in order to greet him. In the present instance, this is way the Buddha avoids to grect his father in this way after a long absence from home. (ndsti ca so satvo vd satvakdyo vad yasya tathdgate pratyupasthihante na saptadhd mirdimam sphaleya)37 In the Suttanipata, a Brahmin challenges another one to give him sone food: "If you do not give to me, who asks you, may your head split into seven pieces on the seventh day!" 36. For the rules governing such discussions see M. Hara, Mittabi, ['Three times'], BukkyS kydri no kenkyi, Tamura Yoshiro hakase kanreki kinen ronshii, Tokyo 1982, pp. 527-543. 37. Cf. also Mv, Vol. 2, p. 26.11 saptadha ii teyatt (of a goddess!), cf. transl. p. 23. ceeThe Case of the Shattered Head 383 (SN 5,983 sqq.: sace me ydecamdnassa bhavam ndnupadassati, sattame divase tuyham muddhd phalatu sattadna)38 And in the Dhammapada Comm. 3, (vol. 1 p. 41), in the story of Tissa Thera, one ascetic curses another one in a similar way: "The sun with its thousand rays and hundreds of flames, which is the dispeller of darkness, when it rises in the morning, may split your head into seven pieces.” Note that here it is the sun, similar to the Yaksa of the earlier texts, which carries out the threat. In the present case, the curse had to be revoked voluntarily; even a strong intervention by the king did not help.3? Against this background the few similar references from the Brahnana texts gain special importance. It is especially the wording of DN 3.1.20 that is closely matched that of a late4° Vedic text, JB 1.46. The passage speaks about the way to heaven by a desceased: 38. The period of seven days looks as if it had been inspired by the seven pieces the head splits into, but cf. Yajfiavalkya's threat towards Sakalya: he will die on "such and such a day", i.e. after a certain number of days, most probably 7, 14, 21, or jo days 39. Cf. The Dhammapada Comm. of Buddhaghosa, transl. by the Dept. of Pali, Rangoon Univ. 14 sqq. - This case again looks very much like the usual saccakiriyd; cf. also Dhammapada Comm. ed. 1, p. 17, 24: "then ny head must split into seven pieces" sacdham ... santikam na gamised- mi, mddhd me sattadhd phdleyya; Av., emend. Speyer: niyatam devasya eaptadhd mirdhdnam sphdlaydmi "I will cause to burst" (but MSS 3rd sing. Cf. also DN 3.1.21:1, p. 95, SN 2.1.9: 1, p. 503 typical saccakiriyde in these texts: Jat. 1, p. 543 Jat. 4, p. 320; Jat. 5, Ps 92; Fausbéll, Ten Jat., Copenhagen, 1872, p. 94; Miln 157 satadhd va sahassadhd va(!); DhA 1.17 muddhd me sattadhd phaleyya. Dhammapada, Fausb. 1855, 87, 140; DhA 1.41 muddhd te phalatu satadhd; VvA 68; for adhipateti, see Liiders, Phil. Ind. 179 sqq. (phal/phat). - There are a number of similar cases in the Epics as well: Mbh. crit. ed. 3-275.33 (3.16564) yadé [Ravayo] hy akdmdm dsevet striyam anyan apt Ghruvam, Satadhdsya phaled deha (v.1. mirdha) iti ukthah; furthermore: Woh. 7.6265; Mbh. 3.1603 tasya mirdhdsya deadya, paphdla; Ram. 2.64.23 saq. yady etad adubham karma na me kathayeh svayan. phalen miirdhd sma te, rajan, sadyah éatasahasradhah, ... saptadhd tu bhaver. murdhd munau tapasi tigthati; Ram. 7.26.56 yadd hy akdmap kdmdrto bhavitd tada (cf. Mbh. 3.275.331). 4o. It comes froma later part of Jaiminiya Brahmana, see Bodewitz, tr. JB 1.1-66, p. 9 sq.; repeated at: The daily Evening and Morning Offering (Agnihotra), Leiden 1976, p. 5 sa-384 Michael Witzel "To him one of the seasons, who holds a hammer in his hand, having come down along ray light, asks him: 'Who are you, man?! In case he knows something, he may withhold (his mame). He (the Rtu deity) then kills him (‘strikes forward! = ‘strikes off his head') with his han- mer." (tak ha ptindm eko yah kitahasto ragmind pratyavetya, prechatt: ko 'si puruget?, oa kim vidvdn pravyfijydt, tasya ha praharati)4! A similar punishment occurs also at other occasions, see for exanple JB 2.269 sqq. § 151: Yavakri Saumastambhi was about to sleep with an Apsa- ras when a Gandharva, a metal hammer in his hand, approached to punish hin. (atha hedam evdyahkitahasto gandharvo 'bhivicakrame ...) The prdyascitti for this was: Yavakri must cut off (Siras chinddhi) the head of all living beings of his father's trek (grdma, see Rau, Staat u. Gesellschaft im alten Indien, p. 54) before the next sunrise. However, according to some, a deaf carpenter killed Yavakri (pragaghdna) when he saw him next morning at sunrise, going round and slaying everything, and therefore believed him to be mad; according to others, the Gandharva did so: “People said: ‘Surely, the Gandharva did kill him.' (‘stroke forward [the head] of him.')"4? (bddhirah ... takgd ... sa evdsya prajaghdnety eke ... tasyodite sa eva gandharvah prajaghénety eke. yathd ha tu mamdra tathdsa. Saévad dhdsya sa eva gandharvah pragjaghana.) Obviously, in the late Vedic period and at the time when the Pali texts were composed, someone to be punished by a supernatural being, like as Yaksa, a Gandharva, or a Rtu-devatd, is killed by a blow of a metal Al, Bodewitz., tr. JB, p. 115 wrongly translates "strikes at him with his hammer", 42. It is interesting to note that the authors of JB do not know or do not want to say whether it was the village people (the carpenter) who killed him or the Gandharva. Is this already a deliberate nystifi- cation of killing a mad man?The Case of the Shattered Head 385 hammer and his head splits (inte seven pieces, as the Buddhist texts 43 ). regarded as something quite real and is also described in realistic way - say In these passages killing is shattering someone's head with a hammer. It is time now to return to the actual situation when the warning Fronounced: "don't ask beyond the deity/do not overquestion - or your head would split apart!" In the Brahmapas and Upanigads these warnings occur, as has been mentioned above, in a situation of 'theological' contest, or one of questioning on the fine points of the ritual or of other sacred/spiritu- al knowledge. A Brahmana challenges others or is being challenged himself to answer difficult questions on the ritual or on the brahman, We are, to some extent, informed about the history of such theological discussions: From the Brahmapa texts we know that they usually took place when a yagandna or a group of people (i.e. in a pre-clacsical Sattra, see H. Falk, Bruderschaft u. Wiirfelspiel, p. 32 sqq.) performed a larger (Soma) ritual or when a king undertook an elaborate ritual like the Agvamedha*! and was about to select his priests (see $B 11 and 14; cf. ChU 1.10-11). Later on kings like the famous Janaka of Videha are reported to have initiated such discussions to further "theological" or philosophical discussion. In some rituals, like the Agvamedha, or the Sattras and their later counterparts, the Dvadagahas, they are part of the ritual itself. But these duels are so much older: Already the Rgveda centains some Brahmanicel "riddles", brahrodya. 43. The seven pieces are those of the bones of the head which grow togeth- er only after birth, at the fontanels. It is at these seams that the head splits in cremation; cf. below. ~ Note that killing with the blow of a hammer on the frontal bone is, according to SB 3.8.1.15, the "way of men" when slaying an animal at sacrifice, or that of the pitrs (slaying it behind the ears); the "way of the gods" (~ and therefore the only proper one in sacrifice —) is, as always, different from that of men: it is to strangle the victim, as to avoid obvious bloodshed. - Cf. also the "head of stone": of an enemy, to be killed, AB 8.28 (end), aeyaémamirdhd bhavaté 44. At $B 13.2.6.9: before killing of the horse; cf. the Yajfavalkya story, $B 11.6.3 (see above § 4, and § 7), cf. Kuiper, TJ 4 (1960): The Ancient Aryan Verbal Contest, reprinted in: Ancient Indian Cosmogony, Delhi 1983, pp. 151-215.386 Michael Witzel 6. They occur at several instances, notably in the long, enigmatic hynn RV 1.164; its verses 34-35 are one of the few (old) cases where question and answer were given. The whole situation, however, is referred to in details at RV 10.88.17-19: early in the morning, before sunrise, the Brahman and Hotr priests take their seats at the place of sacrifice and test each other with questions like: "How many fires are there, how many suns, how many dawns, how many waters then?" (cf. Geldner's commentary); compare also RV 10.166 and cf. PS 10.2, a sapatnaghna hymn consisting of sorcery mantras directed against rivals. In later texts, like VS 23.45-62, the answers to such riddles are provided as a matter of convenience, cf. also the brah- modya at $B 13.2.6.9 sqq. where questions from VS 23.9 sqq. are answered in both in the Samhitd and in the Brahmana portion.4> These "riddles" can - as far as their form is concerned - also be detected in other old Indo-European texts, for example the Zoroastrian Gathds, as in Y.44, which contains questions addressed to Ahura (see below § 9 and cf. C. Watkins, Aspects of Indo-European poetics, 1982). The Av even contains charms intended for succesé in speech contests. AV 19.28 = PS(K) 13.11 = PS (Or.) 12.22, is a sorcery hymn against rivals in speech contests. Vs.4 runs: "split, © darbha, the head of our rivals, of our haters, o amulet, make their head fly apart, as the rising (sun) does (split) the skin of the earth." (whitney) ‘© (bhindht, darbha, sapdtnandh hidayam dvigatdm, mane, udydn tvécam iva bhimydh Stra esd vt pataya) Such outbursts may be regarded as typically Atharvavedic, but reality hides behind this wish full of hate: this can easily be noticed in other ritual practices, and especially those of the "pre-classical" Vedic sacri- fice; for example in the graphic descriptions found in the stray references 45. For the term and function of the brahmodya, see F. B. J. Kuiper, The Ancient Indian Verbal Contest (see ann. 44). 46. Cf. the similar statement in Sarkara's quotation, BrahmasOtrabhasya 3.3.25, from the lost *Paipp.-Aranyaka which included the *Paipp.Up. sarvan pravidhya, hrdayam pravidhya. dhamanth pravpjya, Siro ‘bh pravrjya tridha viprktach> ity ddih; see Journal of the Nep. Res- Centre, 1, 1977, p. 143.The Case of the Shattered Head 387 referring to the original Sattra rituals,47 that exist both in the YV-Samhi- tas and in the extent Brahmanas. These Brahmana texts, however, also reveal, when searched conscien- tiously, a number of passages which shed more light on the whole context. They are quoted in extenso. (a) JB 2.303: Some Brahmins want to offer for Janaka of Videha (with the janaka-kratu ritual, the 7 day rite).48 He refuses: J: "There was no Brahmin among you who was more learned in brahnan (brahmiydn), and who did execute (vy~a-dhdt) this form of the sacrifice (yajfakratu). - Make me offer (as the yajdmana) with this well-known one (kratu of sacrifice) for me (md)." - Br: "Yes". With that one they made him offer. - (2.304) [detail explained.] - With that (ritual) Kratu Karfradi, who wished wealth ( sr) and food, offered for himself. ... In this way are these two forms of the yadfakratu Ky "As such one (i.e. possessed of food, etc.), I will put from the middle of myself food!" As one who is 'food eating’ (recipient of services and goods, see W. Rau, Staat und Gesellschaft, p. 32), I will become the best of my (relatives). - That will support me, who has been surrounded by food, throughout the year (sanatsarena). Then , I will step out towards the top (of the sky) (varema), to Sri, to the (highest) back (of the sky, prstha). - To him, a Brahmin said: "You (bhagavdn) will live; because of food you will become the best of the relatives. But without head you will f its (the 482 this does not hurt (read *klédyati "cause pain, torture"); it is cut off (only) die.” .., The Mahvrata is the head of the sacrifice. K.K.+ yajria's) head is cut off, one should conceal it under the armpits. this far / so long (i.e. off the body of the sacrifice)" he (Kratu Kariradi) said. 47. See H. Falk, Bruderschaft und Wiirfelspiel, Freiburg 1986, p. 30, sqq., and for the speech contest in the abd, p. 94 sqq.; cf. the earlier treatment in ZDMG-Suppl. 1985. 48. Note, in contrast to FiSer's opinion (AO 45, p. 70 sqq.-: the $B/BAU tales are too artificial to be historical) that Janaka appears here quite naturally; at this instance he does not seem to be intentionally introduced as to show the progress of Upanisadic speculation; cf. ann. 19. For Kariradi see JUB 2.4.4. 48a. See K. Hoffmann, Aufs. p. 1933; or are the bushes along the two sides of a river intended, cf. hiding the,pravargya pots on an island of a river, Kathd 3.233:96.6: githati patrdyt dpi va dvtpar hareyuh.388 Michael Witzel Bre: like that: As long as Kratu Karfradi lived he was, as ‘eater of food,' the best of his relatives. Without head, he died. But (?) knowing ones cut off his head, the fearsome ones (bhdyakdh). ~ (Br.:?) "That is the well-known auspicious form of the sacrifice, which he/one should offer (in one's own "Knowing ones will cut off his head" he said. And it was just interest), with you (= J.?) as a spectator." (itz)? (tam u ha Brdhmana uvdea: yadvad jivigyaty anndd eva Sregtha svindn bhavigyati, visirgds tv eva marteti. Siro vd etad yajasya yan nahé vratan. tad yathd gira’ chittva, tad upapakgayor upagihet. tad vai na Sligyati [v.1.: Svigyat? read: *klisyati??], tadrg akyd iti hoviea, dananta evdsya Siraé chettdra iti, tad dha tathaivasa. ydvad dia Kratuh Kartrddir jijivanndda eva Sregtha avdndm dea. visireds tv eva mamdra, jananto ha tv evdsya Siraé cicchidur bhdyakdh. ea haiga kalydna eva yajiakratur yas tvatnendnupadrdstre yajeta.) (b) Even a comparatively late text of Anubhramapa type, VadhP 3.94, retains more direct statements:°° When the horse that is offered in the Agvamedha sacrifice is to be cut up and its skin is to be taken off, one leads a boy to the offering ground to do this: 49. This passage has many difficulties: who is the upadrdgty? - Probably it is J.; read yas tv *enenopadrdstre? or yas tvatenenopadragtre yajety cf. VadhB. 4, 28d, p. 136 tasmdn nu ha yadmasyavid eva ayin nopadragtd "therefore one should be someone knowing the ritual, not only an onlooker. - However, it seems clear that indeed somebody's head was cut off - by knowing specialists! Note that in the Agnicayana sacrifice, heads still are cut off and interred, see already the opinion of Hillebrandt, Rit.Lit. Intr. p. 9 sq., and cf. Heesternan, The case of the severed head, W2KS XI, 1967, p. 22-43; repr. in his coll. articles: 'The Inner Conflict of Tradition’, Chicago 1985; cf. below ann. 104 about the importance of this feature of Vedic ritual and myth. So. For the date of the text (late Br., early Up. period), see author, StII 1, p. 93; for the translation of this passage, cf. Falk, Wirfel- spiel, p. 160.The Case of the Shattered Head 389 "They lead, - crying, as if someone were going to dic, - the (said) son of a noble sta, a young, (specially) adorned boy who has not yet spilled semen, as cutter (vigasitr) of the horse. - One says: ‘Who formerly used to take off (flay) the skin (of the horse) as the first, his head would fly apart." (Gnayanty eta siitaéresthasya putram kumdvam asiktaretasam aévasya vigasitdram alamkytya rudanto yathd marigyantam evar. yo ha amety dhur etasya purd prathama dehyati, murdhd ha emdeya vipatatiti.) Immediately after this, a legend follows which tells of happenings that apparently took place only three or four generations before the time of the author of the passage: "When KeSin Maitreya offered an ASvamedha, one lead (in the same way) the grandfather or great-grandfather of Ekayavan Kandama tama Mamateya heard about this and sai +++ Dirgha~ ‘Come, boy, I will tell you how you shall cut the horse, so that your head will not fly apart!" (kundreht, te'ham tad vakgydmi yathd tvam evdévak visasigyast, no te nirdhd vipatigyatt. ) Dirghatama then told him to cut with the blade of a knife (and not the edge: thus Falk, differently from Caland), using a particular 6-verse hymn (IS 5.2.12). Somebody else would then approach him and tell him to cut in the traditional manner ... Exactly this happened: Someone came and said: "Boy, what are you sitting there, torturing the horse? Thus indeed you will cut up!' And he took the knife and cut it open. His head flew apart. That, one should do in exactly this way." (umdra, kim tdam avam kliénann dsso, ittham vd ava visasigyastty asim ddéydeachau. tasya ha mirdhd vipapdtas tan nu haitad evam eva kurydt ) Again, somebody who does not know the correct way of ritual action gets killed, - a fact that has been noted already earlier. Here, this tra~ ditional practise is ascribed to a time only slightly before the end of the Brahmana/Upanisad period. At the other passages too, and notably again in VadhB, it is mentioned that people, even accidental passers-by like merchants, are robbed for not : ee st knowing certain ritualistic details. 51. See VadhB 3,75: AO 4, p. 180 on the Asvamedha.390 Michael Witzel From the work of H. P. Schmidt we know that the Revedic and other forms of the ‘pre-classical' Vedic ritual generally involved straightfor- vard and "visible" killing of victims, for example that of an animal victin which vas decapitated at the offering pole.” According to J.C. Hee- stermann, every pre-classical sacrifice even involved a fight between two parties.°3 However, there apparently were two (or more?) strands of tradi- tions, namely the "violent" open killing of the victims and other, less obvious forms which co-existed even in Rgvedic times. None superseded the other(s) for a considerable length of time. (One must not, of course, con- fuse this with metaphorical statements, such as the one about not killing the ‘cow’, i.e. poetry). That, also with other items of the ritual, several (regionally or clan-wise) divergent customs existed next to each other is well known (see Hillebrandt, Ritualliteratur, p. 16) and should not sur- prise; compare for example, that even today, there remain several types of ‘violent! sacrifice next to each other with the Trobriand islanders, fanous in anthropology. One of them is an outright ‘bloody! sacrifice, while another form uses strangling to kill the victims. The progress of the ahirisd concept (cf. L. Alsdorf, Beitrage zum Vegetarismus and H. P. Schmidt, Mélanges Renou) had succeeded in pushing the bloody sacrifice (viz. the less overtly killing of the victim by strangling) somewhat into the background by the time of the Upanigads and the Buddha, (cf. above § 5 on rue sacrifice"). However, this trend did 52, Falk (and others, incidentally) use the same terminology; "orthodox brahmins are those who succeeded in ordering the various sacrifices in a hierarchy, by eliminating all wild aspects, and in 'civilising’ it", see Bruderschaft, p. 37 ann. 94. - Cf. H. P. Schmidt on Revedic sacrifice in his article on panthds, ITJ 15, p. 1 sqg-, see esp. P+ 353 cf. in general: Heesterman, coll. articles: 'The inner conflict of Tradition’, Chicago 1985." $3. I wonder, however, how this could agree with the clearly stated aver- sion of priests, even in the RV, against overt killing, see the hynns about the horse sacrifice RV 1.162-163; contrast RV 8.lo1.15 with the metaphorical use of the word, "Do not kill the innocent cow, the Aditi": and cf. also aghnya- (see H. P. Schmidt, KZ 78, 1963, Pp» 1 sqq.; J. Narten, Vedisch aghnyd und die Wasser, Acta Orientalia Neer- jandica, Leiden 1971, p. 120-134); note, that in YV, even the sinple cutting of a blade of grass is avoided in ritual and must be done with care (at the "joints", as with animal bones): MS 4.1.1.:3.9 sq- 8 31.1, TB 3.2.2.1 (cf. TS 1.1.2, $B 3.6.4.10; for the animal protected while being cut by grasses: $B 3.8.2.12).The Case of the Shattered Head 391 hold sway throughout history. Some Tantric forms of worship continue, for example in Nepal, with animal sacrifices (by decapitation but also by slowly letting the animal bleed to its death). Even human sacrifice was practised at Kamakhyd in Assam and probably elsewhere (and it is rumoured to have occurred until recently at Harisiddhi, South of Kathmandu). It has been known since long that the "classical" Vedic sacrifice excluded “overt killing; this was done outside the offering ground by helpers, and executed no longer with an ‘open' shedding of blood but by strangling. The victim had "to agree" or was "pacified". °4 Even in classi- cal Vedic Srauta ritual, however, one had to obtain a human and several aninal heads which were inserted into the gradually built-up Agnicayana altar.>> All of this is not too surprising when viewed against a background of such violent actions as the one reported at RV 10.171.2. "(Indra,) when you took off the head of the resisting Makha from its skin ..." (¢vép makhéeya dédhatah Gird 'va tvacd bharah). Apparently, the flaying of Makha's head is referred to here; his skull is brought by Indra, perhaps, to the sacrifice, cf. the following pada: "You went to the house of the one who alvays (Suffix ~int) possesses Soma." (4gachah comtno grhdm). This reminds of many ‘barbaric’ customs, such as those of some Scythian and Germanic tribes who drank from bowls made from the heads of their enemies, or it reminds of the enigmatic skull sculptures found in Celtic temples in France. As for Vedic India one can hardly argue against the existence of similar custons, given the evidence of the texts themselves, even if they indicate a gradual "humanisatio " of sacrifice. (c) VadhB 4.108 (p. 229) reports that one formerly indeed offered a man as victim for Prajapati at the time of building the altar of the Agnicayana, as for example Karpdjaya did. Dhartakratava Jatakarni (his 54. Cf. Oertel, Euphemismen, SB Miinchen 1942, Heft 8, p. 6-8; T. Goto on labh/rabh, Indogaku Bukkydgaku Kenky® 24, 1976, p- 1015-1007 (= Engl. sect. p. 23-31). In the present day Nepal, the victim ‘agrees’ by shaking its head sideways (the human gesture for 'yes'), after some water had been poured on the head. 55. Heesterman, The case of the severed head, W2KS XT, 1967, p- 22-435 for other details see StI1 8/9, p- 173, ef. VadhB 3.59 and $B 7.5.2.392 Michael Witzel descendant?) did not wish to eat the idd portion of the offered person; the gods therefore exchanged man as a sacrificial animal with a horse (and ‘retracted further from mankind'!). When his descendant, Rahabksita Jata- karni, did not want to eat of it, they again changed the sacrificial aninal to a hornless he-goat. By the time of the composition of VadhBh, one even Jet loose these animals, and offered only animal figures made of rice, barley or clay. (4) While these comparatively late texts underline the existence of human sacrifice, or of the loss of the head as part of a ritual, a passage fron JUB accidentally deals with the ritualistic or competitive discussien and the dangers to one's life involved: It mentions in some detail, but unfortunately does not, as all the other Brahmana texts, describe the conditions of the challenge, the social and ritual set-up, and the dire consequences. It is, therefore, worth to be quoted extensively: 57 "Sudakgina Ksaimi, Practnagali and the two Jabala (brothers) were fel- low Veda students. They repeated (after their teacher, = recited) much of what is to be mumbled (in Yajurveda ritual) and of other (Rgvedic, Sa- mavedic texts). Now S. K. used to ask (them and their teacher), that what is quickly (understood), what is well known. They, being distracted (in study), ran about, crying out: "The Sidra (i.e. S. K.) is one who is not learned!" [Oertel: "we don't know"]. Thus, both Pracinagali and the two Jabalis, used to cry out (vy.d.krus) about S. K. - S. K. used to say: "Where the best Kuru-Paficdlas have come together, at that (place) shall be our discussion (samvdda)! Let us not discuss (sam.vad) without a specatator 8 (an-upadrasty), like Sidras 1" 56. Cf. also the observation of Falk, Wiirfelspiel, p. 37 sqq.: There are some references to cannibalism at TS 72.10, KS 34.11. 57- I thank H. Falk, who has read the Paris draft of this paper, for point- ing out this passage to me. 58. Cf. above on upadrdsty; apparently there were many spectators, just as they are attending solem yajrias and pujds today; but in Vedic times they came to disturb, “creeping up" to ask questions (cf. Falk, Wirfelspiel, p. 34); only the 11th day was avivdkya "without dispu- tation". ~ Note that open discussion in public is a point of honour ("Sidras do not act like this")! - For samvdda cf. Falk, Wiirfelspiel, p. 189, ann. 554; ApGS 22,19; Heesterman, Kautalya, WZKS XV, 1971, P+ 16 sq. - For a ritual discussion between the Kurus and Paficdlas, see JB 1.262, Caland § 87, Oertel KZ 58, p. 81.The Case of the Shattered Head & The two Jabalas, Sukra and Gogru, consecrated themselves. Practnasali vas chosen as Udgatar by the two.” Sudaksina got to know this: 'The two Jabalas have consecrated themselves.‘ He told his chariot driver: "Hey (ave!), come! (Grayasva). The two Jabalas have consecrated themselves. We will go there!" His relatives were, by and large (iva), in tears: "This one (just now) went to one (of his two, or three possible lots) [digtam?]!"©° Now, one used to think of someone who formerly ‘spoke a brahmavddya' (took part in a ritual discussion), ‘he went to! (his death: cf. pra.i, preta-). Now, they used to serve him like one who really vas (more or less) dead.°! The driver said to him (S. K.): "Now, Sir, as there is nothing pleas- ant for you by these two, (they treat you badly), why do you speak thus (shy do you say you want to go?)" (S. K. said): "Gi (so be it, yes)! I have to go. The teacher had thought them easily tameable."©? He mounted the chariot and made (his driver) drive off.°3 they (the priests, Pracinasali, the two Jabalas, etc.) saw him. "Do you know him?" (the other Sattrins asked), - "(It is) Sudakgina" (the Jabdlas) answered. - "He must not get down here now (and sit down here, on the offering ground, idam, i.e. to 59. Note the same situation in KB, see Ved.Ind. s.v. Jabala, Citra Gaugra- yani. Here it is Citra who comes from the North of the offering ground and helps out in the discussion. - Gaugla also occurs as an interlocu- tor at a Vigvajit rite, AB 6.30. 60. For the meaning of iva see now: Bereton, JAOS 102, 1982, p. 443-450. - It is not known at this instance whether there was an idea of pre~ destined (digtam) span of life (dyus) in Vedic times, but see Oertel, KZ 58, 1942, p. 154: He can adduce only AV 3.11.8; cf. however RV 10.59.4 dyibhir hitd jarimd, cf. comm. of Geldner with parallels from MS, KS; cf. also dégtré 'fate' at RV 10.85.47; for digtam ‘fate! cr. also Pali ditthia < digtyd "thank god!" see 0. v. Hiniiber, Das iiltere Mittelindisch im Uberblick, Wien 1986, p. 151 § 339. - Note, however, the case form of anyatardn (acc.sing.f., acc.pl.m.?) 61. Cf. Oertel on pra-i: Euphemismen, SB Bayr.Akad., Minchen 1942, Heft 8, p. 9-13. 62. At JB 1.285, Aruni is the teacher of Samgamana (1!) Kgaimi.; ~ note that ont tyes', also at JB 2.77 (Vidagdha Sakalya), AB 7.18 (Haris— candra). - Note that S.Ks. already answers, as if already taking part in ritual, not with tathd 'yes', but with or 'yes'. 63. For the meaning of (pra.)dhdv, cf. Bodewitz, ITJ 16, p. 93; note, however, besides "driving around/towards a goal", actually "making one's driver drive towards ...", the special meaning "to drive towards a discussion" is obvious in some texts, cf. $B 11-6.2.43 JUB 3+2-4-85 at 3.2.3.5 a king leaves for a samiti JB 1.234 § 87-394 Michael Witzel question us)!" - "It is only pim."©4 He (S. K.), having stepped dovn fron the steps (of the chariot?) in(to) the offering ground (antarvedi), said: ‘Well now, thus! ... 0 Grhapati ... (leader of the Sattrins)?" He (J.) did not wish to get up after him (S.K., although having been challenged’ He (S.K.) said: "Be one who will get up after me! You are one who wears the black antilope skin."° 64. He is, however, not as stupid as the two suppose him to be: see JB 1.285 where S.Ks. puts an intelligent question to Satyayajfia Paulusi which the latter cannot answer; cf. Oertel KZ 58, p. 81. 65. PW col. 1304: anu.ut. sthd 'to get up after someone’, as for example in ritual Agv$S 4.10.7; - but contrast BAU 3.8.2 Gargi: upa.ut.sthd ‘to charge someone (with questions)". Cf. ann. 67. 66. Apparently the same situation, with reversed roles, is found at $8 11.8.4: KeSin's samdj cow was killed by a tiger during his Sattra. He went to Khandika Audbhari who knew the atonement but who also vas his enemy. Kegin knew this, and said "If you will have me dic, I shall be torn apart along with the sacrifice that is torn apart." (yddy uma marigydet, yashdn vtkrstam dnu vtkrakgya ttt; note the wording, simi- lar ‘to the one about the "shattered head") - Kh. seeing him coming said: "As there are those skins on deer, we break their ribs and cook (them, the deer). Have you dared to drive up to me, (thinking) "The skin of the black antelope is attached to my (Kh.'s) neck'2" (yde nv btdny evajtndni mrgégu phdvanty, dthaigin pretir apisirya pacanahe. kergndjindm me grivasv dbaddham tty evd meddn adhygd 'bhydvasyanttisn ttt). Kh. then told him the atonement, knowing that Kesin would be successful on this earth and that he, Kh. would die (apparently during his own sacrifice: he wears the antelope skin) and have "more nights up there" in heaven. - kh. thinks that K. challenges him at his om sacrifice, and "wants his head", wants him killed and "cooked" (Cf., on eating human flesh during a sattra, Falk, Bruderschaft, p. 37, of- above, ann. 56). Wearing the antelope skin makes the wearer equal to an animal which can be hunted and eaten! - In the passage under discussion (JUB), it is the challenger, S. Kg., who wears an antilope skin (either because he still is a brahmacdrin, or, more likely, as he has prepared himself for ritual, (note his "sacred speech": his answering with or); the Grhapati (Jabala) and the other Brahmins taking part in the Sattra wear black dresses and in addition, skins (perhaps sheep skins, see Falk, Bruderschaft p. 20 sq.)The Case of the Shattered Head 395 The Kuru-Paficalas knew this; they said: "He (J.) is one who will get up after you (S.K.)." The younger brother (of the two Jabalas) told him (J.): "Get up, Sir, only after the Udgatar (Pracinagali)!" He (J.) got up after him (Practnagali).°7 He said: "Three times, o ‘householder', man is born [etc., on the efficacy of the Udgdtar to help beyond the death thrice].°” Having said this much, he mounted (his) chariot and made (his driver) drive off (having overcome Prdcinagali, the Udgatar). To the (elder) Jab@la wao had come back (from the discussion), the younger brother said: "Which speech, Sir, has the little Sidra spoken (just nox)?" - "You have sought a ford with an elephant" (i.e.: You made an ele- phant out of a molehill). To him who had said "what speech, Sir, has he spoken?" the (older brother) declared: He who knows the driving across the three deaths by means of a Saman, is the Udgatar. He drives across death." He said: "Your father indeed had thought of him (S.K., amanyata) as an tdgatar!"7° The Pracinagala knew which of them was the one chosen as an Udgatar here (at this sacrifice, i.e. now S-K. and no longer PracinaSali). They did not agree with this. They said: "Run after kandviya!" They ran after him. They made Kandviya the Udgatar and Pracinagali the Brahman priest.’ 67. Apparently, to get up “after someone" means to challenge someone, to answer questions in a brahmodya. The challenger and the challenged who already was present on the offering ground (antarvedt) seem to go to another part of the sacrificial area, apparently in the North. This is another term which has to be studied in detail, cf. ann. 16 and JMB 3.2.1, JB 1.337-338 § 115. Apparently, he gets no answer; for getting up cf. also AB 8.28 (end): not getting up before the enemy, in a different situation, though. Cf. getting up as a sign of politeness, TU 1.11, KathSiU 11, at W2KS 23, P- 28 and 24, p. 66. 69. where did he come back? Or does the vivdkya take place elsewhere? i seen cae a = For the address ‘Sidra’ cf. ChU 4.2.3 (Janagruta, apparently the king of the North-Western Mahavrsa land, is addressed thus by Raikva). Jo. That means: he thought him capable to be an Udgatar; the teacher mentioned here apparently could, according to JB 1.285, have been Aruni, - who is not a Jabalal 71. Who has to remain silent during the ritual; just like Pr., who apparently did not know an answer.396 Michael Witzel (S.K.) having looked down on him, he said: "Thus has this Brahmin not been tired of useless (mogha) talk. He does not seek the subtle of the adnan." — He did this beyond him (overcame him) ./*" (2, Sudakgino ha vai Kgaimih Prdcinaédlir Jabdlau te ha sabrahnacd- rina dauh. te heme bahu japyasya canyasya cdniietre Prdcinasalié ca Jabétay ca. 3. atha ha ema Sudakginah Kgaimir, yad eva yajfasydijo yat suviditan, tad dha ematva prechati. 5. ta u ha vd apoditd vydkrosamandé cerué 'Sidro im Kgaimim dkroganti Prdcinaéalté ca dabdlau. duraniicdna’ itt ha sma Sudakg 6. 8a ha emdha Sudakganah Kgaimir, yatra bhiiyis bhavitdras, tan na ega saywddo ndnupadrdgtre sidra iva saqvadigydnaha iti. 7. ta u ha vat Jdbdlau didikgate, Sukraé ca Gosrus ca. tayor ha Prdcinaéi- lir vpta udgdtd. 8. sa tad dha Sudakgina 'ububudhe: Jdbdlau hadékgate. ea ha samgrahttdran uvdednayasvdre! Jadbdlau hadikgigdtam; tad gamigydva itt. (3.7) hah Kurupatiedlds sandgatd (1. tasya ha frdtikd aSrumkhd ivdsur: anyataray va ayan updgdd itt. 2. atha ha sma vai, yah purd brahmavadyap vadaty, anyatardn updgdd iti ha smainam manyante. atho ha smainan mptam tvaivopdsate. 3. tap he sangrahito 'vdedtha yad bhagavas te tabhydm na kuéalap; kathettham dttheti. 4.on itt hovdea: gantavyam md; dedryas euyamin amanyateti. 5. sa ha ratham dethiya, pradhdvaydm cakdra. tam ha sma pratikgante. 6. kam jdniteti. Sudakgina itt. na vat minam sa idam abhyaveydd itt. ea evett. 7. sa ha sopdndd evdntarvedy avasthdyovdediga nv itthan grgapata3? iti. tam ha ndnidatigthdsat. sa hovdednitthdtd ma edhi. Kpgydjino 'stltt]. tad ime Kurupatedld avidur: anitthdtatva ta itt hocuk. 8. tan ha kaniydn bhrdtovdednittig¢ha bhagava udgdtdram! itt. tam hanittasthau. 9. sa hovdea: trir vai grhapate purugo ddyate ... (3.8) 3.9.8 etdvad dhaivoktvd, ratham Gsthdya, pradhdvaydn cakava. 9. tat ha Jabalam pratyetap kantyén bhrdtovdea: kam bhavai chiidrako vaean avéditi. hastind gddhan aigir itt. 10. pra hatvainam tac ehasaisa, yah kathan avocad, bhagava? iti yas traydpdm mptyiindm sdmativahan veda, ea udgitd. mptyum ativahatiti. (3.9) 72. PW ati.kr ‘to do something in excess', as in ritual: TS 6.6.2.1 cf- also: paro bhi at SB 11.6.3/JB 2.77, see above § 4.The Case of the Shattered Head S 1. tam vdva, bhagavas, te pitodgdtdram amanyateti hovdea. tad u ha Prdctnaéald vidur, ya egdm ayam vyta udgdtdsa. tasmin ha nanuviduh. 2. te hocur: anudhdvata Kaydviyam! iti. tam hdnusasruh. te ha Kaydviyam udgdtéran cakrive, brahmanam Prdcinaédlim. 3. tam hdbhyavekgyovdcaivam ega brahmano moghdya vdddya ndgldyat. ea ninu samo 'nvicchatiti. ati hatvanan tac cakre (3.10)) The role that personal prestige played in the Brahmanical society becomes quite evident in this passag one has to react to an open chal- lenge, if that is pronounced in public. Not to accept it means to be like a Sara to be lesser than the three twice-born castes; Sadras do not dispute publicly in a formal way. To lose or to win in such a public discussion has an immediate effect on one's standing inthe community of the Twice-born castes. - The passage also stresses the risks one undergoes when taking part in a ritual discussion: one is regarded already as more or less ‘dead', that is at least liable to die or to be put to death; or one was, until very recently, supposed to have been put at this risk. Therefore, one's relatives "formerly" used to cry when one went to such a discussion. One virtually risks one's head when going (‘driving off") to a brahmodya. In addition, such passages indicate that the situation of challenge persisted in or still was closely connected with ritual, even in such com— paratively late texts like the VadhB and JUB. Both the Sakalya legend and the Soma story (see above, § SB 3.6.1. 23) look like a satyakriyd, like wagering on part of the participants of a discussion viz. of a ritual. The custom of wagering one's head during a discussion lasted for much longer, in fact well into the first millenium A.D.; this can be gathered, for example, from a statement about Hsuan Tsang. He spared the life of a Brahmin who had staked his head in a dis— ; : 73 cussion with him, but then kept him as a servant. 73. See: Hui-li, The Life of Hiuen-tsiang, transl ; Delhi 1973, p. 161-65; cf. W. Slaje, Nalanda, int Grazer Morgenlandi- sche Studien 1, ed. H. D. Galter, Graz 1986; cf. also (ref. kindly provided by R. Thurman) the case of Dharmakirti, as an apprentice of Kundrila, in: Taranétha, Hist. of Buddhism, transl. by Chattopadhyaya, and also: Obermiller, Hist. of Buddh. by Bu-ston. by Samuel Beal, New398 Michael Witzel 7. Against this background, the only reported death resulting from a dis- putation, the one of Sakalya, gains a new significance. It is reported both in $B 11.6.3, JB 2.77, and then, slightly differently, in an Upanisad (BAU 3.9), a (comparatively) late text thus. The loser is no longer expect- ed to die after a discussion or during the ritual in this late-Vedic peri- od, just as one is not killed any more at such occasions in VadhB or JUB. However, this is exactly what is reported in all four versions of the story. Why then is Vidagdha Sakalya killed? Why did he not die on the spot, or rather, why was he not killed immediately? The introduction to the $B, JB and BAU stories tell that Janaka begins a "sacrifice with many gifts for the priests." The general context thus is the usual ritual one. One can therefore suppose that the loser, Vidagdha, would be killed and, that, perhaps, his head would be used in the ritual. Nothing of this is said, however, while on the other hand, VadhB, a comparatively late Anubrahnana and even the earlier $B, still speak of the heads of the give sacrificial animals (including man). Yet $akalya's bones are found in the wilderness (or are they robbed from a caravan?) Have they been cast away intentional- ly, just as Yajfiavalkya's threat implied: ‘Not even your bones will reach home' (for a proper burial?). Did Vidagdha die of fear, was he indeed kill- ed? Why did the robbers take his bones for something else?”4 Im the present case, I think, we will have to reckon with another factor, ~ that of literary composition and of legend forming.’> First of 74. Note that in the Vedic period described by Middle and Late Vedic texts, the burial of the ashes and bones took place, after cremation, in a low, square caitya, according to the classical Vedic érauta/grhya form and not in a round caitya which is described by SB 13.8.1.5 as eastern and ‘asurya'; note the burial in effigie of people who have disappeared or have died in a foreign land, BaudhPiS, 3,7-8, cf. Srautakosa, Engl. tr. p. 1117 sq.; for the ritual of cremation and burial, see Caland, Todtencult, § 4o § 91, saq., esp. § 77 and 112; for a dead body lying on a Smaédna cremation ground, half devoured by animals, see Chag.Up. 2. 75. This has, to some textent, been taken into account by Figer; however, he simply dismisses (p. 78-81) the whole story and its interesting details as being “of little value and credability [with regard to Yajfavalkya as an actual historical person] ... legend forming is already in full bloom; ... [it is written in] unusually pompous style, +++ an unfortunate mixture of the brdhmaya element [$B 11, discussion, and the Upanisadic curse [Satyakriya]", - with a curious misunderstand- ing of the satyakriya type of curse as a late development. But note already the use of a satyakriyd in RV 3.33 (crossing rivers), or cf+ AV 2.2 = PS 1.7 (used when playing dice, cf. with Katha 1.101).The Case of the Shattered Head 399 all, Yajflavalkya was, according to all accounts, an extraordinary person, as is obvious from the - comparatively - lively accounts in $B and also in JB, VadhB. Even if not all the quotations attributed to him are genuine, he was taught, at least by the compilers of $B, to have been a remarkable personality 76 According to the texts, he was not only a ritualist but also an Upani- sad thinker.’’ One of his characteristics, not treated extensively in 76. The final compilation of SB, made up of several independent portions, is probably a comparatively late one; yet the compiler was able still to put cross-references into the Vedic text: cf. the case of the Pra- vargya legend in $B (see author, on frame stories, in Festschrift U. Schneider): the compilator still knew Vedic well enough to produce such sentences referring forwards and backwards in the text. On the other hand: the compiler was different from the (much later) redactor who seems to have lived many generations after Yajfiavalkya, even ac— cording to the various Varigas found in $B and BAU. I suspect that he was a contemporary of the Kanva dynasty or the Satavahana dynasty. (This problem will have to be treated seperately). It is only the redactor that was responsible for glorification of Yajfiavalkya and for his authorship of the White YV; note that this information is added as the very last words of SB, - though still with Vedic accents (1, cf. ann. 97); note that the redactor already describes Janaka as present- ing land to Yajfiavalkya (see ann. 97, loo but cf. $B 13.2.4.2-4: boundaries of villages (grdma) ideally are contiguous). Yet even the Satakarni inscription, 2nd cent. A.D., (see ann. 82) still mentions only presents of cows given as dakgind to Brahmins, and not a donation of land; cf. however, W. Rau, Staat und Ges., p. 58 sq. For the person of Yajfiavalkya see now: I. Figer, AQ 45, 1984. - (Cf.also Ruben, ZDMG 83, cf. ann. 8. Actually, several others that I know of (F. Sprockhoff, J. Bronkhorst, myself) also wanted, in the early Eighties, to study the person of Yajfiavalkya, but having heard about Fi¥er's article, then planned, refrained from doing so. - I may, however, mention two points here which seem to be of importance to me: A close study of portions of the texts ascribed to Y. reveals, as for example in BAU 4.3, that the author of this short text is very innovative in his language: he coins many new words not found before him and often not found even in later literature. I investigated this when T. Vetter and I jointly conducted a seminar on Upanigads at Leiden, in 1978/79. The results of this will be published later. ~ I. Fier misunderstands the hapax problem: Comparatively infrequent attestation of a particular word of course docs not mean that the passage in question is necessarily late! - The question "One or two Yajfiavalkyas?" is not treated here (cf. Horsch, Gatha- u. Sloka-Lit., Pp. 474). Both the witty ritualist of the Br. portions and the thinker/mystic of the Up-s seem (to me) to reflect a single, extra~ ordinary personality. However, only a thourough comparison of the use of language in both the $B and BAU passages ascribed to Y. can decide the question.4oo Michael Witzel secondary literature, (with the recent exception of I. FiSer), is his wit: he usualy will give an unexpected, quick and undefeatable answer, such as the one to the Carakédhvaryu who told him not to perform a sacrifice in this or that way - otherwise his breaths would desert the body. Yajfiavalkya, who had followed the other procedure all his life, just pointed to his gray-haired arms saying "these old arms ~ what in the world has become of that Brahmin's words!" (SB 3.8.2.24 sq.) In the $B (11) and the JB variant of the Sakalya legend, he again shows his wit by the ploy of not giving a direct answer to the question asking whether he indeed is the most learned Brahmin: He simply says: na~ mo'stu brahmigthdya! go-kdmd eva vayar sma. “Reverence of the most learned one! We are only after cows," ($B, JB 2.76). What follows then is a simple questioning in two parts: the first one ending with Yajfavalkya's answer that there really is only one god, to which Sakalya agrees; the second part provides their identifications: the one god is "breath". Sakalya should have stopped before asking this last question and should have been content with the answer: "1-1/2 gods = Vayu". This simple narrative scheme has been elaborated in BAU 3.1-9 (SBM 14.6), as has been noted already by Deussen. It starts with the same intro- duction, be it somewhat more elaborate, but then, the Up. version inserts 8 questions by 7 other persons: Aévala, Artabhdga, Bhujyu, Usasta, Kahola, Gargi, Uddalaka, and again, Gargi. Of these, the stories of Bhujyu and Udddlaka are two slightly different versions of the same frame story, both with Western dialect characteristics that are umusual for BAU (cf. StII 7, p+ 114). The double questioning by Gargi and the confusion in Sakalya's questioning are also remarkable: The passage clearly is a patchwork, as has already been pointed out by Deussen. 7° These preliminary observations on the structure of the BAU version of the legend can be supplemented by others: The last sentences in BAU have ‘ 9 been remodelled into memorative verses, a clear sign of later redaction.’ The grammatically wrong form md vyapaptad instead of mi *vipaptad’? is 78. Cf. also Oldenberg, Dic Lehre der Upanischaden u. die Anfange des Buddhismus, Géttingen 1923, p. 133, and I. Fixer, AO 45. 79. This has already been noticed by Deussen, 66 Up.s; cf. also the trans- lation of Hume, The 13 Principal Up.s, repr. London 1971. 80. Or did one make a wrong dissolution of Sandhi: ma+paptad < mapaptad?The Case of the Shattered Head dot matched by a similar mistaken use of the moods in combination with mi in 8 comparatively late texts.”! Here, the situation asks for an imperative or rather, a prohibitive statement to be expressed by an injunctive; the form of the verb, however, seems to have been influenced by the parallel Passages which have: na ... vyapatigyat (see above, § 1: TB 3.10.9.5, ChU 13-14, 5.18, GB 1.3.14). The very occurrence of, or the possibility for naking such a fundamental mistake again is a sign of the redac- tors 's shaky grammatical knowlesde. He probably already spoke a form of post-Vedic Sanskrit and Prakrt.°2 Otherwise, the form vyapaptat seems to belong to a group of cases in late Vedic where the use of the augment deviates from normal Vedic: - Such as impf. used as a kind of plupft. in $B, see Caland, SBK, introa.83 = Or ChU vyapatigyat colloquial "would have burst" instead of “would burst" (cf. Hoffmann, Aufs. p. 370); here it merely seems to indicate an action taking place earlier than another.°4 Sl. See K. Hoffmann, Injunktiv, p. 94-98; note the similar case mi ... atyagad he adduces from Mbhar. 3,269.22 (N.B. also found at Ram. 1.71.1308*); he underlines the frequency of augment after the preverb vi, which could have resulted in a habitual use of augment in such situations; see also the “augmented aorists + mi" and other "remark- able augmented forms" (all after vy, abhy- at: van Daalen, Valmi- ki's Sanskrit, Leiden 1980, p. 83 (and compare further the augmentless imperfects of Ram.); cf. 0. v. Hiniiber, Notes on the preterite, MSS 36, 1977, p. 41, with materials on Epic and Pali. 82. For the period of the redactor of the Up., note: the teachers’ names in the Vafiéas ending in ~i~putra; such names do not occur in earlier Vedic texts but can be compared to the Pali name of Ajdtagatru: Aja- tasattu Vedehiputta at DN 2.1, etc. (for the date of the Pali sources, compilation in the 3rd cent. B.C., writing down during the period 80-77 B.C., see 0. v. Hiniber, Das altere Mittclindisch im Uberblick, p. 30); cf. also: J. A. van Velze, The names of persons in early Skt. lit., Diss. Utrecht 1938. Cf. also Horsch, As. Stud. 18-19, 227-246. 83. Cf. now, author, Festschrift U. Schneider, Freiburg 1987. 84. Cf. also those hypothetical statements in JB, noted by Oertel, where the first part of the sentence is missing: JB 1.234, 1.285 and 3.156 yat pratyavakgyat, 1, 262 yat pratyavakgyan, see Oertel, KZ 58, 1942, 80 sqq. - Cf. finally also the use of the Opt. as preterite in Middle Indian in non ~2- formations of the Optative; see 0. v. Hiniiber, e pret., MSS 36, p. 4o sq-, and the same, Uberblick, p. 181 sq., 193; cf. in general also Bechert, Uber den Gebrauch der indikati- vischen Tempora im Pali, MSS 3, 1958, 55 saa.402 Michael Witzel - One can now also add, in my view, the tendency of the diaskeuasists and of the Padapathakara, Sakalya, (see Hoffmann, Injunktiv p. 149 sq.) to "reconstruct" RV injunctives as pseudo-aorists, supplying them with an augment that had no business there. While the diaskeuasis was a long pro- cess, Sakalya lived, as will be seen immediately, during the late Brahmana period: He will have regarded the augment as an indication of past or pluperfect that had to be restored in the interpretation of RV verb forms. Finally, the list of the persons that question Yajfiavalkya in the BAU version is of great interest: 1. ASVALA, the hoty of King Janaka. (2, Jaratkarava Artabhaga) (3. Bhujyu Lahyayani: he travels in Madra) (4. Usasta Cakrayana) 5. Kahola KAUSITAKEYA (6. Gargi Vacaknavi: he travels in Madra) 7. Uddalaka Aruni (8. Gargi Vacaknavt) 9. Vidagdha SAKALYA The discussants, numbered 1, 5, 9 above, are of special interest as they all reflect names of Revedic schools: (1) ASvala represents the Agvalayana school (Aitareya-Br., Ar., and ASvalayana-Srautasitra); note that Agvalayana is the reputed author of AA 4 (while Saunaka is the one of AA 5, see Keith, tr. AA, p. 19 sq.)+ (4) KahoJa Kaugitaki is the reputed author of texts of another RV school, namely the Kaugitaki Br., Ar. (and $ss,)85 (7) The famous Uddalaka Arupi is often reported to be a Kuru-Paficala Brahmin, and to have travelled in the Madra land (Panjab). He stems fron the territory lying to the West of Videha, the home of the Vajasaneyi school, to which BAU belongs. 85. Cf. BSS 2.3 on the choosing of the priests: the Sadasya should be a Kaugitaka (gotra) Brahmin. The question is whether, at this compa- ratively early time, this refers to a gotra only, or whether a Vedic school (Sdkhd) was named after a particularly prominent gotra.The Case of the Shattered Head 403 (6/8) Gargi as representative of womanhood needs a special treatment. Her name makes her a member of the Garga clan: While Garga is not mentioned before the Sitras (ASS, SSS, KSS), his descendants, the Gargéh Pravareydh cceur, with other ancient ritualistic authorities, already at KS 13.12 (see StII lo, p, 232). Gargi therefore seems to represent the (originally) more Western schools like the one to which Uddalaka belonged.°° These representatives of rival Vedic schools have been portrayed as opposing Yajfavalkya of the jasaneyi school quite futilely. (Note that his name sometimes is Vajasaneya, as in JB and VadhB). In the list of oppo- nents, the heavy stress on the Rgveda schools surprises. One could equally have expected Sdmavedic and other Yajurvedic Sakhds. However, the threefold occurrence of the Revedic schools seems to have a purpose.” This can be best understood when studying the mentioning of Uddalaka, a famous representative of the rival YV school of the more Western peoples, the Kuru-Paficdla. He is quite prominent, and even occurs as a teacher of Yajfiavalkya in $B, but he also appears, in the texts of Samavedic priests who composed the Chandogya-Upanigad, as someone who is made to concede victory in a discussion to other Brahmins and to a certain yajaména, apparently a keatriya.®° In the present text he is shown as being less sophisticated in discussion than Ya jfiavalkya. As far as the Revedins of the list given above are concerned, it should be pointed out that SB seems to know a RV different from the Sakalya ver- sion, which is the only one that has survived to this day. For example, SB reports in the Purdravas legend that the Purdravas hymn (EV 10.95) has 15 verses. However the vulgate (the Sakalya version) has one of 18 verses. Sakalya, the "clever one" (vidagdha!) will have been opposed to this Easter 8. Note that there were many authorities on ritual in this family, see McDonnel-Keith, Ved. Ind. I 226; cf. also s.v. Gargi-putra, Girgya, Gareyayana, Gargyayani, and see ann. 82 on names in ~é-putra. 87. It could be that the author here intends to oppose the Upanigads be- longing to the Reveda, the Kaus.Up. in particular. It teaches the way to Brahman, cf. trans]. and discussion by P. Thieme in Wiss. Zeitschr. d. Martin-Luther-Univ. Halle, Jg. 1, 1951/2, Heft 3, p. 19 sqq. = Kl. Schr. p. 82-99. 88. See ann. 19; note also that a similar occasion occurs at BAU 4.1: Janaka mentions the opinions of various teachers, among them are fam— ous Satyakaéma Jabila and Vidagdha Sakalya; all these are refuted by Yajfiavalkya.dod, Michael Witzel version: it is he, (or perhaps already his school, the Sakalya group of Revedins), who invented the padapdtha and ordered the whole of the RV in his/their way, that means differentely from the one of the Eastern Rgvedins. Secondly, while the carly Brahmana text of the RV, i.e. Aitareya Br. 1-5, clearly belongs to the West, to an area overlapping with or close to that of the Katha school, (i.e. Eastern Panjab and W. Kuruksetra, see author, Fel. Vol. Eggermont), the later part of AB, with pavictkds (!) 6-8, knows about all of Northern Indian but is more familiar with the East and the South-East (even the area SE of the Vindhya, inhabited by "foreign" people (dasyu) like the Pundra, Sabara, Pulinda, Maitiba, up to Andhra, AB 7.18).89 some features of the language of AB 6-8 indicate as well,”° that there must have been a (sudden) movement of the Aitareyins towards the East, to the Videha, and perhaps, Magadha countries. This will have meant a strong rivalry between them, the representatives of the Kuru-Paficdla “ortho doxy" and Srauta orthopraxy on one hand, and the predominant school of the 91 East, the White YV (Vajasaneyin) on the other. BAU indicates this: ASvala already functions as the Hotr of king Jana~ ka; ive. he has an important position at the court. This is also reflected in the later part of AB itself: While pamictkde 1-5 deal almost exclusive- ly with the Soma ritual, the very last part of pamictkd 8 (AB 8.24-28) treats the office of the Purohita of a king in detail. Note also that AB 89. The last point may be indicative of a comparatively late redaction of AB 6-8: was it made under the early Magadha kingdom which had closer ties with the iron producing tribes of Chhota Nagpur? - A Magadha kingdom is nowhere mentioned in the Vedic texts, and it is only in KA (SX) that a magadhavdsin brdhmana occurs. The Pali texts, however, know of large Brahmin settlements in Magadha and Ahga. - It may very well be the case that the Vedic texts intentiona}ly do not mention the Kingdom of Magadha (note also the opposition of the Mbhar. towards Jarasandha of Magadha), as this was, after all, an area were Brahmins were not supposed to venture without the fear of losing their (ritual) purity; cf. R. Salomon, on the injunction against crossing the Karmand- $a River, the Western boundary of ancient Magadha, Ady. Libr. Bull. 42, 1978, p. 31-60. Qo. See author, in the proceedings of the Paris dialect conf. of Sept. 1986, to be edited by C. Caillat. 91. Nowadays, the YV usually is strongest im number among the varous Ve~ das, and followed by RV and only then by SV; the AV comes last, with very few adherants, But in Yajfiavalkya's time, the Aitareyins may have meant a serious threat; note their importance in royal ritual at AB 7-8.The Case of the Shattered Head 405 7-13-23 is a very detailed and long treatment of the Rajasiya and allied royal rituals (which in the parallel school of the Kausitakis were only incorporated into a later text, the $$S). The coming of the Aitareyins to the East set off an innovative trend, quite similar to the one instituted by the introduction of the Sandilya texts $B 6-10, and that of the Kanvas (SK), from the West. Unfortunately we de not know the exact time of these movenents. They all must have taken place in the late Brahmana period. In any case, already the authors and collectors of the JB and $B 11 (which contain the Sakalya story), and later on also the redactor of $B and BAU, must have intended to show the Revedins in a subordinate, if not a defeated position, as to contrast this with their own, extra-ordinary champion, Yajfiavalkya. He is, after all, regarded as the traditional "author" of the White YV; see the last sentence of the Vafisa at the end of $B and paU.° If thus, the hand of the authors/collectors of $B 11 and JB vas strongly involved in the formulation of the story, one can now ask the question: Is Sakalya's very name "Vidagdha" perhaps a double entendre, a lega? Vidagdha means, at the same time, "clever" (attested in Mbh.), but also "burnt up, cremated" (SB), or "decomposed ...". To use this mame may have been an indication, a warning of what would become of Sakalya already in the $B 11/JB story. The detrimental outcome is intentionally underlined in the later, i-e. the BAU, version of the story, and it therefore was conceived already by the redactor of BAU. This is also evident from the twofold insertion of the questioning by Gargi (BAU 3.6 and 3.8): She warns Vidagdha and the others, if Yajfiavalkya could answer her two questions (BAU 3.8.1) then certainly nobedy would be able to overcome him in a discussion. And again, at the end of the discussion, BRU 3.8.12, she says: "Brahmins, you must think highly of yourselves if you can get away from him by merely showing him your obeis— ance. Certainly, nobody of you will be able to conquer him in discussion." 92. Cf. I. Figer in AO 45, 1984; acc. to the Puranas, Sakalya swallowed the Veda which Yajfiavalkya had spit out again; this legend is compara— ble to the Tittiri story found only in the S. Indian "Chardi-Br." (of the Vaikhanasa school); the text was noticed by Caland, and actually is present in the Utrecht Univ. Library but has remained untreated so far.) Such legends apparently have their origin in the tradition on Vidagdha Sakalya.406 Michael Witzel The redactor of BAU succeeded in making his point by unusual means: It is a woman, Gargi, who is better than all the other Brahmin men that took part earlier in the discussion. Just as by the introduction into sone Upanisads of Ksatriyas and kings who occasionally know more about a particular ritual or philosophical topic than all the Brahmins, the eminent position of Gargi again underlines the still more extraordinary position of Yajfiavalkya. Her eminence is recognised by Yajfiavalkya himself vho, among all the other discussants, warns only her not to overask: The others had given up much earlier in the contest ...27 Everything therefore points to the narrational climax of Sakalya's persistent questions and to his subsequent defeat, in spite of Yajfiaval- kya's warning. In short, Sakalya's death in BAU is clearly intended as a climax, while in the older version (SB 11.6.3, JB 2.76-77), it is simply reported, like some of the other statements about ‘shattered heads’ in the Brahmanas (see above, VadhB 3.94, JB 2.203 sq., JUB 3.7-3. 11).94 Differently from such legends, however, all the Sakalya stories, as told in the BAU and $B, JB versions, set a date: Before the day "x", Sakalya will automatically die, most probably before the actual beginning of the sacrifice which Janaka is about to offer. The discussion among the contestants probably was intended by Janaka as a preliminary to his chosing the priests (révigvaraya) for the ritual he was going to undertake. The loser's, i.e. Sakalya's head might have been used for the ritual. His body, however, would then have been useless and therefore will have been cast away (as to fit Yajfiavalkya's threat); apparently it was devoured by wild animals: Robbers find only his bones and taken them not knowing that these are bones of someone brought to death.?> 93. The other persons next to Gargi are, with the exception of the Reve- dins and Uddalaka, not very prominent in the extent Vedic literature: Artabhaga, etc. 94. This outcome of a discussion is threatened to Yajfiavalkya himself at BAU 3.7.1 = SBM 14.6.7.1. 95. Cf. also the words of UrvaSi to Purdravas: "Do not die! Do not throw yourself (into a ravine: md prapaptah!), may the hostile wolves not devour you!" RV 10.95.15. For suicide in ancient India, see now: P.-A- Berglie and C. Suneson, Arhatschaft und Selbstmord, in: Kalydnamitra~ raganam, ed. by E. Kahrs, Oxford (NPU/OPU) 1986, p. 13 sqq. cf. also RV 10.67.5 - FiSer thinks Sakalya's bones had been put into an urn and then had been mistaken for treasures. This is also the later inter-The Case of the Shattered Head 407 What is significant in all of this material, is this: We have got here a late Brahmaya legend, sustained by many similar, even late accounts, of a brahmodya, a "theological" discussion, where one's head was at stake. Similarly, in certain ritual situations, one was apt to lose one's head - if one did not know the correct procedures. By the time of the early Upanigads and their (probably somewhat later) collection, or that of the much later redaction of the Pali texts, all of this is still remembered vividly; it was, perhaps, rarely actually practised, but was surviving as just another ritual "fossile" certainly and it therefore was frequently used as a threat in ritual and ‘theological’ discussion.°° pretation made by the commentaries; the prophecy, however, simply was: ‘Even your bones will not reach home ...'. How the bodies of decapitat- ed victims were treated by white YV priests is described by Heester- man, WEKS XI, p. 33, ann. dl a (i.e. throwing them away). 96. I am not of the opinion that the early Upanisads like BAU, ChU are necessarily contemporary with or even later than the Bud- dha. The Pali texts clearly indicate a much later and much more de- veloped society, with towns and with Brahmins prominently settled even in Magadha and Aiga. - This was either not the case yet at the time of the Vedic texts, or such passages were, (as fas as towns are concern- ed), intentionally excluded from the texts. Towns were densely settled, and mostly by non-Brahmin/Ksatriya groups; they represent a ritually undesirable and imperfect type of settlement. - On the other hand, there may be passages even in the early Up.s, - like the composite story of Vidagdha Saikalya in BAU, - which are of a later date. However, this is a topic that will have to be approached by a careful study of the language of the texts, passage by passage, and by a careful comparison of their contents and cultural setting with that of the Pali texts. Sweeping statements and superficial comparisons of two or three unrelated facts (Bronkhorst, Two traditions, on Svetake- tu, Stuttgart 1986) are misleading. To use, as an argument for chronology, e.g. the Vedic accents ("chan~ das" type language) referred to in Buddh. texts is besides the point, as accented texts were composed even much later than the (supposed) lifetime of the Buddha, see e.g. (almost unknown) Vaisnava hymns of the VaikhMp., or even the last sentences of SB, cf. am other hand, Katydyana (ca. joo B.C.), Vartt., already did not know the accents of Panini's original grammar, and had to infer them and to indicate them artificially (e.g. ddy-uddtta): this can be due to the break in tradition, but also to the distance of the North-Western home land of Panini and the Eastern one of Katy., who most probably had 2 tones only in his local form of Vedic Sanskrit: in the so-called bhagika accent of $B, see above ann. 76.408 Michael Witzel 8. Summing up the evidence presented here it can be said that asking too far, beyond the limits of one's own knowledge, and inquiring into sone thing one is not allowed to as (anatipraénya), necessarily leads to one's destruction. Secondly, the answer given to such questions, must be the "final" solution of the problem. A partial answer is not allowed, it will lead to one's destruction just as well as not answering at all; compare for exanple the ChU discussion on the sdman: If the opponent cannot answer the final question he is warned in time (ChU 1.10.9) and is destroyed if he does not pay heed to this. One can only avoid this fate by announcing one's defeat publicly and by becoming the pupil of the opponent (SB 11.5.3.13, ChU 9.12.2) .97 In the Buddhist texts, a quite similar type of questioning occurs, It is called sahadhammika; the rules are clearly specified:? answer at the third time the question is put (cf. M. Hara), - one must - one must answer completely, not only partially, - if one does not/cannot answer, death is imminent. The Pali texts indicate, perhaps more clearly than the elusive Vedic texts, that such questioning is a formal one: it takes place in a kind of open challenge or tournament, similar to the Vedic bralmodya. The Vedic texts sometimes stress the formal conditions as well: ChU 1.10.9 “after I had announced this to you ..." The clearest example is the JUB (3.7 sqq-) legend of Sudaksina Ksaimi who is nicknamed '$idraka' but challenges other Brahmins at a sacrifice. - An informal _ exchange of views does not lead to such dire results like the shattering of one's head. The whole situation involving this threat is described in both the Buddhist as well as the Brahmanical texts in an almost identical way: A superior being, ive., a Gandharva, viz., Rtu in JB, or a Yakkha in the Pali texts, appears. He is sometimes seen by the questioner. When the question 97. Note that this probably is the cause for a number of statements found already in the RV: to denounce someone is regarded as a severely evil action, especially if one accuses someone as a sorcerer, e.g+ RV Totody a shall kill such persons with his weapon (the vajra); RV 0-1.8.16. 98. On the sahadhanmika question, see Franke, transl. Dighanikaya, p- 94 Frage, bei der es auf die Tatsachen ankommt"; cf. M. Hara, Mittabi, see ann. 36.The Case of the Shattered Head 409 is not answered, the Gandharva viz. Yakkha splits, with a metal hammer wta JB, aya-kita DN), the head of the unfortunate person. The actual (ayaheF result of questioning, that is the death incurred, sometimes takes place seven (SuNi) or an unspecified number of days later ($B 11.6.3.11, JB 27 7-78). In most cases both the interrogation as well as its result, death, remain hypothetical (SB 11.4.1.9; DN 5.21, Mv.2, p. 114, SuNi 5.1.3). The threat alone suffices in most cases to deter from the actual confrontation. It remains to be aske: : Why is it that even a formal questioning, viz. @ speech contest holds such dire consequences for an unsuccessful partici- pant??? In preparing the way for an answer, it must be observed that the Vedic examples all deal with knowledge which is ‘secret! in one way or another; it may be known only to an eminent person, a teacher who will not pass it on readily even when he is questioned; or it is known to a class of ritual specialists who will not share their esoteric knowledge with rival groups. Such a case is described in $B 13.6.1.23 where lack of knowledge about the true nature of the sadas shed leads to the death of someone entering it unvittingly. Finally, wrong procedure during the Agvamedha automatically results in the splitting of one's head (or, at least in being robbed by the Ksatriyas taking part in the ritual, VadhB 3.79, p. 186). Under such circumstances, even an unwittingly wrong action, or an unsuspecting question automatically leads to one's death. This resembles very much the well known "automatic consequence" per- valent in Vedic magical thought: a given procedure must lead to a certain result. On the other hand, the Buddhist saccakiriyd, Skt. satyakriyd comes to one's mind readily. If a certain truth is stated, it will automatically yield the result desired by the speaker, as could be noticed in some of the 99. The winner, on the other hand, wins fame, one (or 100, looo, once 10.000) gold coins, and - perhaps - land (but cf. ann. 76 and 11), as seems to be given to avalkya at BAU 4.2.4, along with the servi- tude of Janaka and his people, or villages given by Janagruti to Raikva at ChU 4.2.4; cf. in Pali, DN 4.1: a village in Anga, (already regarded as part of Magadha!) had been given by Bimbisdra, the father of Ajatasattu, to a Brahmin, Sonadanga, with all rights (pertaining, such as in later in history, typically to an dgrahdra village?).flo Michael Witzel Pali texts quoted above, § 5. In the rather formal type of questioning found in the Vedic and Buddhist texts, the result is the automatic destruction of the unsuspecting, uncautious or proud challenger. It should not be forgotten that such questioning also involved the social status of the contestants. JUB 3.7.2 sqq. expressively states that such discussions were held only among the Brahmins and Ksatriyas (and Vai- Syas?), but not among the Sidras. Sudaksina Ksaimi who apparently could ask only simple questions about the ritual is therefore nicknamed 'Sidraka'. To challenge some Brahmin or Kgatriya in an open assembly (of the Kuru-Paficd- las) and to win the contest means a gain in social status; to lose, a loss of face. 1°° Apart from the discussion on ritualistic details the results of which are found, often with indications to their authors, in the Brahmanas and 1 Samhitas, these discussions’ deal, both among the Vedic brahmavddins as well as among the contemporaries of the Buddha, with some esoteric, secret a knowledge, be it about dtman, brahman or about the dhamma (or, simp! secret, as in the case of the origin of the clan of Ambattha which is known only to him and a few others), It is the (last) question(s) into the Ultimate which one has to avoid if one does not know the answer oneself: The question about the “ultimate deity" (devatd), or something irrefutable which is only known to the Buddha. As is well known, both conditions are also met with in the saccakirtyd: the truth stated or quoted must be unknown or irrefutable. Both the saccakiriyd and the anatipragnya / saha- dhanmika statements deal with truth, and both do so in a formalised context: either a discussion with a challenger and one or more opponents. This leads to a final point: given the similarity of both the ana tipraénya and the sahadhamnika questions and the general rules of dis- cussion found in the Vedic and Pali texts, it must be asked: how old are these rules and in what contexts could such rules for discussion arise? The anatiprasnya question of Middle and Late Vedic texts, clearly finds its Joo. References to this are already met with in the RV; one wishes to have a poet which wins a discussion. See above, ann. 21.The Case of the Shattered Head ay origin on the brahmodya riddles put at the sacrifices. The earliest cases are to be met with already in the Reveda (see above); later the questioning by riddles was formalised in the classical Vedic Asvamedha sacrifice and at the Sattras (where it could take place for 1o days, until the eleventh, the avivdkya, the "no discussion" day). !°! These so-called riddles, however, aim at stating ultimate truths, insights about the gods, the cosmos and man's position in it. It is the "work" of the “inspired poet", (vipra) who had to "fashion" (take) “new poems" about these truths, (See Geldner, HOS 36, p. 191 s.v. Neuheit, and RY 3.31.19, 10.23.6). They are formulated in such a way that they often remain enigmatic, as RV 7.87.4 itself reveals: "He who knows the track should tell (the secret 21 names of the cow, i.e. poetry,) like secrets if he wishes to serve as inspired poet to the later generation". !°? It is well known that, in the world view of the Vedic poet, only one quarter of speech was known to the uninspired men and was used in everyday language: three quarters remain hidden or secret (RV 1,164.45). Bternal truths are either told in poetic formulations (brahman, manmani RV 7.61.2) involving ‘secret! words, or they could be directly composed in riddle form. In the context of ritual, the riddle of the cut-off head of the sacrifice was one of the most important questions for the Brahmins which demanded ever-renewed answers. lol. Bodewitz, ITJ 16, p. 90 ann. 17, (repeated in "Agnihotra" p. 188 ann. 2) warns “against a confusion of all kinds of brahmodyas, against attributing aspects to late contexts which are foreign to them and against over-sensivitiveness with regard to chariots" [used in coming to and leaving from a brahmodya; cf. Heesterman in Pratidanam, p. 446]. He has not noticed that the "rather incidental brahmodyas of SI are not fundamentally different from the "Ancient Indian Verbal Con- test" of F. B. J. Kuiper, but merely their continuation, in a rel. gious and social context that has changed. Right down from the Revedic period, the brahmodyas were common also outside the New Year ritual, as could be seen above, for example in the Sattras. - Note that even very late texts like JUB and VadhB still show clearly that (a degree of) violence was maintained until the period of the authors or until very recently before that. For riddles, see TS 2.5.8.33 Ss 13.2.6.9, 13.5.2.11; 4-6.9.20; AB 5.25.22; VS 23.9-12, 45-62, and literature, at Ruben ZDMG 83, p. 247, Ann. 2. 102. vidudn paddsya gihyd nd vocad yugdya vtpra upardya étkean; see C. Watkins. Reacts tse Indo-European Poetics, in: The Indo-Europeans in the fourth and third millenia, ed. E. C. Polomé, Ann Arbor (Karoma) 1982, pp. lod-120.412 Michael Witzel It is here that the circle closes: sacrifice, which in the weltanschau- ung of the brahmins, both procudes and sustains the world (cf. RV 1.164.50 yajiiéna yajridm ayaganta devdk) is not possible without violence and destruction; even the gods did not know 'the head of the sacrifice! and had to learn about it from ‘outsiders', i.e. Dadhyafic Atharvana and the Advins. Only then they could succeed with their Sattra at Kuruksetra.!03 This myth is, in my opinion, the central one of Vedic Srauta "orthodoxy" but it has not yet received, also not in Heesterman's article on the "severed head", the attention it deserves: It serves as the justification and apology of ritual killing and at the same time, of the Brahmin's and especially the Adhvaryus' position in ritual (and society). All the five "sacrificial victims," i.e. man, horse, cow, sheep and goat could lose their lives in the ritual, until the first one or two on the list vere substituted by the rest; during the late Vedic period, even the rest of the sacrifical animals was substituted by figures made of flour (pipfapasu). Their heads, however, still were necessary for the building of the Agnicaya- na: they have to be interred in the altar.'°4 rt is only at this late point in time that in rituals such as the Agnicayana the priests have to resort to such measures as to seek out the head of someone killed by lightening or 103. Note that normally, this results in guilt which has to be passed on, for example to Trita Aptya; similarly, it is Thraétaona Athwya in Avesta, Yt. 8.33 who kills the "dragon". ~ However, in the case of the head of Dhadhyafic, there apparently is no guilt: It is not Dadhyafic's head but its replacement, that of a horse (a sacrificial animal), which is cut off: the slaughter of man / demi-god is substituted by that of an animal: Note that this substitution has become the standard myth, the apology and justification of all 'classi- cal! Srauta ritual, - a fact not noticed by the modern theoreticians of Srauta ritual. Cf., in general, Heesterman, Brahmin and renouncers The case of the severed head, WZKS XI, 1967, pp. 22-43. - Note also the case of the horse sacrifice in VadhB: a young pre-pubescent boy must be killed as atomement for the slaying of the horse which has spilled semen during its suffocation, see Falk, Bruderschaft, p- 160- = Guilt normally is passed on to one's neighbours, and then onvards, "to the farthest distance", see AV 5.22. PS (K) 13.1 = PS(Or) 12.1-23 BSS 2.5 (pdpmano vinidhyayah). lod. See for example TS 5.1.8.1: man's head is impure without the "breaths"; it is deposited on (?)/or replaced by an anthill which is pierced in7 places just like a human head (openings of the eyes, etc.) cf+ Heesterman, Severed head p. 39.The Case of the Shattered Head 413 in battle. Even in the ‘classical’ Vedic ritual therefore, violence is not completely excluded, though the texts mostly pass it over silently or hide it by euphemistic expression.'°> Against this background, the loss of one's head in a ritualistic discussion is not unlikely even during the late Vedic period. In fact, VadhB expressively states that such cases were common even three or four generation before the composition of the text. On the other hand, the severed ‘head of the sacrifice’ is nothing but a transposition into the sphere of the sacrifice (yajia) of the cutting off of the head of the puruga, be it a human or the divine one.!°° 9. The offering of the puruga has parallels in other Indo-European re- ligions, for example in the myth of the primordial Ymir and the crea— tion of the world on Old Norse mythology (Véluspa 3 sqq., Grimnismal 40-41) - which resembles the myth told on RV 10.90. Old Norse religion provides another parallel: that of the repeated questioning in riddle form about the origin of the world, of the gods and of man (Véluspa 21, 22, etc.; Wafthrudnismal 11 sqq.). This has, of course, been known since long, yet it acquires special importance in the present context: the very form of questioning, which is also met with in the questions of ZarathuStra put to Abura Mazdd in ¥.44, seems to reflect an old idea about ultimate truth: it is attainable only at the risk of, or - at least - at danger to one's life. Odin lost one eye in order to acquire secret knowledge (Véluspa 21-22), and Zarathustra had to face poverty and other hardships like emigration, in the 105. Cf. the study of d.labh by T. Goto and Oertel, Euphemismen, see ann. 54, Heestermann, WZKS XI. 106. The RV hints at the real procedure, in case of Vrtra: "The head is severed from the skin", 1o.172.2 sird 'va tvacdh. One may ask wheth- er this is not the actual procedure in all cases: Note that Vrta is, as Benveniste and Renou have shown long ago, the mythylogisation of vetra, the "resistance" of enemies (and demons); the Tndo-Iranians certainly knew about a "dragon" which was to be killed (ahi/a%i), but they did not yet call it Vrtra. (See now C. Watkins, How to kill a dragon in Indo-European, (Studies in Memories of W. Cowgill), Berlin- New York (de Gruyter) 1987). - In later texts, the splitting of the head has been reinterpreted: ChU 8.6.6 / KU 6.16 state that the dtman ascends, at the time of death, by a vein from the heart to the skull/head, and then to the sun; still later (cf. Laghusannyasa-Up. 24.1-25.2) one shatters the head of a deceased Sannyasin and then buries him. Sannydsins are not cremated, see Sprockhoff, Numen, te44 Michael Witzel course of his quest, both according to tradition and to the testimony of ¥.46.107 Vedic man tried to face these risks by creating poctry and, later on by ever new rituals: deep insight gained in poetic concentration or inspi ration, aided by the effects of Soma, lead to metrical poetic or to formalised prose statements (brahman). These were subsequently arranged in sets of single and multiple identifications between items of macrocosm, microcosm and sacrifice (yajvia), which in turn gave rise to ever-increasing strings and lists of an apparently infinite number, or to more complex, orderly structures like circles and frames, all of which are found in the YV-Sahitas and in the Brahmanas. Ultimately an immense web of cosmic and microcosmic interrelations was created that can now be found in the ad hoc discussion of the texts and that can and still has to be pieced together from the extant Brahmana portions of Vedic literature. This by its very nature open system vas constantly questioned, tested, and improved upon in speech contests and challenges, as well as in friendly discussions of Brahmins (and Kgatriyas). Ultimately, it became both the basis and model for the post-Vedic, still more formalised type of discussions and their pertinent rules. Consider, for example, the art of discussion found already in Patafijali's Mahabhigya (ca. 150 B.C.) with its standard pirvapakga, dedryadesiya, and stddhdnta setting, and compare also Caraka-Samhita 3.8.3 and Nyayasiitra 4.2.50. How long and how strongly this tradition survived can be witnessed even today, for example in the discussions of contemporary pandite or those of Tibetan lamas, which are accompanied by quite expressive gestures, (visible for example, in the film on the examinations in Tantra of the present Dalai Lama). Or, one can compare the statement of a comparatively 1964, p. 130 sqq. - For the whole problem, see Sprockhoff, Sannyasa, Pp. 62 saq.; cf. also RV 1.181.4 stimakha~ “having a good makha-", the name of one of the Agvins. 107. For a description and an evaluation of the state of the art with re- gard to Indo-European poetics, see C. Watkins, Aspects of Indo-Furo- pean Poetics, in: The Indo-Europeans in the fourth and third millenia, ed. E. C. Polomé, Ann Arbor (Karoma) 1982, pp. 104-120.The Case of the Shattered Head 415 late Dharmasiitra text, the Visnu Smrti, which is based on the lost Kathaka Dharma~Sdtra; the passage seems to reflect the earlier tradition of loss of life but it also softens the issue, in the tradition of Hsuan Tsang and Dharmakirti: "If one answers improperly or the other asks improperly, that one will perish or occur hate: (adharmena ca yah prdha yas eddharmena prochati tayor anyatarah praiti vidvegam vadhigacchati) Thus, in the Upanigad legends quoted at the outset, yet already in the Revedic passage 1o.88.17-19 mentioned above, we witness the beginnings of the ca. three millenia of the art of discussion in India, actually recorded in ritualistic, philosophical or scientific (édetra) texts in Sanskrit.