STUDIES IN YAMA AND M,~RA
ALEX WAYMAN Berkeley I. THE NAMES OF YAMA Ehni I points out that in the Vedas there is a word ydma (accent on initial) meaning "tamer, guider (especially of the carriage horse)", and another word yamd (accent on final) meaning "twin". This latter word is the name of the deity Yama. The Taittirfya San.~hitd gives this explanation :" "The gods and Yama were at strife over this world; Yama appropriated (ayuvata) the power and strength of the gods; therefore Yama has his name." The B.rhaddevatd has this? "He giving (prayachan) offspring here, and gathering (them) goes forth (to the other world): therefore the seer Yama calls him, the son of Vivasvat, Yama." In the epic, "He is called Yama, it being he who keeps mankind in check. TM Both the learned and popular idea of Yama among Hindus and Buddhists in India of the first millenium, A.D. is succinctly presented by one and one-half glokas of the Amarako~a (Svarga-varga, 53 and 54A of Bibl. Ind. edn.), in the form of fourteen names of Yama, for which both the Sanskrit and Tibetan equivalents are here given: King of the Law (dharmargja, chos kyi rgyal po) Lord of the Fathers (pit.rpati, pha giiu bdag) Impartial Judge (samavartin, mtshufis .hjug) King of the Dead (paretar~j, ggin gyi rgyal po) Ender (krtfinta, mthar byed) Brother of Yamun~ (yamun~-bhr~t.r, ya-mu-na yi spun) Destroyer (gamana, ~i byed) King of the Yamas (yamargj, ggin byed rgyal po) Yama (yama, g~in) Destructive Time (kftla, grofis can) Club Bearer (dan..dadhara, dbyug .hdzin) i 2 3 4
J. Ehni, Der Vedische Mythus des Yama (Strasburg, 1890), 41-2. ArthurBerriedaleKeith, The Veda of the Black Yajus School, HOS Vol. 18, p. 137. ArthurAnthony MacDonell, The Brhad-devat~, HOS Vol. 6, p. 43. V. Fausboll,Indian Mythology (London, 1902), p. 136.
STUDIES IN YAMA AND MARA
45
G o d of Rites for Deceased Relatives 0fftddhadeva, mtshun gyi lha) Son of Vivasvat (vaivasvata, fii mah.i bu) The End (antaka, mtha.h can) A longer list is included in a lexicon of perhaps the twelfth century, A.D. by the Buddhist *Sridharasena and available in Tibetan t r a n s l a t i o n ) The lexicon presents first the names of Y a m a (interlinear Tibetan note: " T h e Lord of the Southern District has twenty-six names"~). It is immediately apparent f r o m the Tibetan that this list overlaps the one above in these names: dharmardja, pit.rpati, yamardj, ~amana, yama, samavartin, paretardj, kt'tdnta, dan..dadhara, mivasvata, and grdddhadeva. Then it is transparent that antaka is present with alternate Tib. translation (byas mtha.h), and that T. grafts can is a textual corruption of grohs can, so that kdla is also present. Hence, this list omits only yamundbhrdt.r. However, it has the equivalent in the " B r o t h e r of K~dindi" (Kfilindi is a name of Yamun~), because T. lto boas ka-li-ndf necessarily translates kdlindisodara. In effect, the entire Amarako~a list is incorporated. The name kdlind?sodara is also found in H e m a c a n d r a ' s Abhidhdnacintdmat.~i (ed. by O. Boehtlingk and C. Rieu, 1847, p. 32). Other names f r o m our list overlap H e m a c a n d r a ' s names, and thereby the original Sanskrit is m a d e certain: " L o r d of the Southern District" (dak.sin.dgdpati, T. lho phyogs bdag po); " P l o u g h m a n " (k?nd~a, T. ki-nd-~a); 7 " W h o has the buffalo as his e m b l e m " (mahi.sadhvaja, T. ma-het.ff rgyal mtshan) ; "Robber" (hari, T. .hphrog byed); and " L o r d of the D e p a r t e d " (pretapati, T. yi dags bdag). A n o t h e r name is in H e m a c a n d r a ' s " A d d e n d u m " (ge.sa, verse 35), " W h o s e presence is ancient" (purdn. dnta, T. r~#i gnas). Two other names are in the Mahc~bh&ata: " W h o has the p u n i s h m e n t " 5 Dpal h.dzin sde's AbhidhanaJastravigvalocana(ityaparabhidhdnamuktavali)-ndma, Derge Tg., Sna tshogs, Po. 80a-3, f.://chos kyi rgyal po pha g~in bdag / .htshe ba.hi dbati phyug ggin rje .hi rgyal / ii byed g~in rje mtshutis .hjug daft / lho phyogs bdag po mthar byed daft / gratis can / byas mthah. / ki-n~-~a / ma-hehi rgyal mtshan [ chad pa .hdzin / kun h dod / rfiift gnas / g~in gyi rgyal / dus .hdzin / h.phrog byed / yi dags bdag / dbyug hdzin lto bcas ka-li-ndi / fii mah.i bu daft mtshun gyi lha / chad pa can daft dga.h byed do//.hdi yi churl ma dud pas hgebs//groti ni yafl dag .hgog ces bya//mdun pah.i mifl ni fii ma can//k~- .ni-tsi ni rnam spyod pa//pa-fitsa-ka-tva-gri-sa-nn~-li//sna tshogs sbas pa / yi ge pa//gYog ni / gtum po / gtum chen dag [/ lho phyogs bdag po la miti tier drug. 7 Cf. Manfred Mayrhofer, Kurzgefasstes etymologisches W6rterbuchdes Altindischen, I, 215; F. B. J. Kuiper, "Rigvedic Loanwords", Stud& Indologica (Festschrift far Willibald KirfeI) (1955), 155-6; H. W. Bailey, "Indo-Iranian Studies- III", Transactions of the Philological Society (1955), 68, where the solution is "ploughshare" (*kfna or *kind) plus "possessing" or "concerned with" (-dda). The difficulty of analyzing this term was doubtless felt as well by the Tibetan translator as by Western scholars: hence the transcription into Tibetan.
46
ALEX WAYMAN
(dan. din, T. chad pa can); 8 " W h o holds, or wields the punishment" (da.nd. adhdra, chad pa .hdzin). 9 Then, assuming the Tib..htshe ba!li dbali phyug is a textual corruption of .htsho ba.hi ~ it translates S. fivitega "Master of Livelihood". 1~ Various Sanskrit possibilities for two expressions did not lead to names of Y a m a : T. kun .hdod, probably "who has everything he wishes", T. dgat.~ byed, "gladdening". There is one name left in the list of twenty-six: T. dus 13dzin, which appears to mean "holding, or holder of time". Now, there is a name of Yama, kdlaku.ntha, in the lexicon Sabdakalpadruma? ~ Mayrhofer ~* gives the meaning "blunt, dull" for ku.nfha, but "Nicht sicher erklfirt". The Monier-Williams dictionary, p. 289, lists ku.ntha under the " r o o t " kun. th "to be lame or mutilated or blunted or dulled". These dictionary meanings convey no obvious sense in construing a compound of which the first member is kdla "time". The same dictionary lists under the passive participle kun. thita, besides the expected "blunted, dulled", meanings from Wilson, "grasped, held, encircled". If Wilson is right - but has his contributions to this dictionary ever been evaluated? - we would be justified in interpreting kdlakun, t.ha as "holder of time", which suits Yama, and then certainly this Sanskrit word is the original for the T. dus .hdzin. However, the latter conclusion cannot be accepted even tentatively, because further on we shall see that Y a m a has a name indicating that he is a cripple. Thus, it is conceivable that kdlakun, t.ha means "Time's Cripple", i.e. "Cripple throughout Time". Is Yama's Vedic foot-fetter (pad.bfda) as his own shackle as well? The hobbling of a horse was expressed bypad, bfda, ~a and further on Y a m a is shown to be associated, or identified with the sacrificial horse. The entry by * Sridharasena concludes with associated names, which are also in Hemacandra except for the "judgment seat" and the "register of human actions". His consort is Dhfimor.nfi. His city is Sa .myamana. in The name of his 8 S. SSrensen, Index to the Names in the Mah~bh~rata, p. 768. B6htlingk and Roth, Sanskrit W6rterbuch, Vol. 3, 496: da.nd.adhdrasya yamasya. ~o Monier~ Sanskrit-English Dictionary, p. 423, equates this word with jivitanatha and cites Raghuva.m~a xi, 20. ~ B6htlingk and Roth, Sanskrit W6rterbuch, Vol. 2, 252. ~ Op. cit., I, 225. ~3 A . A . MacDoneU, Vedic Mythology (Strassburg, 1897), p. 172.
~ MacDonell and Keith, Vedic Index, I, 42 and 470. ~5 This is the spelling in Fausboll (op. cit.), p. 136, where some names of Yama's "kingdomare found. Willibald Kirfel, Das Purd.na yore Weltgebiiude (Univ. Bonn, 1954), p. 303, also gives references to Sa.myamani. In the Tibetan translation of Candragomin's U.nddiv!'tti(Derge Tg., Sgra redo, Re, f. 121b-1,2) the transcribed word Marflka.h is glossed gain ~jel.ligroti ("city of Yama"); but no verification of this was noticed.
STUDIES IN YAMA AND MTURA
47
(pratfMra) is Vaidhyata. xB Kan.ici1~ is his judgment seat (vicdrabh~). PafijikfLis his register of human actions (agrasanAdhdng).18
doorkeeper
Citragupta is his scribe. His servants are Ca.nd.a and Mah~ca.n.da. Hemacandra has some names of Yama that are not included by *Sridharasena: "Death (personified)" (mrtyu); "Having shrivelled feet" (d?rn.dhghri) 19 (cf. the previous reference to Yama as a cripple); "Son of the Sun" (arkas~nu) - a substitute for Vaivasvata; and in his "Addendum" (~e.sa), "Foremost of Time" (kdlakfqa); "Great T r u t h " (mahdsaO~a); and "Born from the Yamugra" (yamugraja). Some other Sanskrit names were found by Scheftelowitz in late Brahmanical passages: 2~ "The indigocolored one" (n~la); "Righteousness (personified)" (dharma); "Who is victorious over death" (m!'tyum.jaya); Citra - an appearance of Yama ;21 Citragupta - identification with his scribe; "Coming from the Udumbara tree" (audumbara); "Coagulated Milk" (dadhna); "Wolf-bellied" 1G This particular entry especially demonstrates how the Tibetan text of the lexicon is often misleading or incomprehensible without the utilization, as has here been done, of both the Amarako~a (in the Skt.-Tib. Bibl. Ind. edn.) and a lexicon such as Hemacandra's. The Sarat Chandra Das Tibetan-English Dictionary, p. 1249, under ggin rje (Yama), repeats this entry as found in a native lexicon which incorporated material from the Tanjur lexica. The T. mdun pa is then translated as "minister", but the dictionary does not even list mdun pa in its ordinary place (which would be on p. 675), and the word for "minister" is there given as mdun na bdon. T. mdunpa means literally "the one in front" and this suits the usage "doorkeeper". Besides the difficulty with mdunpa, the T. ~i ma can "solar" would hardly suggest a Sanskrit original of vaidhyata. But since doubtless the Tib. did translate that Skt., we must consider the possibility that the doorkeeper here is Vi.s.nu, because the latter is stated to be the doorkeeper of the gods in J. Gonda, Aspects of Early Vis.t3uism (Utrecht, 1954), p. 93, and it is relevant to out further arguments that "Vi~gu is requested to guard the s o m a . . . "(ibid., p. 95). 7 This word, understood to be the original of the Tib. transcription kO-tff-tsi, is not given this explanation in the Monier-Williams Dictionary,, which, however, does record vicdrabht-~, the presumed original of T. rnam spyodpa. s An interlinear Tibetan note reads: dka.h .hgrel ni mchog dam pa sgrt~bs. This is an entirely inadequate translation, but facilitates the restoration, from the corrupt transcription, of the correct Sanskrit: pa~jikd tv agrasa~.ndhdni. This is because T. dkal.~ .hgrel is the standard translation of pa~jikd (as a type of commentary), and T. rnchog frequently translates agra. The rest follows easily because agrasa.mdhdnTis in the Monier-Williams Dict. The term occurs in B6htlingk and Roth, Vol. I, 42, from Puru.sottama's Trika.nd.aies.a, I, 1, 73. Furthermore, the corrupt Sanskrit transcription pa-~tsa-ka (S. pa~eaka) "group of five" may have given rise to a spurious name of Yama in the Das Dictionary, p. 1248, T. h~a pa.hi bdag "lord of the group of five." x9 This has the alternate form ~ir.napdda. B6htlingk and Roth, Vol. 7, cites Trikd.n-
d.age~a, I, 1, 71. 20 J. Scheftelowitz, ,,Die Zeit als Schicksalsgottheit in der indischen und iranischen Religion," Beitriige zur indischen Sprachwissenschaft und Religionsgeschichte, IV (Stuttgart, 1929), pp. 19, 29-30. He cites Baudh. Dharm. 2, 5, 9, 11, and the Utsarjanaprayoga. Cf. Mahfimahop~dhy~ya Pandurang Vaman Kane, History of Dharma~dstra, Vol. IV (Poona, 1953), p. 593. .z~ Cf. Emil Abegg, Der Prete.~'alpa des Garud.a-Pur~.na (Berlin, 1921), p. 54.
48
ALEX WAYMAN
(v.rkodara); "Destroyer of all Creatures" (sarvabhatak~ayakdra);2" "Standing at the head" (parame.st.hin); and "Who shatters" (vihant.r). Some of those names of Yama stem from the Vedic literature. In the RV, he is called by his patronymic Vaivasvata. The names "Brother of Yamunfi" and "Brother of K~lindi" derive later fi'om the RV story that Yama and Yami are twins. Yama has a name in the RV, " L o r d of Settlers" O,iJpati), 2~ and this seems consistent with the names "Ploughman" (k?ndga) and "Master of Livelihood" (jh, itega). " ' Y a m a hath given the settlement on earth (to this S a c r i f i c e r ) ; ' . . . ,,24 The RV legend that he was the first mortal seems behind the names "Foremost of Time" (kdlakat.a) and "Whose presence is ancient" (purdn. dnta). Also in the RV he is said to have chosen death, finding out the path for many, and from this tradition doubtless come such names as " L o r d of the Fathers" (pit!Tati), "King of the Dead" (paretardj), and " G o d of Rites for Deceased Relatives" (grdddhadeva)?"5 The AV concept of Yama and related deities is summarized by ShendeY We find here an increasing emphasis on his role as "Lord of the Fathers" and ruler of the other world. Since he is a stern ruler, we see the tradition that later gave rise to the all-important name "King of the Law" (dharmardja), as well as to the name "Impartial Judge" (samavartin). Death personified Onrtyu) is Yama's messenger (data), and he is more fearful than Yama; but since Yama is now sometimes identified with Mrtyu. Yama also becomes more fearful. MacDonell writes, 27 "In the later Sam.hiffts Yama is mentioned beside Antaka, the E n d e r . . . " This Antaka is one of Yama's names in the Amarako~a. Thus, by the time of the later Sam.hiffts he ig sufficiently fearfnl as to be called in the course of time the various names showing him as a destroyer. The spirit involved in identifying him with time is demonstrated by a passage in the Mahdbhdrata: 2s "Time, which brings to pass the destruction of all creatures, is passing on." However, the concept of time in the beautiful " H y m n to Time" in the Atharva-Veda has a different spirit. ~9 ,.2 Cf. the MBh. passage cited in note 28, below. ,:3 Wl'.en Yama is identified with Agni: cf. I~. Senart, Essai sur la Ldgende du Buddha (Paris, 1882), p. 191. '-'~ Satapatha Brahmal3a, tr. Julius Eggeling, SBE XLI (Oxford, 1894), III, p. 298. .z5 Vedic Mythology, 172, for the RV references. ~6 N . J . Shende, The Fotmdations o f the Atharva.nic Religion (Poona, 1929) 248-63. 27 Vedic Mythology, 172. '-~ Santi P., Crit. edn., 169, 1 : atikrdrnati kale 'smin sarvabhfttaks.ayavahe. 29 For example, Maurice Bloomfield, Hymns L?f the Atharva-Veda, SBE XLII ~,Oxford, 1897), pp. 224-5.
STUDIESIN YAMAAND I~1'.A
49
Regarding the name "Who has the buffalo as his emblem" (mah~s.adhvaja), Yama has the mount of buffalo in post-Christian Indian iconography, but in Vedic times he had a horse. Thus Bloomfield translates, 3~ "With the blood of the brown horse of Yama thou hast verily been sprinkled." Bergaigne 31 shows that Yama (among others) is identified with the sacrificial horse. The names "Coming from the Udumbara tree" (audumbara) and "Coagulated Milk" (dadhna) must be discussed together. Bloomfield observes :32 " A tree as the seat of the gods occurs in RV. X, 135, 1, ydsmin vrk.sd supald~e devai.h sampibate yamd.h, 'the tree of beautiful foliage within which Yama drinks with the gods.'" Eggeling translates, 33 "They [the gods] said, 'Come, let us lay into the Udumbara tree whatever pith, whatever vital sap there is in these trees : ' . . . hence that (tree) is always moist, always full of milky s a p . . . " For the latter phrase, the ~Br. (M~dhyandina, edn. A. Weber) VI, 6, 3, 3, has tasmdt sa sarvadardra.h sarvadd k.sM. The word k.sMn means it possesses k.sira 'thickened milk'. The word dadhi ("coagulated milk") is an explanation of soma in Kau. Br. 3~ Coomaraswamy states, 3a "There are in fact many texts identifying the soma with the essence in the waters, sap in trees, and seed in man and animals." He goes on to give examples, one of which is especially pertinent now: "In AV., X/X, 31, 12, an amulet of udumbara wood is called virile (vr.san)." An amulet of that wood answers to the name "Coming from the Udumbara tree". Consequently, this name shows Yama in the aspect of virility. Moreover, the foregoing suggests that the milky sap of the Udumbara tree symbolizes semen or its vital essence. Yama, in order to be the first human, must start out this way. It is necessary to discuss, even if briefly, the name " G o d of Rites for Deceased Relatives" (gr6ddhadeva). Kane has recently provided an indispensable coverage of the subject. ~6 He quotes 37 an early explanation that the gods went to heaven owing to sacrifices, and that the men who duplicate those sacrifices will come to dwell with those gods and Brahman. It is also held in Hinduism that the ftavour of the sacrificial material pleases the pitrs and that they respond with benefits to men. 38 The factor of the so Ibid., p. 21. sx A. Bergaigne, La Religion Fddique (Paris, 1878 and 1883), I, 274. ~ Op. cit., p. 416. 3a ~atapatha Brdhma~.la, III, p. 256, 7. 34 Ha.msarfija, Vaidikako.sa.h (Lahore, 1926), p. 203: somo vai dadhi. 35 Ananda K. Coomaraswamy, Yaks.as, Part H(Wash., D.C., 1931), p. 21. 8s The work cited in note 20, above. 87 Ibid., p. 349. 38 Ibid., p. 335. Ludwik Sternbach, reviewingthis volume of Kane's work in JAOS,
50
ALnX WAVMAN
merit of nourishing the pretas seems to be the main reason for the Buddhist offerings, aa The basic point in the pre-Buddhist period is that the Fathers (pitt) dwell in Y a m a ' s Heaven, and a dead m a n may or may not arrive at that Heaven to be one of the Fathers. Since Y a m a first showed the way, he must have been the first human to perform the sacrifice. 4~ His brown horse was the external sacrifice. He himself was the personal sacrifice. But it was also held that the state of mind of a dying man determines his destination after death. 41 It appears that opponents of ritualism, i.e. the Upani.sadic sages and the Buddhists, stressed the latter factor. Of course, a devotee of Sr~ddha could maintain a consistency of the two factors by saying that the person who performs the rites will naturally have the proper state of mind as he dies. The name ~rdddhadeva also settles one point: Y a m a is called a deva, n.b. in connection with certain rites. He is also the deva in the Devadf~tasutta, as shown by Mus. 4'z The MBh. names of Yama, deve~a and sureia, 43 both meaning "master of the gods", show that Y a m a consorts freely with the gods, but do not necessarily show that he is one of them. Certainly Bergaigne is right in considering Y a m a "un personnage de nature 6quivoque, tant6t dieu, tant6t homme". 4~ But since the latter judgment was made for the Vedic literature, it probably involves an inference; as Apte says for the .Rg,45 " I n the case of Yama, for example, it is only implied that he is a dev6!" The name " L o r d of the Departed" (pretapati) concerns a subject much treated in Buddhist texts. Oriental art shows the pretas as frustrated spirits wandering among men unseen. They dwell in the 'Realm of Desire' (kdma-dhdtu) without the means of gratifying their desire, and so usually the word preta is translated (inexactly) "hungry ghost". Mus gives a comprehensive treatment based on the verses of Dharmika Subhati. 46 Another treatment is in the ,ffrya-Saddharmasm.rtyupasthdna, exposed briefly by Lin. 47 It so happens that the present writer investigated the latter Buddhist source independently of Lin's work, learning about the 74 (Oct.-Dec. 1954), p. 272, mentions the view of W. Caland "that the motive for ~raddha to the dead was merely the fear of the dead (preta)... ". so Paul Mus, La Lumi~re sur les Six Voles (Institut d'Ethnologie, XXXV) (Paris, 1939), 250, f. (the note to verse 49). 40 This agrees with K. F. Geldner, Der Rig- Veda, III, HOS Vol. 35, p. 157, note to 5d. 41 Kane (op. cit.), 185 f.; Franklin Edgerton, "The Hour of Death", Annals of the Bhandarkar Institute, Vol. VIII (1927), 219-49. 4~ Op. cit., p. 68, f. 43 S6rensen (op. cit.), p. 768. 4~ Op. cit., I, 111; cf. also II, 96. 4~ V. M. Apte, "Varu.na in the l~.gveda", NIA, Vol. VIII (July-Dec. 1946), p. 145. 4e Op. cit., pp. 248-61. 4~ Lin Li-kouang, L'Aide-M~moire de la Vraie Loi (Paris, 1949), 16-8.
STUDIES IN YAMA AND M.TtRA
51
sf~tra list f r o m a native T i b e t a n abhidharma work, then l o c a t i n g it in the satra itself, ascertaining that the Sino-Japanese translations preserve m o s t o f the Sanskrit n a m e s in t r a n s c r i p t i o n ; a n d then a t t e m p t i n g to restore these names in Sanskrit b y utilizing the J a p a n e s e r e a d i n g o f the Chinese l o g o g r a m transcriptions, the T i b e t a n t r a n s l a t i o n o f the names, a n d the contexts. 4s This p r i o r study e n a b l e d the writer to m a i n t a i n c o m p l e t e l y i n d e p e n d e n t j u d g m e n t a n d later to r e a d Lin's exposition s o m e w h a t critically. I n the following, after the English t r a n s l a t i o n the parenthesis will include first the r e s t o r e d Sanskrit, a n d next the J a p a n e s e r e a d i n g in those cases where the s o l u t i o n differs f r o m Lin's. A l s o it m u s t be said t h a t his solutions were i n v a l u a b l e in a n u m b e r o f instances. 1. " L i m b l e s s - t r u n k P r e t a " (*kap6li- o r *kabandha-preta, kabari), 4a 2. " N e e d l e - m o u t h P . " (s~cimukha-~ 3. " V o m i t - e a t e r P . " (vdntabhak.sa-~ 4. " O r d u r e - e a t e r P . " (vi.st.hdbhak.sa-~ 5. " F o o d l e s s P . "
(abhak.sa-~
6. " O d o r - e a t e r P . " (gandhabhak.sa -~ = gandharva), " D o c t r i n e - e a t e r P . " (dharmabhak.sa-~ 8. " W a t e r - e a t e r P . " (salilabhak.sa-~ 5~ 9. " H o p e f u l P . " (d~aka -~ ashaka), 51 10. " S p i t t l e 7.
eater P . " (khet.abhak.sa-~ 11. " G a r l a n d - e a t e r P . " (mdldbhak.sa-~ 12. " B l o o d - e a t e r P . " (raktabhak.sa -~ rakitta), 13. " F l e s h - e a t e r P . " (mdm. sabhak.sa -~ = pi~dca), 14. " I n c e n s e - e a t e r P . " (sugandhabhak.sa -~ 15. " M a l e v o l e n t - c o n d u c t P . " (abhicdra -~ abhishara), 52 16. "P.
~8 The beginning ~ the preta secti~ in the Tibetan translati~ is in Derge Kg" Md~ Vol. Ya, f. 284b-6: de n a s . . . The section ends f. 313b-1. Most of the space is given to description of the 36 families, the order of which is followed in my numbering, in agreement with Lin. Preceding the individual descriptions is a list of the 36, with slight irregularities in the order and some alternate misleading translations of terms. The Japanese text was made accessible by my wife Hideko, who also makes possible the other Sino-Japanese references in this article. The Japanese translation employed is in the Kokuyaku lssaiky~, 1st Set., Ky6-shfi-bu, Vol. 8, beginning p. 292 (Japanese numerals). ~9 Lin was justified in restoring the Sanskrit as kapali because after the transcription the word is translated by the logogram meaning the 3-legged vessel. The subsequent description of the particular preta family is of the preta with no head, arms, or legs, for which one would expect the Sanskrit kabandha. The Tib. has lto gug "prostrate belly" which suits either kapali or kabandha. 50 Salila must be the word, verified by the description, which usually involves river water, or running water. However, the Japanese reading bariran is justified by the TaishO Vol. 17, p. 92b-1 ba graph. Lin presumably and justifiably took this as a graphical error for a similar graph (Ueda Daifiten No. 10717, pronounced sa), but a note by him would have been helpful. ~1 However, adaka is given by Lin in a note with a question mark. s2 The original Sanskrit is made certain by Tib. drag gul spyod pa = abhicara. Moreover, this completely fits the description of the preta, who is bent on bringing harm to people and giving them nightmares. The source of the mistranslation into
52
ALEX WAYMAN looking for the opportunity" (avaMrapreks.in, tendara), ~3 17. "Underworld P." (pdtdla-~ 18. "P. of great magical power" (.rddhi-~ = maharddhika), 19. "Blazing P. . (jvahta. . . . ,jabart), . . 5~20. "P. looking for the opportunity regarding human infant" (*manu.syaputraavaMraprek.sin, tendara), ~ 21. "Having attractive b o d y " (kamarapin, kama), 22. "Seashore P." (*samanantaradv~pa -~ or *samudraffra -~ sammudarateiba), 23. " Y a m a policeman" (*yamadan. .din = yamardks.asa, enra), 24. "Child-eater" (bdlabhak.sa), 25. " ( H u m a n ) vital heat eater P." (u.smabhaks.a-~ 26. "Brahman demon" (brahmardk.sasa), 27. " H e a r t h P." (kun..da-~ 28. "P. of unpleasant street" (a~ubharathyd-~ 29. "Wind-eater P." (vdyubhak.sa-~ 30. "Ember-eater P." (a~gdrabhak.sa-~ 31. "Poison-eater P." (vi.sabhak.sa-~ 32. "P. of forest" (at.av~-~ 33. "P. of charnel ground" (gmag~na-~ 34. "P. of tree" (v.rk.sa-~ 35. "P. of crossroad" (catu.spatha-~ 36. " M e m b e r of Mhra group" (m~rakdyika, maragaya)3 s
Because of some relevance to what will follow in this or a subsequent section, three of the families should be described - only essential points of the text will be given: No. 18 " O f Great Magical Power" is god of the evil spirits and lives on a lofty mountain or at the seashore; alone has supreme pleasure; is surrounded by innumerable suffering pretas who observe his pleasure. No. 23 " Y a m a policeman" is ordered by Y a m a to record wrongdoings of persons; of fearsome aspect, he ties up the deceased wrongdoer and drags him to Yama's palace for sentencing. No. 36 " M e m b e r of M~ra group" is tortured in the three evil destinies (durgati); especially visits bhik.sus at training, mealtime and meditation periods, attempting to frighten them with evil voice and bring them nightmares. While the text here gives no hint that thepreta " O f great magical power" is Y a m a himself, this preta is outstanding as being the only one who enjoys himself. Mus 57 gives information that Y a m a is the variety of preta maharddhika ( " o f great magical power"). The fact that this is a family of Chinese, "marchant vite", is probably the fact that the final statement, "That is why he is called *abhicarapreta", is immediately preceded by a reference to the preta's fastgoing, but that final statement referred to the entire description of the preta. sa The Skt. is decided from the Tib. glags lta ha. Lin's cettara-preta cannot be defended. 54 However, the T. mtsan mo lus rab tu hbar ba "Whose body blazes at night" suggests a longer Skt. form, perhaps ratrijvalapreta. 55 Tib. mi rnams kyi bus pa la glags lta ba suggests the Skt. given. This preta seeks to take the life of a human infant as soon as it is born. so Tib. is bdud kyi sde. ~7 Op. cir., p. 286, note.
STUDIES IN YAMA AND M.~RA
53
pretas is also consistent with the multiplicity of yamas implied by the name " K i n g o f the Y a m a s " (yamardj). Mus 5s explains this multiplicity in connection with the different hells. So far the writer cannot determine if the ~frya-Saddharmasm.rtyupasthdna regards Y a m a as a maharddhika-
preta. Concerning the messengers, D k o n m c h o g .hjigs reed dbafl po writes in his Rin po che.hi gru gzihs, based on Vasubandhu's Abhidharmakoga, as follows :59 Now, are the 'Policemen of Y a m a ' (yamardk.sasa) sentient beings (sattva, sems can) or not? 6~ The Vaibhfi.sika maintain they are persons (pudgala). ~1 The Sautr~mtika maintain they are unconscious substance (or, "soulless matter") (jdd.a, hem po) produced f r o m the differentiations o f elements (bhata) and their derivatives (bhautika). Both the Mfidhyamika and the Cittamfitra maintain the same as the latter. All [four doctrinal schools (siddhdnta) o f Buddhism] maintain that the Y a m a o f Hell is a person (pudgala). The description o f these "Policemen of Y a m a " (tragically modern) leads to consideration o f Y a m a ' s two servants, Can..da and Mahfican..da. The policemen ("die Schergen") Pracan. d.a, Ca.n.daka, a m o n g others unnamed, have in the Garud.a-Purdn, a the function of Y a m a ' s policemen. 6z It appears that Y a m a ' s servants are the Epic substitution for the two dogs of Yama, sons o f Saramfi (sdrameya), his regular messengers in the Vedas. " D e lighting in lives (asut.rp) they watch men and wander about a m o n g the peoples as Y a m a ' s m e s s e n g e r s . . . Their functions therefore seem to
58 Ibid., pp. 307-9. s9 Dam pa.hi chos m~on pa mdzod kyi don legs par bgad pa rin po che.higru gzi~s, III, 33b-4, f.: / ho na dmyal ba.hi bsrufl ma de dag seres can yin nam rain 2e na / bye brag smra ba ni gaff zag tu .hdod cifi / mdo sde pas .hbyufi ba da6 h.byufi .hgyur gyi khyad par las grub pahi bern por h.dod do / dbu sems gfiis kya6 de daft h.dra / dmyal bah.i g~in rje ni gaff zag yin par thams cad kyis h.dod do / The author of this abhidharma commentary is considered to be the rebirth of the Jam-yafl-zhad-pa referred to by Th. Stcherbatsky, Buddhist Logic (photomechanic reprint, 's-Gravenhage, 1958), Vol. I, p. 57. Accordingly he is called the second Jam-yafl-zhad-pa. The "whole library of works" mentioned by Stcherbatsky may well have been written in part by Dkon mohog h.jigs med dbafi po.
s0 Cf. Mus (op. cit.), "Les Gardes infernaux sont-ils des 6tres?" (p. 209-11); and Abhidharma-koga, tr., Chap. III, p. 152-3. 81 This statement shows that the /[rya-Saddharmasm.rtyupasthana is a Vaibb~.sika work, since in its standard way, it gives a presumed offense as a man leading to birth among the pretas as a policeman of Yama. "Vaibh~.sika', however, can mean various "Hinay~ma" schools. This agrees with the subtitle of Lin Li-Kouang's work, "Recherches sur un Sfitra D6velopp6 du Petit V6hicule". e~ Abegg, op. cit., p. 63.
54
ALEX WAYMAN
consist in tracking out among men those who are to die, and in keeping guard on the path over those who enter the realm of Yama. ''6a Bloomfield writes, n* "The 'four-eyed bitch' is Saramft the mother of the two four-eyed dogs of Yama, ~y~ma and ~abala, which I have explained as the sun and the moon; see Journ. Amer. Or. Soc. XV, 163 ft." Hariyappa makes forceful reply, 6~ " I f the hymns contemplated any mythical motive that the Hounds of Heaven are the Sun and the Moon, or that Saramgt is the Storm-Goddess or Vasi.st.ha is the Sun, one wonders why the Veda would not state it; what harm? On the other hand, what harm is there to believe that there were two real hounds in the service of Y a m a ; . . . " It seems to the writer that the solution lies in the connotation of the word for "dog" in a particular culture. Even today in Japan the word for "dog", inu, is used disparagingly to refer to a "spy" or "secret police". Because dogs are noted for ability to track out by scent, a person of analogous occupation might reasonably come to be called, at least colloquially, a "dog". The same word could be employed metaphorically for a function. In terms of word usage, we can arrive at a position midway between Bloomfield and Hariyappa. Of course there were two real hounds in the service of Yama, but this is only true mythologically. Likewise, in accordance with the well-known Buddhist doctrine of karma, the Sautr~ntika, M~dhyamika, and Cittam~ttra are of course indicating that our own wrong actions become the "policeman" (in psychological terms, the "superego" or punitive self) that drags us away to retribution. Vasubandhu tells us in de La Vall6e Poussin's translation :6e "Les seize enfers sont crdes par la force des acres de tousles a t r e s . . . " What is called the Vaibhfts.ika viewpoint here is merely the popular objective representation. Modern psychologists know that we constantly personify or dramatize in dream life and while in that state believe in the objectivity of the representations. Perhaps in this way "mythologies" are created. The name "death personified" (mrtyu) of Hemacandra's list is sometimes identified with Yama (Dharma) in the Mah~bh~rata. 6v Still, not only the Amarakoia, but also the more comprehensive list by the Buddhist *gridharasena, fails to include it. The expression is frequent in Buddhist passages of an admonishing type, of which a number are collected in 6a Vedic Mythology, 173. 64 Op. cit., p. 404. He has a note to the word "Yama", " I n RV. I, 29, 3 the two messengers (dogs) of Yama are personified as females." e6 H . J . Hariyappa, "~,gvedic Legends through the Ages", Bulletin of the Deccan College Research Institute, Vol. XI (March, 1951), p. 142. 65 Chap. III, p. 155. 67 S~rensen (op. cit.), p. 489.
STUDIES IN YAMA AND M~.RA
55
Tsofi-kha-pa's Lam rim chen mo in the section devoted to meditation on suffering (du.hkha). For example, the Kani.skalekha of Matrcet.a has this verse, translated by Thomas :68 "60. Uncompassionate, the lord of death slays accomplished persons without reason: with the slayer close at hand, what wise man busies himself with cherishing vanities?" The Uddnavarga has this, translated by Rockhill: 69 "I, 17. As a cowherd with his staff gathers his cattle into the stable, so disease and old age bring mankind to the lord of death." As will be shown in a later section, mrtyu is one of the four kinds of M~ra, or metaphorical death. M~ra has built up an evil connotation in Buddhist literature, while Yama is "King of the Law" (dharmaraja), a model of justice. Perhaps for this reason, some sections of Indian Buddhists did not follow the general Indian identification of M.rtyu with Yama. But in the Devadutasutta, Yama is the Deva and m.rtyu. 7~ Shende tells us from the AV: 71 "The gods, like men, were mortals first, and as such were subjected to M.rtyu. But they overcame death by means of celibacy and penance (11.5.19)." As will be seen, likewise the Buddhist ascetic aim is to overcome the M.rtyu-m~ra. If this implies overcoming Yama, it must be the Yama of the Atharva- Veda tradition. Logically the R.g-Veda Yama is not to be overcome, because, as Kane writes, 72 "In .Rg. IX. 113.8 the poet prays 'make me immortal (in heaven) where dwells king Vaivasvata, where the sun is confined (i.e. where it never sets) and where the divine waters flow'." As " L o r d of the Fathers" (pit.rpati), Yama rules the men who succeed in reaching the world in the Intermediate Space (antarik.sa) between earth and heaven 7z - presumably what is later called in Buddhism the "Realm of Form" (rapa-dhdtu) between the "Realm of Desire" (kdmadhatu) and "Formless Realm" (arf~pa-dhdtu). The pit.rs have an extensive literature, not always consistent, and here only a brief presentation may be made. Kane points out that the Fathers (pitr) called Barhi.sada.h (who sit on ku~a grass) and the Agni.sv~tfft .h (tasted or licked by Agni) are mentioned in .Rg. X. 15.3-4, 11. 74 He says, 75"In V~yu 72.1 and 73.60, Brahrn~.nd. a as F.W. Thomas in Indian Antiquary, 32 (1903). n9 W. Woodville Rockhill, Udanavarga (London, 1892), p. 4. N. P. Chakravarti, L'Udanavarga Sanskrit, Tome Premier (Paris, 1930), p. 3, Sanskritizes the Tibetan of the second haft as follows: evanj vyadhig ca jara ca janan mrtyurajasakaga.m nayati. The Tibetan here does not agree with extant P~li and Prakrit versions. 70 Mus, p. 68, note, already cited above (note 42). 71 Shende,p. 253. 72 Kane, IV, p. 156. 7a Cf. Kane, 342-3. 74 Ibid., 194-5 and 201. 75 Ibid., 344.
56
ALEX WAYMAN
(Upodhghata 9.53), Padma V. 9. 2-3, Vi.s.nudharmottara I. 138. 2-3 and other Puran.as the classes of pitrs are said to be seven, three of which are formless (amartimat) and four have forms (mftrtimat) and they and their offspring are described in detail." Presumably the Agni.svattft.h would be the formless (or, "incorporeal") variety since they have been "licked by fire". The ones that sit on kuda grass obviously have not yet been licked by fire. Giving .Rg-V. references, Kane says, 76 "The pit.rs are often said to regale themselves in the company of gods, particularly of Yama . . . . The pit.rs are said to be fond of Soma drink . . . . Fire is supposed to take the spirit of a cremated person to the pitys . . . . It was supposed that the departed spirit, after the cremation of the body, was endowed with an ethereal body and became associated with Yama, the gatherer of departed men, and with pitrs." Again, 77 "The Baud. Dh. S. II.8.14 summarizes a brahma.na text stating that pit.rs move about in the form of birds. The Au~anasa-sm.rti and Devala quoted by the Kalpataru say the same thing. In the Vayu-pura.na it is stated that at the time of ~raddha the ancestors enter the brahman.as (invited) after assuming an aerial form . . . . " The Skt. passage quoted for the last statement shows that vdyubhftta is the original for "assuming an aerial form". Basing himself on the Matsya Puran.a, Kantawala writes, 7s "Somapa (lit. drinkers of Soma) pit.rs of Vedic antiquity, the progeny of Svadha and the residents of the M~nasa region situated above the universe, are eternal and the embodiment of D h a r m a par excellence and are held to be higher than BrahmL Being proficient in yoga and having attained Brahmahood and having discharged the work of creation etc. one and all, they reside in the M~nasa." This may be a good place to touch upon the procedure in these studies of utilizing pre-Buddhistic, Hindu, and Buddhist sources. Whether or not it be easier, it is usually safer, but certainly less illuminating, to stay within a single tradition and thus present an appearance of consistency. Both the Hindus and Buddhists have preserved an enormous literature of Indian legend, practice, and philosophy. Sometimes the same words are used with different meaning. Sometimes different words are used with similar meaning. Disagreements are therefore sometimes definite (even if not obvious) and sometimes seeming (even if not so). When people have the will to disagree, they surely can disagree. Yet, in that disagreement, 78 Ibid., 341-2. 77 Ibid., 339--40.
vs S. G. Kantawala, "The Cult of Manes as Depicted in the Matsya Purataa", Journal of the Oriental Institute, Baroda, Vol. V, (July 1956), p. 407.
STUDIES IN YAMA AND MARA
57
each side preserves the other to some implicit e x t e n t - trying unsuccessfully to "forget" the opponent. The Buddhist texts - with which the writer is most familiar - do not use the "father" (pit.r) terminology as exposed above. The Northern Buddhists have a heaven for the faithful called the "Western Paradise", or the " H a p p y Land" (sukhdvat~), ruled by Amifftbha, whose name "Immeasur~ able Light" reveals his solar nature; and recalls King Vaivasvata's heaven "where the sun is confined". On the other hand, the Buddhist Y~ma Heaven in the Kdma-dMtu has no apparent connection with Yama. TM In Abhidharma Buddhism he is the King of the Pretas and his capital is 500 fathoms beneath Jambudvipa. 8~ Buddhist texts have much to say about the dhydna heavens in the "Realm of F o r m " (rapa-dhdtu), which is the Buddhist middle region and preeminently the realm of yogic attainment. In some Buddhist schools it is taught that one becomes a complete Buddha in this " R e a l m of F o r m " as the Sa .mbhoga-kaya Vairocana, another solar name, 81 but this Vairocana is by no means associated with Y a m a in Buddhism. The name " K i n g of the Law" (dharmardja) will be the last one discussed. It not only refers to Yama's popular role as judge of merit and demerit, but in the case of the profound Kat.ha Upani.sad could reasonably include his position as guru. The Buddha's Dharmakdya is also associated with death in the Buddhist Tantras. Thus, Tsofl-kha-pa (1357-1419 A.D.), founder of the Gelugpa school of Tibetan Buddhism, quotes in the Prajfi~-jfiana Initiation section of his Shags rim chen mo the work Mukhdgama regarding the experience of the D h a r m a k a y a in five states
(avastha) :8~ Because one experiences the Dharmak~ya, Joyful, equal to the sky, for only an instant At the time of (1) death, (2) faint, (3) Going to sleep, (4) yawning, and (5) c o i t u s . . . Coomaraswamy might well have seized on this passage as a point of departure for his extraordinary insight and learning, neither of which the 7~ Cf. Lin (op. cit.), p. 21-2, for a discussion of the three chief personnages of the Y~mas, with no suggestion that any of the three is Yama. 80 Abhidharma-koga, Chap. III, commentary on stanza 59. 81 The author's "Contributions regarding the thirty-two characteristics of the Great Person", Liebenthal Festschrift; Sino-lndian Studies, Vol. V, Parts 3 and 4, pp. 245-6. 82 Snags rim, Peking edn., f. 286b-2: / ~at tuff las / ehos sku tab dgah mkhah, mfiam pa / ~i dab brgyal daft gfiid log daft [ glal daft bldarig dus skad cig tsam / myofl bar h.gyur bas [ ~es ~i ba la sogs pa gnas skabs lfiar chos sku myofl bar gsufis pa.
58
ALEX WAYMAN
present writer can match. When he speaks of "the sacrificial initiation which involves a temporary or symbolic death, and a rebirth ''8~ he may also have indicated the relevance of the Dharmakaya of the Buddhist Tantras to the Ka!ha Upani.sad. Then who or what is this Yama, whose names have been set forth and a number discussed above? Taking them all together, it is possible to come to a reasonable conclusion. Yama is the male side of mankind, the maker of laws, the judge, the destroyer. His joy is heaven ; frustration, thepretas; and suffering, hell. He is both the Dionysian and the Apollonian man. But those names derive from various Indian traditions, which in the preChristian period were handed down in family circles. It does not follow that a learned man of 6th century B.C. India, and interested in the subject of Yama, knows what is taught in all the various lineages in different places of India, any more than it would be true today, even with our vastly superior methods of communication. Hence only certain aspects of the synthetic Yarna sketched above are found in any given text. II. YAMA IN THE MYTHOLOGICAL PANTHEON
Deities are in some respects similar to words. One word will come to acquire much of the connotation of another word, not necessarily a synonym, and perhaps thereby cause that other word to fall into disuse. We could carry the comparison to religions. One explanation given for the disappearance of Buddhism in India is that the orthodox schools yielded somewhat, absorbed the "telling points" of Buddhism into their own schools and thus "stole the thunder" from Buddhism. Yama is signified by his names, but he does not come alone, any more than words do. A deity has his functions among the functions exhibited by the various deities of the given culture. One deity can replace another by acquiring the major characteristics of that other, and yet be of different character. Or, as sometimes happens, when one deity acquires certain characteristics of another, the latter may simply lose those and continue on with altered or diminished character. A simple way of showing Yama's place in the "scheme of things" is by a table of the lokapdla prepared from two Tibetan passages :s4 83 "Notes on the Ka~ha Upani~ad", NIA, Vol. I (April, 1938), p. 45. 84 The commentary on the Subahupaript'cchd, Tshig gi don b~ad paOi brjed bya~ (TOhoku Cat. No. 2672), Derge Tg., Rgyud, Thu, 71b-3, f.: / gsafi bah.i dkyil h.khor las / ji skad du / brgya byin gyi ni rdo rjer gsufis / me lha.hi phyag rgya Jugs thab yin / g~in rjel.aiphyag rgya be con yin [ bden bral gyi ni ral gri yin / chu bdag gi ni iags par bgad [ dut~ gi phyag rgya rgyal mtshan yin / ku-be-ra yi dbyug to yin [ dbagt phyug gi ni
59
STUDIES IN YAMA AND M.~RA
Directional Regents
(lokapala)
Seal (mudr~)
(E.) Indra (S. E.) Agni (S.) Yama (S. W.) Nir.rti (W.) Varun.a (N. W.) V~tyu
Thunderbolt (vajra) Hearth (agnikuntja) Club (da.n0a) Sword (khad.ga) Noose (pfiga) Banner (dhvaja)
(hi.) Kubera, or Vaigravan. a (N. E.) Igfina, or I~vara (Below) P.rthivi (Earth Goddess) (Above) Sfirya-Candra (Visible Sun and Moon)
Mace (gadS) Trident (tri~fila) Water Pot (kumbha or kalaga) Disk (man.tJala)
Retinue (pariv~ra) Gods (deva) Seers (.r.si)8s Mothers (m~t.rka) Demons (rfik.sasa) Serpents (nfiga) Flying Wisdom-holders (vidyfidhara)8e Secret Folk (yak.sa) 8r Creatures (bhfita) 88 Local Genii (k.setrapati)
T h e w o r d " S e a l " (mudrd) needs s o m e e x p l a n a t i o n . This is the variety known
as " S y m b o l i c Seal" (samaya-mudrd).
Buddhaguhya
writes, s~
" M o r e o v e r , the " S y m b o l i c Seals" are the ideas (sam.jgd) ' t h u n d e r b o l t '
(vajra), ' h o o k ' (ahkuga), ' a r r o w ' (data), ' ( ? ) d r u m ' (*nanda, T. m~es pa), a n d so f o r t h . "
T h u s this p a r t i c u l a r " S e a l " is the idea possessed b y the
deity i n q u e s t i o n . F o r e x a m p l e , the E a r t h G o d d e s s (p.rthivT) has the i d e a rtse gsum gsuils / sa yi phyag rgya bum pa yin / fii zlah.i phyag rgya zlum por bya / ~es gsufls te/. The work .Hphags pa drd-bi-da.hi bum pabi cho ga (Tfhoku No. 3130), Derge Tg., Rgyud, Pu, 240b-3, 4: / brgya byin gyi .hkhor ni lha rnams so / me.hi h khor ni drail sroil rnams so / g~in rjeh.i h.khor ni ma mo rnams so / bden bral gyi .hkhor ni srin po rnams so / chu lha.hi hkhor ni klu rnams so / fluff lha.hi 13.khor ni rig shags h.chail rnams so / rnam thos kyi bu .hkhor ni gnod sbyin rnams so / dbafl ldan gyi .hkhor ni .hbyufi po rnams so / sa.hi lha mo h.khor rnams ni sa bdag rnams so /. The former work gives the entries Kubera and ~vara; the latter, the entries Vai~ravan.a and I~fina. 85 D . R . Patil, Cuhural History fi'om the Vdyu Pura.na (Poona, 1946), p. 22: "It is said that they are called mahar$is because they see (r.santimahat)..." "The Brahmar.sis... 'see' B r a h m a n . . . " "The Devar~is... 'see' D e v a s . . . " 86 See the illustrations in Stella Kramriseh, The Art oflndia (New York, 1954): NO. 68 "Flying Vidyfidhara, on East Wall, VirQp~k.sa Temple"; and No. 77 "Vidy~dharas flying with lotus bud and mirror, Svarga Brahm~ Temple". s7 The translation is a combination of the "good folk" (puttyajana) of the AV as pointed out by Coomaraswamy, Yak~as, II, p. 3; and the "secret one" (guhyaka), an epithet of the yak.sas. 88 Apart from the direction, this may be the oldest line of all. i~.na, or Tgvara, i.e. Siva, with trident, and master of creatures, may, as has been postulated, have descended from the horned "lord of animals" (pagupati) of Mohenjo-daro. 89 Tantrarthdvatara, Derge Tg., Rgyud I.zgrel, Vol. t:Ii, 6b-3 : / dam tshig gi phyag rgya de dag kyail rdo rje daft / lcags kyu daft / mda.h daft / miles pa la sogs par .hdu ges ste/.
60
ALEX WAYMAN
of "water pot", probably filled with water and adorned with young sprouts. 9~ When we look upon earth this way, we are thinking in mythological terms. Yama has the idea of "club", the Skt. word dan.d.a also meaning "punishment". The eight directions of the guardians are standardized in the Hindu Pur~n.as - post-Christian literature, although some of their contents are older; and earlier texts vary considerably in the directions ascribed to the guardians aswell as in the particular guardians themselves. 91 Coomaraswamy writes, 92 "The earliest assignments of deities to the four quarters are those of YV., I, 8, 7, where we get Agni (E), Yama (S), Savit.r (W), and Varun.a (N), B.rhaspati (Zenith), and ib. VI, I, where we find Pathy~t Svasti (E), Agni (S), Soma (W), Savit.r (N), and Aditi (Zenith); lb., II, 4, 14, Indra is guardian of the East." This shows that from the earliest assigning of deities to directions Yama was in the South in one system, and this assignment won out over alternate possibilities. As a matter of fact, Yama has been a fairly consistent winner. It may be that the half brothers Manu and Yama "originally" corresponded to the Egyptian Osiris and Set pair. If so, the Indian Yama took on the opposing characteristics and became both Osiris and Set. It is not in the scope of the present article to discuss such a possibility. Yama's relation with Varun. a requires some treatment. Kane translates .Rg. X. 14.7:93 "Hasten, hasten by the ancient paths (to that place) where our forefathers that went before us passed. May you (the departed) see the two kings Yama and god Varun.a rejoicing as they will." Kane writes, 9a "In .Rg. IV. 5.5 it is said that those men who are bereft of.rta and satya, being sinful, create a deep place for themselves." Varu.na was, of course, the upholder of .rta: "Durch die Wahrheit schtitze du mich, o Varu.na. ''gs Kar.na speaks in MBh. Vana P., edn. of Krishnacarya and Vyasacharya, 303, 6A: " N o t so do I fear death as I fear untruth" (bibhemi na tathd m!'tyor yatha bibhye 'n.rtddaham), showing that being in disgrace 90 cf. the illustrations of filled vases in Coomaraswamy, Yaks.as, Part II, Plates 28, 29, 31, 32, 33. 91 See Jitendra Nath Banerjea, The Development of Hindu Iconography (Univ. of Calcutta, 1956), 519-29, for a discussion of the directional guardians. 9.2 Yaksas, II, 31 (note). In Dr. Subhadra Jha's "Studies in the Paippal~.di Atharvaveda", JBRS, Vol. XXXIX (Sept. 1953), p. 347, we find in one list Yama in the South, and in another, not containing Yama, Indra in the South. 93 Op. cit., in the translation of X. 14, pp. 191-4. ~4 Ibid., p. 154. J. Norman Brown, "The Rigvedic Equivalent for Hell", JAOS, 61 (June 1941), p. 79, locates Nirrti (our next topic) "personification of Malevolence and Destruction" beneath the earth in a place of non-rta (an.rta) and non-existence (asat). ~5 Heinrich Ltiders, Varuna, I: Varuna und die Wasser. Aus dem Nachlass herausgegeben yon Ludwig Alsdorf (G6ttingen, 1951), p. 31.
STUDIES IN YAMA AND M.~RA
61
with Varu.na is more terrible than being in Yama's power. As we come into the Christian period, "truth" was still of lofty prestige in India under the word satya (in modern times scholars have discussed the meaning of Varu.na's rta, while they do not doubt the meaning of satya). The word dharma had undoubtedly risen to higher prestige, through being one of the trivarga of Hinduism, "righteousness" (dharma), "endowment" (artha), and "desire" (kdma), and through being one of the triratna of Buddhism, Buddha, Dharma, and Safigha. So .~ryagQra in the Jdtaka-m6Id, XXXI, "The Story of Sutasoma", while in earlier stanzas (especially No. 22) speaking highly of "truth" (satya), has the Bodhisattva say in Verse 62: "Who, stationed in righteousness (dharma), ought fear from death?" (dharmasthitah. ko maran,dd bibh~ydt). Varuna is out of the picture. The departed will see Yama, "King of Righteousness" (dharmaraja). That brings up the subject of nir.rti. The S.W. corner of this deity is translated into Tibetan as bden bral "truthless". In Vedic times Nirrti is especially the goddess of destruction, but she has a creative side. Renou 96 explains this word as meaning "dis-order" ("d6s-ordre", for nis+rti), implying dis-organisation. "Disintegrat!on" also seems compatible with his interpretation. For him, rta means "order" rather than "truth". Dandekar 97 explains .rta as "the unbreakable, invulnerable law." Kane says, 9s "In .Rg. X. 165.4 Yama is identified with M.rtyu (death) and the owl (an evil omen) or a kapota is said to be the data (harbinger) of Yama." Now, in the Atharva-Veda, the owl and the pigeon (kapota) are the messengers of the goddess Nir.rti.99 It should not be concluded that Nir.rti and M.rtyu are identical: Renou 1~176 cites examples showing the two in contrast. Furthermore, in the~atapathabrdhma.na Nir.rti is the Earth, and the Earth is Yami, while Yama is Fire. T M In the latter text (Madhyandina, ed. Weber, VII, 2, 1), "Fire" is of course Agni, and "Earth" is iyam. ("this"), or bhami. It follows that when Yami attempts to seduce her brother Yama, and he resists in .Rg. X. 10, there is more to this than at first appears. Yama actually refused Nir.rti. Consistent with etymology, he prefers the Ascendant (.rta) to the Down-fall (nir.rti). When in verse 7 of that hymn she expresses her desire to lie with Yama on the same bed 96 Louis Renou, "V6dique Ni~:rti", hzdian Linguistics (Chatterji Jubilee Volume), Vol. 16 (Nov. 1955), p. 11. 97 R. N. Dandekar, "Asura Varu.na", Annals of the Bhandarkar Inst., Vol. XXI (April-July 1940), p. 184. 98 Op. cit., 159. 0a Bloomfield(op. cit.), p. 166--7. loo Op. cit., p. 13. 10~ Eggeling,Part III, "The Altar of Nir!:.ti"(pp. 319-25).
62
ALEX WAYMAN
(Geldner III, p. 135), we must remember that she is earth. The SBr. says (VII, 2, 1, 11): "This (earth) is Nirrti. And this (earth) makes him decay who passes away" (iyam. vai nir.rtir iyam. vai tam nirarpayati yo nirrchati). But it also says (idem): "Whoever comes into being, comes into being on this (earth)" (asyd~7.~ vai sa bhavati yo bhavati). When we compare what is said about the goddess Nir.rti with the two lines in the table for Nirrti and P.rthivi, it appears that the lines present the destructive and creative aspects, respectively, of the Vedic Nir.rti. Yama's retinue of Mothers (mdtrkd) requires some discussion. We rrdght have expected the retinue to consist of Fathers ~it.r), as in MBh. Vana P. 41.9, where the pit.rs are called martyamfcrtidhara ("mortal corporeal"). But, as Bergaigne (II, 98) shows from the .Rg, he is "le mari des femmes" and"Tamant des filles". Also Yama is the regent of Bhara.ni (either the 27th or 28th asterism), three stars in the shape of the female organ (bhagasam. sthdna). 1~ Fausboll tells us from the MBh. :a03 "His wife is named Dhfimor.n~ (XIII, 7637). Daxa [dak.sa] praj~pati gave 10 of his daughters to Yama (I, 2577). But in XII, 2252 ~ri is named as being his consort, thus also in I, 2578." Gonda states, 1~ "Before being constantly described as specially connected with Vi.s.nu - this stage is not reached before the younger parts of the Mah~bh~trata - ~ri was associated with various gods, all of them assorting well with her. One of them was Kubera . . . . Another was Dharma, whose wife she is in the Mahabharata and even in Vi~n.uitic pura.nas like ViP. 1, 7, 21 ; cf. also Mbh. 1, 66, 13 etc. where Lak.smi belongs to the thirteen daughters of Dak.sa who were given in marriage to Dharma, Tutti "Satisfaction", Pu.s~i "Thriving", .Rddhi "Growth, Success", Kirti "Renown" etc. being her sisters and co-wives." Now, Winternits says, ~~ " . . . in the whole of the Mah~bh~rata the idea prevails that Yama, the god of death, is one with Dharma, the personification of Law." A curious feature of this situation is that thirteen daughters of Dak.sa are also stated by the Mahdbhdrata to have been given in marriage to Kfiw Praj~pati, ~~ who, as the father of Vivasvat, is Yama's "grandfather" on his father's side) ~ But in this respect, Yama also resembles his "grandfather" on his mother's side, Tva.st..r (whose xo~ Sujitkumar Mukhopadhyaya, ed., ~qardalakar.navaddnam (Santiniketan, 1954), text p. 51, Cf. Nak~.atra Table in H. T. Colebrooke, Asiatick Researches, Vol. IX (1809 reprint), between p. 322 and 323. lo~ Op. cit., p. 137. lo~ Op. cit., p. 223. 10~ M. Winternitz, A History of Indian Literature (Univ. of Calcutta, 1927), I, 297. lo6 Fausboll (op. tit.) pp. 1-2. 1~7 John Dowson, A classical Dictionary of Hindu Mythology... under Ka~yapa.
STUDIES IN YAMA AND M.~RA
63
daughter Saran.yfl with Vivasvat gave birth to the twins Yama and Yami). MacDonell writes, 1~ "Probably because of his creative agency in the womb, Tva.s~r is closely allied with celestial females (gn~.h,janaya.h)or the wives of the gods, who are his most frequent attendants (I, 229, etc.)." The group of ten, and group of thirteen daughters of Dak.sa given in marriage to Yama presumably come from among the 27 asterisms (nak.satra) because in the Mahdbhdrata the 27 Nak.satras are the daughters of Dak.sa (also the wives of Soma, the moon god). l~ The group of thirteen can be interpreted in one or both of two ways: (1) the 13 Yamanak.satra, (2) the 13 lunar months that approximate the solar year. (1) The 13 Yamanak.satra. MacDonell and Keith write, 11~ " . . . the Taittiriya BrShma.na divides the Nak.satras into two sets, the Deva Naksatras and the Yama Nak.satras, being 1-14 and 15-27 (with the omission of Abhijit) respectively. This division corresponds with one in the third book of the Br~hman.a where the days of the light half of the month and those of the dark half are equated with the Nak.satras. The Br~hma.na treats the former series as south, the latter as north; but this has no relation to facts, and can only be regarded as a ritual absurdity." That division is presumably related to the Atharva-Veda teaching, "The gods go to the other world by a path known as devaydna and a dead man, a prospective Pitr by means of a path known as pit.rydna. Ultimately they reach the same place. ' ' m This implies Yama's name "Lord of the Fathers" (pit.rpati). (2) The 13 lunar months that approximate the solar year. Whitney writes, ~1~ "Through all the known periods of Indian history, down even to the present, the current appellations of the lunar periods into which the year is divided have been asterismal, and taken in each case from the nakshatra in (or near) which the moon, during that particular synodical revolution, reached her full." He further states, 113 " . . . even in the earliest of the Br~hmanas, the month date is always given by the same asterism as at present," and goes on to discuss the problem of the year in which (or, for which) the system was established. What is meant is that the Hindus traditionally had a series of 12 solar months, Pau.sa, down to M~rga~irs.a, derivatives from the names, almost all of feminine gender, of 1oa Op. cit., p. 117. lO~ S/Srensen(op. cit.), p. 496. 11o Vedic Index, I, p. 414. 111 Shende(op. cit.), p. 253. 112 WilliamDwight Whitney, Oriental and Linguistic Studies, See. Ser. (New York, 1874), p. 360 (in his essay "The Lunar Zodiac"). ~13 1bid., p. 361.
64
ALEX WAYMAN
the twelve asterisms near which the m o o n first became full in a certain year: Pu.syft, down to Mrga~ir~t. The derivative names are transferred to the lunar months in the Luni-so!ar Year, with one name (Caitra) repeated in modern practice to make thirteen, n4 Hopkins shows that, while the 30-day month is also Vedic, anciently, beginning with the RV, ten months were the regular ascription to the period of pregnancy, 1~ and these are of course lunar. Also the rite of one year of the Taittiriya Sam. hitd began either with the full m o o n in Phalguni or in Citr~. ~1~ This is not a solar year, which would only rarely begin with such a stipulated full moon. Also a year beginning with a full m o o n insures thirteen full moons in the year, whereas a solar year more often has only twelve full moons. I f Yama's thirteen wives are the corresponding nak.satras, Y a m a might be the sacrificial year. The group of ten daughters married to Y a m a may imply the development in the womb during ten lunar months. This phase of Yama's career was already alluded to when discussing the name "Coagulated Milk" (dadhna). He was there connected with the Soma. Sometime in the Vedic period Soma becomes definitely a name of the moon; ~17 but since the latter marries all twenty-seven daughters of Dak.sa, he cannot be Yama, who marries only ten or thirteen. The solar nature of Y a m a indicates that he is a kind of "sun by way of the moon". The Soma of the m o o n is really the contribution of the sun, for as MacDonell points out, 1is " I n the post-Vedic literature Soma is a regular name of the moon, which is regarded as being drunk up by the gods and so waning, till it is filled up again by the sun." This is partly implied in .Rg. I, 105, 1, as cited and translated by Hillebrandt: a~9 candr~ma apsv ~nt~tr ~t suparn. 6 dh~tvate divi / nh vo hiran.yanemaya.h pad~t.m vindanti vidyuta.h - / / "es 1/iuft der Mond in den Wassern, ein Vogel a m Himmel. Nicht findet man euren Ort, o goldgefelgte Blitze." The m o o n is a bird, but why "gold-rimmed"? This is not the color of the 114 See the convenient tables in L. Renou et J. Filliozat, L'Inde Classique, Tome II (Paris, 1953), 729 f.; example in Sftrya-Siddhanta translation (Univ. of Calcutta, 1935), p. 34; discussion and bibliography in Vedic Index, I, under "Nak.satra"" some recent views in Philip Yampolsky, "The Origin of the Twenty-eight Lunar Mansions", Osiris (1950), 62-83. 11~ W. Hopkins, "Epic Chronology", JAOS, Vol. 24, p. 19. 11~ A.B. Keith, HOS Vol. 19, p. 607. 117 MacDonell (op. cit.), p. 112. H8 ldem. ~9 Alfred Hillebrandt, Vedische Mythologie (Breslau, 1927), I, 335.
STUDIES IN YAMAAND M/~RA
65
m o o n , but of the sun which has descended into the moon and in this guise is in the water. Ltiders 12~devotes a chapter to the subject "Die Sonne im Wasser", and Brown says, 121 "Possibly the Adityas knew that the sun was in the water as an embryo . . . . " MacDonell writes, 122 "Soma is described as purified with the hands (9, 8634), by the ten fingers (9, 84, 15 s etc.), or, figuratively, by the ten maidens who are sisters (9, 17.65), or by the daughters (naptO of Vivasvat (9, 145)'', one of whose daughters is Yank. Also, 1~3 "Soma is the drop which grows in the waters (9, 85 l~ 89"). Hence he is the embryo of the waters (9, 9741; SB. 4, 4,521) or their child, for seven sisters as mothers are around the child, the newly born, the Gandharva of the waters (9, 614)." As another well-known formulat;.on of Yama's ancestry, MacDonell states, 10-4"In their dialogue in the RV. (10, 104) Yama and Yami call themselves children of Gandharva and the water nymph (apyd yo.sd)", the latter being an apsaras, 125 who could be called a "divine courtesan" (divya-vedya), for Bergaigne (II, 506) points out regarding the mother of the twins (cf. his. II, 98): "Nous avons d6j~ vu que Sara .ny~ est la femelle c6leste qui se d6robe ~t l'attente des hommes." Ehni informs us, 12s "In sp~iteren Stricken wird Saran yu mit der Sindhu (Luftmeer) identifizirt." Wherever this river may be, Law says, 1~7 "As described in the .Rg Veda (X, 75) the Sindhu in might surpassed all the flowing streams." Just as Yama's mother Saran. ya is a river, so also is his sister Yami, because Yama was shown to have the name "Brother of Yamun~", and the Yamun~ is a river. We need not raise our eyebrows at Yami's ability to be in one text the earth and in another a river. After all, Yama is many things; why deny Yami the right to change her mind? It appears that when Yama and Yami are co-uterine she is a "river", and that when he becomes a ploughman, she becomes the earth. That brings up the topic of rivers, about which Lriders (in the posthumous work already cited) has gathered much material. ChildO 2s mentions that among clay tablets of Mohenjo-daro, "one shows a river gushing out of a goddess's womb." In the MBh. Vaitara.ni is a river in the 1~o Op. cit., 294-307. 121 W. Norman Brown, JAOS, Vol. 62 (June 1942), p. 91. 122 Op. cit., 105. 12~ Ibid., 107. 124 Ibid., 172. 125 Cf. the entry, s.v. in Mayrhofer, op. cit., I, 40-1. I~G Op. cit., p. 18. 12~ Bimala Churn Law, Rivers of India (Calcutta, 1944), p. 10. The same author's Historical Geography of Ancient India (Paris, 1954) is a convenient reference work for the geographical names sacred to Hindu and Buddhist texts. 128 V. Gordon Childe, New Light on the Most Ancient East, 4th Edn., p. 195.
66
ALEX WAYMAN
region of Yama, and in the regions of the Pitrs the Gafig~ becomes Vaitaran.i. 1~9 The Vaitaran.i is called a cow (dhenu) in the Garu~laPurdn.a? a~ According to Dh~rmika Subhftti the water current of the river Vaitarani is of blazing copper, and the man whose evil deed sends him into this river is perpetually burned by fire. m According to Vasubandhu, on both sides of this river the "policemen of Yama" (yamarak~asa), armed with sharp weapons, prevent any of the hell beings from emerging, t32 Buddhist doctrine speaks of the three "evil destinies" (durgati), hell beings, animals, and pretas, ass Ui Hakuju's Concise Buddhist Dictionary (in Japanese), p. 365, mentions a river to be crossed on the seventh day after death. There are three current speeds of this river according to the karma of the previous life. The implication here is that the three currents represent the karma of the three "evil destinies". If the river can be crossed, presumably one would go to one of the "good destinies" (sugati), men and gods. Asafiga (4th cent., A.D.) in the CinMmay~ BhCtmiof his Yog~c~ra-bh~mi, quotes this verse from the Kumdrikd-pra~na-gf~thd: ta~ / katha .m vih[trabahulo bhik.su.h paficaughatir .has taratiha .sas..tha.m / ] evam. dhyfiyi vipulg .m kftmat.r.s.nfi.m tir.nobhavaty apratilabdhayoktf~// "How does the monk with a multiplicity of states of existence, 1as Having forded the five turbulent streams, ford here the sixth? And how the meditator who has not achieved union Ford the extensive thirst of desire?" Asaflga comments on the six turbulent streams (ogha): .sad. ime ogh~ cak.sur ogho rapg.nft.m darganftya ] y~tvan mana ogho dharm~.nft .m vijfiapanftya / "These are the six turbulent streams: the eye (cakeus) is a turbulent stream through seeing forms (r~pa);... (down to) the mind (manas) is a turbulent stream through knowing natures (dharma)." Hence the five turbulent streams are the five senses through which we perceive the outer world. The meditator has turned away from that world x~o S6rensen (op. tit.), p. 703. tso Cf. Abegg (op. cit.), pp. 117-22. 13t Mus (op. cit.), p. 237. ~8~ Abhi. Koga, tr. III, p. 151-2. t83 Cf. W. E. Soothill and Lewis Hodous, A Dictionary of Chinese Buddhist Terms, p. 62b, entry on the "three mires". ~34 A large part of this bhf~mi intrudes in the Bihar Society Ms. of the ~r~vaka-bhfm~i, which the writer is editing. The verse is in Taish6 30.372c-29 f. ~s5 The translation "states of existence" for vihara is due to Edgerton, Buddhist Hybrid Sanskrit Dictionary, p. 505.
STUDIES IN YAMA AND M~RA
67
because he is referred to as having "forded the five turbulent streams". But he still has the problem of the mind (manas). In order to ford the turbulent stream of the mind, one must "be not aroused, not excited, and be mindful" (ndbhisam. skaroti / na kupyati / sm.rtimdm, d ca bhavati), expressions which Asafiga then goes on to explain. The first two Stages (bhami) of Asa6ga's Yogdcdra-bhf~mi are "Stage associated with the five perceptions" (pa~cavijgdnasam. prayuktd bh:tmi) and "Stage of Mind" (manobhftmi). The "Stage of Mind" is for Asafiga the Stage of "mental substance" (citta), explained as the "store perception" (alaya-vij~dna), the store of seeds or tendencies; " m i n d " (manas), the moral agent based on the "store perception" and the condition immediately succeeding (samanantara-pratyaya) any of the six sense-based vijhdna; and the "perception based on mind" (mano-vijgdna), the mental orientation (manaskdra) having a nature (dharma) as support condition (dlambanapratyaya). In the ~ravaka-bhftmi, only the older triad of citta, manas, and vU~dna is employed for the benefit of the drdvaka. Here the word vij~dna means all six vijgdna - the five based on the outer senses and the manovijgdna. Asa6ga similarly defines vijgdna-dhdtu and vijgdna-skandha (Taish6, Vol. 30, 430b-28, 29 and 433c-17, 18) to include all six vij~dna and also the triad. In the case of the skandha he says: / yac cittam, mano vij~dnam, te punal.l s.a.d-vijhdnakdydh./... This citta is certainly the dlaya-vijfidna of non-&dvaka terminology, and there is a good reason for his adherence to the word citta in the ~rdvaka-bhf~mi, while he frequently uses a different word, dlaya-vijgt~na, outside of that bhfcmi. ~se To show an equivalence with the Sh .mkhya buddhi would be more difficult, if indeed it be the case. When Asafiga says that the mind (manas) knows natures (dharma), that does not imply that the mind knows citta. The distinction is made clear by Asafiga's discussion of the four sm.rtyupasthdna ("Applications of Mindfulness"). Two of the Applications are to citta and to dharma, each of which have in this context twenty varieties (Taish6, Vol. 30, 440b-22, f aae The reason is the celebrated verse of the Salpdhinirmocana Sfttra. l~tienne Lamotte, in his edition and translation of the Tibetan text (Louvain, 1935), cites the Tib. and Skt. for the verse on p. 58. His Skt. is presumably the corrected form in Louis de La Vall~ Poussin, VUaaptimatratasiddhi(Paris, 1928), I, 173, further modified by the suggestions of Sylvain l.,6vi, Mat&iaux pour I'll,rude du SysNme Vijaaptimatra (Paris, 1932), 103 (note). The verse runs as follows: The profound and subtle Appropriating Perception (adanavij~ana) Flows with all its seeds (bTja)like a turbulent stream (ogha). I did not teach that to the "children" (bdla), Indeed, lest they should imagine it to be a self (atman). Perhaps this is the psychological Sindhu whose "might surpassed all the flowing streams".
68
ALEX WAYMAN
and 440b-25, f.). From the ~rdvaka-bhf~mi Ms. the first two citta are sardga and vigatardga, and the first two dharma are rdga and rdgavinaya. The mind (manas) holds the dharma ("what is held") of craving (rdga) in the corruption category (sam.klega-pak.sya) or taming of craving (rdgavinaya) in the purification category (vyavaddna-pak.sya). What becomes corrupted or purified is the citta. Accordingly, it is either "with craving" (sardga) or "free from craving" (vigatardga). The mind (manas) can choose between natures (dharma). Having held one, the union leaves a seed (b?ja) in the citta or dlaya-v~ihdna. The springing to life of such seeds is what the writer understands as the sam.skdra that is the second member of Dependent Origination (prat?tya-samutpc~da). The foregoing might appear to be a departure from the subject of Yama. The relevance will be greater in later sections of this paper. Here the appropriateness of introducing such material appears when considering what happens at the time of death. A later current of Asafiga's school - Yog~chra or Cittam~tra - sets forth that the consciousness at death is the ~laya-vijhdna, often called the eighth vij~dna; while the Abhidharma-koga places the manovijhdna here. 137 The difference is mostly a matter of terminology. One can take the old Buddhist term manovijhdna as implying the triad of the Stage of Mind, above discussed. For example, manovij~dna immediately implies a manas. Again, if the alaya-vij~dna were uppermost at that time, this would not mean that manas is not present. The basic fact is the withdrawal of the "stream of consciousness" (citta-sanydna) from involvement with the outer senses. This is just what the yogin attempts to do. The deceased person must face Yarna. The yogin may therefore evoke Yama without actually having died. This may be in back of the Kat.ha Upani.sad. Now when we say that the deceased person faces Yama, we are speaking in mythological terms. This is not a rational concept of the mind (manas). The mind knows natures (dharma) but not the citta. Consistent with the Abhidharma-koAa, the "perception based on mind" (manovijhana), no longer having material to work on from the outer senses, continues its work of objectifying. Consistent with the Yog~c~tra school, what it objectifies are the natures associated (sa~.nprayukta-dharma) with the ~laya-vij~dna or the citta. Consistent with mythology, what is seen is Yama. Still, the mind (manas) continues to know natures (dharma), and consequently the mind is a turbulent stream. Perhaps this is the river to be crossed on the seventh day after death. The writer neither maintains nor aa7 The Vijftaptimc~trat~siddhi, I, 195, and 196 (note).
STUD1ES IN YAMA AND M,~RA
69
denies that the turbulent streams (ogha) o f the Buddhist text are to be identified with the waters in which grow the embryo Soma. However, the discussion has b r o u g h t out the pertinence o f both groups o f waters to the problem o f Yama. A n o t h e r factor in the situation is that in Buddhism, the fourth member of Dependent Origination (pratTtya-samutpdda) is called " N a m e and F o r m " (ndma-rftpa) and understood as the five personality aggregates in the womb. l~s In the Tibetan "Wheel of Life" this m e m b e r is depicted as two persons in a boat. Vinitadeva, while explaining the drawing of the wheel, says, a39 " N a m e and f o r m are a w o m a n and a m a n . " In mentioning " w o m a n " first, there is little d o u b t that he intends the two to correspond in the given order. We could, however, have decided this independently by the previous finding that the feminine deity gives the name to the solar month. Consistent also is ,~rya-glara's verse in which Lak.smi, the consort o f Vi.s.nu, becomes yathdrthandmd (her name according to its meaning) for the divine king, and the meaning of her name is "characteristic" (lak.sat3a), an Old-Indic meaning o f ndma (usually " n a m e " ) . 14~ A n d this is also consistent with the fact that the Buddha Vairocana (a name meaning " s u n " ) is considered to be the intrinsic nature (svabhdva) o f the personality aggregate " f o r m " (r~pa). m Each o f the five basic winds (v~yu) are the ~as Paul Hacker, "Eigentiimlichkeiten der Lehre und Terminologie ~afikaras",
ZDMG (1950), devotes pp. 258-68 to Ndma-ritpa, with the discussion somewhat more metaphysical than what is presented above, and usually stressing the world rather than the individual, but still not necessarily inconsistent with the Buddhist position. 18~ Vinayavibha~gapadavy~khydna, Tib. translation in Derge Tg., .Hdul .hgrel, Tshu, 149b-4: mifi daft gzugs ni bud med daft skyes pa.ho 140 The kind suggestions of Prof. Dr. Friedrich Weller enable me to make considerable improvement in my translation of two verses describing King ~ibi in my article in the Liebenthal Festschrift, already cited, 255-6. The Sanskrit is in /~rya-gfira's Jdtakamdld (6.18-21). The corrected version given now not only suits better the Sanskrit but also the contexts of the article in which it was quoted and of our present discussion: In him, the multitudes of virtues subsumed under the three classes Took up common residence as though from yoking of rivalries; All their forms (samastarftpd) shown forth (vibabhur), Their glories not undone through opposition and commotion. Lak~mi, who is the scorn of the unrestrained errant, The grievous misfortune of the incompetent, The intoxicating drink of the debased, Became for him her name according to its meaning (yathdrthan~m~). Cf. Keith, translation of Taittiriya Sa~.nhita, p. 80: My name and thine, O all-knower, Which like men changing garments we bear, Let us exchange again; Thou for life, and we to live. ~4t The Liebenthal Festschrift article, p. 245.
70
ALEX WAYMAN
intrinsic nature (svabhdva) of one of the five Tathfigatas, according to the Vajramdld, an Explanatory Tantra (vydkhydtantra) of the Guhyasamdja. The Pa~cakrama quotes this verse from that Explanatory Tantra : m Sarvadehfinugo v~tyuh, sarvaces.!hpravartaka.h Vairocanasvabh~vo 'sau m.rtakgty~d vini~caret. "'The wind that is the intrinsic nature of Vairocana Proceeds through the whole body, Causing all movement of the limbs, And departs (only) from the dead body." In Tsofl-kha-pa's Shags rim chen mo, where this stanza is part of a lengthy quotation (beginning f. 440b-2 in the Peking edn.) from that Explanatory Tantra, the associated discussion shows that this wind is the one called vydna (T. khyab byed). The identification of Vairocana with the vydna wind may have preserved an ancient doctrine found in the ,~atapatha-brdhmaT.la (III, 9, 4, 7), as translated by Eggeling: m " N o w that Up~ .m~usavana (stone), forsooth, is in reality .~ditya Vivasvant (the sun), it is the pervading vital air (vy~ma) of this (sacrifice)." But if the rftpa is the offspring of Vairocana, and Vairocana is equivalent to Vivasvat, who is the father of Yama, it follows that Y a m a is the " f o r m " (r~pa) and Yami the " n a m e " (ndma). This is consistent with their co-uterine state. It is by no means a "coincidence" that Lak.smi is one of the thirteen daughters given in marriage to Dharma, an alias of Yama, and that ~ri is named Yama's consort in the Mahdbhdrata at a stage in Hindu mythology when she is usually the consort of Visn.u. This is not to suggest that Y a m a and Vis..na are identical. The relation is suggested when the Buddhist Tantras interpret the Avatftrs of Vi.sn.u as monthly stages in the womb. Tsofl-kha-pa quotes the above-cited Vajramdld to the effect that in the first five lunar months of uterine existence, the body has successively the forms of fish, tortoise, boar, lion, dwarf. He ~2 L. de la Vall6e Poussin, ed. Pahcakrama (Gand, 1896), p. 19. The Vy~khya Tantras are not considered commentaries, traditionally placed in the Tibetan Tanjur, but are treated as further promulgations (so to say, "telling more") of the Basic Tantra (malatantra), and hence are placed in the Tibetan Kanjur. The textual history of these works is obscure, but they appear to have been composed at the same time as the Basic Tanttas, or soon afterwards. A work like the Guhyasamdja-tantra cannot be understood without employing the Vy~khy~Tantras, or commentaries based on them. The Vajramdld is No. 445 in the T6hoku Kanjur-Tanjur Catalog. While S. K. De, "The Buddhist Tantric Literature (Sanskrit) of Bengal",/VIA, Vol. I (April, 1938) 1-23, does not discuss the Vyfikhy~ Tantras, his article is a useful introduction to Tantric literature. For Buddhist Tantric literature of Tibet, the best introduction is in Giuseppe Tucci, Tibetan Painted Scrolls (Roma, 1949), Vol. I. ~3 J. Eggeling, ~atapatha Brahma.ha,Part 2, SBE Vol. 26 (Oxford, 1885), p. 240.
STUDIES IN YAMA AND M.~RA
71
goes on to say, ~44 "It is said:145 'Fish (matsya), tortoise (karma), boar (vardha), man-lion (nara-sim.ha), dwarf (vdmana), the two Rfimas, Kr.sn.a, Buddha, and Kalki are the ten.' These are the ten Avatars of Vi.s.nu. In agreement with what is maintained by other schools, the states of womb correspond to five and do not correspond to the remainder; but according to the Kfilacakra school 14e they correspond to the ten Avatars." The Vajramdld uses the terminology, first, second, and so on to ten, t.zpho ba, for which the original Sanskrit is probably sam.krdnti, here 'transits,' and this Skt. word is the one used to indicate the sun's entrance into a sign. ~47 Since the construing of certain Vi.s.nu Avat~ras as uterine stages may be novel to Western scholars, ~4s it seems advisable to present more information from another work of Tsofl-kha-pa where he describes more fully these shapes. 149 (lst month): "At that time the body of the being is like t44 The Rdo rjebi bzlas pa.hi rim pa (Steps of "Thunderbolt Muttering"), Lhasa Collected Works, Vol. Cha, 16a-5,6: [ ji skad du / fia dab rus sbal phag pa daft / mi yi sefi ge mi.hu thufi daBt / ra-ma gfiis daft nag pa dari / saris rgyas kal-ki ~es pa bcu / $es khyab h.jug gi .hjug pa bcu g~an dag .hdod pa de dab mthun par mflal gyi gnas skabs lria la sbyar nas lhag m a m a sbyar ro / dus .hkhor las ni h.jug pa bcu ka dab sbyar ro / This work is devoted to the theory and practice ofpra.nayama. 1~5 Not necessarily in the Vajramala. A similar verse is in the Garud.a-Pura.na: Abegg(op. cit.),p. 110. Miss Kamala Ray, "The Ten Incarnations of Vi.s.nuinBengal", 1HQ, Vol. XV (1941), pp. 370-85, explains that while the Avatara theory is very ancient, the standard list is more modern, and (p. 373) "Epigraphic evidences testify to the existence of this theory from the 5th century A.D. onwards [in Bengal]." For some alternate lists, see J. N. Banerjee, "The Avat~ras of Vi.s.nu and their enumerations in some early Indian texts", Bharata-Kaumudi (honoring Dr. R. K. Mookerji), Part I (Allahabad, 1945), 61-8. 246 For an introduction to the background of the Kalacakra school, one may refer to Johannes Schubert, "Das Wunschgebet um Sambhala", Mitteilungen des Institutsfiir Orientforschung, Berlin, I. 3 (1953), 424 f. In Tsori-kha-pa's passage here quoted, and in his further discussion, it is plain that he rejects the Kalacakra position in favor of the Guhyasamaja position in this matter. This often happens in Tsori-kha-pa's works. 247 S~rya-Siddhanta, p. 310 (xiv. 3). Compare "What was the seed (retas), that became the Year (sa .mvatsara)"; A. K. Coomaraswamy, A New Approach to the Vedas (London, 1933), p. 22, from B.rhad-ara.nyaka Upani~ad, I, 2.4. Cf. F. D. Lessing, Yung-Ho-Kung (Stockholm, 1942), I, 119, re the 12 hands of Samvara, "he knows the twelve nManas and the twelve bhava-samkr6nti (T. srid-pa-hpho-ba, projections of .h consciousness, etc., into another form)". 14s The admirable coverage of early Vi~ouism by Gonda (op. cir.) would doubtless have led him to discuss such a theory if this had already been seriously advanced. He himself points out (p. 5, note), " . . . ~Sews, beliefs, customs, which are early superseded in some circles may be long retained elsewhere." The classical form of 10 Avat~ras, listed above, seems to have been interpreted in some quarters as a deity proceeding through major evolutionary stages - sub-human stages, the stage of man as the dwarf, heroic stages of superman, finally the divine stage, thereby "pervading" the whole universe. 1~ The Gsapt ba ~dus pa~i yig ehuti ~ter gcig ("Twenty-one brief writings on the Guhyasamaja"), Collected Works, Vol. Cha, 62b.
72
ALEX WAYMAN
the shape o f a fish. ''1~~ (2nd m o n t h ) : " A t that time, because five elevations - 2 arms and 2 legs (making four) and a head - project, the b o d y is like the shape of a tortoise. ''151 (3rd m o n t h ) : "Because he bears his fifth member [i.e. the head] and bends it somewhat downward, he has the shape of a boar. ''152 (4th m o n t h ) : " A t that time, the upper part of b o d y being somewhat enlarged, he has the shape o f a lion. ''153 (5th m o n t h ) : " H e has the 12 great 'joints' (dMtu, T. tshigs) o f the fundamental spots. ''~Sa According to the Vajramdld, as quoted by Tsofi-kha-pa in the former work, one perfects the five personality aggregates (skandha) in the first five months and the five senses in the second five. x55 Here the five personality aggregates are of course the re-emergence f r o m latency o f residues and tendencies carried in the 'stream of mental substance' (citta-sam. t6na) which transmigrated f r o m a previous tenement of life. There can be no perception (vij~dna) based on the five senses until the five senses are perfected. I f Y a m a ' s role as prototype o f m a n requires him to go t h r o u g h uterine existence, he overlaps five m o n t h s o f Vi.sn.u's Avatftrs. This is consistent with certain facts o f mythological overlapping as set forth previously, i.e. sharing of Sri and Lak.smi, and his full ten months in the w o m b could explain the ten daughters o f Prajftpati given him in marriage. In resembling his grandfather Tva.s.t.r, Y a m a m a y in this phase o f his career be representing phylo-genetic recapitulation of theoretical former 150 de.hitshe sems can gyi 1us fiah.i rnam pa lta bar yod par. 15x de.hitshe lag pa gfiis rkafi pa gfiis te b~i mgo daft bcas pa.hi .hbur bu lfia dod pas lus rus sbal gyi rnam pa lta bu yod do. l~z de.hi yan lag lffa chaff ba la cuff zad rgur pot yod pas phag rgod kyi rnam pa lta bur yod do. 153 deh.i tshe stod curl zad rgyas pa seff ge.hi rnam par yod do. Cf. "upper part of body leonine" (si.mhapf~rvardhakaya),one of the 32 lak~ana of the Buddha. 154 de.hi rtsa ba.hi gnas tshigs chert bcu gfiis yin no. as~ It may interest some readers to give a brief survey of the process as set forth in all three works of Tsofi-kha-pa mentioned. The ten lunar months sequentially perfect the five skandhas, "perception" (vij~ana), "feeling" (vedana), "motivation" (sa.mskara), "ideas" (sar~j~a),and "form" (rapa); then, the five senses associated with the orifices, "eye", "ear", "nose", "tongue", and "torso", In the first month there is the wind pra.na in the heart center; and in the next four months four more winds arise one after another from the preceding wind, in this order: apana (intestinal region), samana (navel), udana (neck), vyana (in the 12 fundamental spots). These winds support the skandhas perfected in the corresponding month. In the last five months, five winds that are "limbs" (a~ga) of the heart-based pra.na pass through the orifices to associate "objects" (vigaya) with the senses, in effect, generate the five "perceptions" (pa~ca-vO~ana) of outer objects. These winds are, in order, naga, k~rma, k.rkala, devadatta, and dhanaK]aya. The nine goddesses of the GuhyasamSja-tantra are the essentiality (tattva) of the winds (omitting wana): Mamaki, Locanfi, Tfir~, and Pa.n.dara for prana through udana; and Rfipavajr~, gabdavajr~, Gandhavajrft, Rasavajrfi, and Spar~avajrgt for naga through dhana~jaya.
STUDIES IN YAMA AND MARA
73
evolutionary stages. MacDonell writes, TM "The RV. further states that Tva.s.t.r adorned all beings with form (10, 1 I09) . . . . He himself is called omniform (vidvarapa) oftener than any other deity in the RV." Pati1157 believes that the original avat~ras were seven in number and says, " . . . cf. also Dikshitar, Matsya Purdn.a, A S t u d y , 24 ft. where the author claims to have found the rudimentary notions regarding the theory of evolution." Unfortunately, the latter work is not presently available to the writer. Seven as the number of original avatfiras suggests a relation to the seven rivers.
(to be continued)
156 Op. cit., p. 190. 1~ Op. cit., p. 116.
