Chapter 6.1 - The Shakta Upapuranas—Introduction

CHAPTER I THE SHAKTA UPAPURANAS The history of the worship of female deities in India goes back to a remote past, as early literary evidence from the Vedic period downwards and the excavations at Mohenjodaro, Harappa and other places amply indicate, but the conception of a central goddess Devi as Sakti, to whom all other female deities were affiliated as her parts or incarnations, and the compilation of Puranic works dealing with her praise, nature, exploits and worship, were matters of comparatively late ages. As a matter of fact, there have been Mahapuranas on the rites, customs and faiths of the Brahmas, Pancaratras and Pasupatas from long before the beginning of the Christian era, but not a single work of this class has ever dealt exhaustively or even principally with Sakti-worship, although chapters on the praise and worship of the different forms of Devi are to be found in the Markandeya-purana, Vamana-purana, Varaha-purana, Kurma-purana, and so on¹. In comparatively late days there arose a number of Sakta Upapuranas of note, of which the following have come down to us: Devi-purana, Kalika-purana, Mahabhagavata, Devibhagavata, Bhagavati-purana, Candi-purana (or Candika-purana), Devirahasya, and a second Kalika-purana (which is also called Kali-purana and Sati-purana and is quite different from the Kalika-purana mentioned above). Of these, the first four are available in printed forms and the rest still exist in Manuscripts². In the present Chapter we shall deal with the printed works only, leaving out the remaining ones for future treatment. It may be mentioned here that these Upapuranas relate very often to 1 See Markandeya-purana 81-93 (=Venkatesvara Press (Bombay) ed., chapters 78-90), Vamana-purana 17-21 and 51-56, Varaha-purana 21-28 and 90-96, Kurma-purana I. 11-12. Also Mahabharata IV. 6 and VI. 23 (containing hymns to Durga) and Harivamsa II, chapters 2-4 and 22 (in which Devi has been praised). * For these Manuscripts see R. L. Mitra, p. 187, No. 415 (Bhagavati-purana), and Julius Eggeling, VI, pp. 1202-05, Nos. 3360-61 (Candi-purana or Candika-purana, also called Bhagavati-purana and Kalika-purana), and pp. 1193-98, No. 3344 (Kalika-purana, also called Kali-purana and Sati-purana). See also R. L. Mitra, I, pp. 208-9, No. 370 (for description of a Manuscript of the Candi-purana).

2 STUDIES IN THE UPAPURANAS the central goddess Devi and sometimes to one or other of her principal forms such as Durga, Kali (or Kalika), Candi, Sati, etc., although the growing popularity of the conception of Sakti in India inspired people to look upon every female deity as a Sakti (Active Energy) of a particular male god, to whom she was associated very often as a wife. Besides the Puranic works mentioned above, there are also a few others, such as the Brhaddharma-purana and the Bhavisyottara³, which contain chapters on Devi-worship; and among the Puranic verses quoted in the Smrti Nibandhas of different parts of India on the praise and worship of Devi we find a large number which cannot be traced in any of the Puranic works now extant. For instance, in the Bengal Nibandhas a good number of verses on the autumnal worship of Durga has been ascribed to the Bhavisya-purana, Linga-p³., Nandikesvara-p'., Brhan-nandiThis Upapurana has been printed as the fourth Book (called Uttara-parvan) of the Venkatesvara Press (Bombay) ed. of the Bhavisya-purana It will be examined in details in Chapter II of the present Volume. For our brief analysis of its contents and a discussion on the problems relating to its date, provenance, etc. see Journal of the Oriental Institute, Baroda, III, 1953-54, pp. 8-27. 4 See Kv, pp. 512, 513; Dv (Sul.), pp. 1-2, 6, 7, 9, 10, 14-17, 19-21, 23-26; Dv (Srin.), p. 52; Kt, fol. 62 a; Vk, pp. 367, 372, 376, 378-9, 385-7, 391, 396-7; Dpt, pp. 5-7, 13-14, 16-24, 30-32, 37, 38, 44; St, I, pp. 67, 68, 75, 76, 83, 87, 88, 90, 98, 101-3. The following abbreviations have been used in footnotes 4-62 of the present Chapter: Dbht=Durga-bhakti-tarangini (of Vidyapati Upadhyaya of Mithila). Dpt=Durga-puja-tattva (of Raghunandana of Bengal). Dv (Srin.) = Durgotsava-viveka (of Srinatha Acarya-cudamani of Bengal). Dv (Sul.)=Durgotsava-viveka (of Sulapani of Bengal). Kr Krtya-ratnakara (of Candesvara Thakkura of Mithila). Kt=Krtya-tattvarnava (of Srinatha Acarya-cudamani of Bengal). Kv=Kalaviveka (of Jimutavahana of Bengal). Ns Nirnaya-sindhu (of Kamalakara-bhatta of Benares). Sp=Samvatsara-pradipa (a Smrti work of Bengal). St Smrti-tattva (of Raghunandana of Bengal). Vk Varsa-kaumudi (of Govindananda Kavikamkanacarya of Bengal). 5 See Kv, p. 512; Sp, folios 23 b (anonymous), 24 a (anonymous); Dv (Sul.), pp. 2, 5, 8, 22; Dv (Srin.), pp. 43-45, 50; Kt, folios 60 a, 62 a (anonymous), 62 b (anonymous), 63 b-64 a; Vk, pp. 368, 371 (anonymous), 375 (anonymous); Dpt, pp. 4, 44; St, I, pp. 65, 74, 76, 87. No verse on Durga-puja is found quoted from the 'Linga-purana' in the Smrti Nibandhas of Sridatta Upadhyaya, Candesvara or any other comparatively early writer of Mithila except Vidyapati, who ascribes only one verse to the 'Linga-purana' in his Dbht, p. 46. • See Dv (Sul.), pp. 7, 8 (?), 9; Vk, pp. 367, 375, 420; Dpt, pp. 2, 7, 38; St, I, pp. 69, 76, 86, 91.

THE SAKTA UPAPURANAS 3 kesvara-p'., Bhagavati-p 8., Kalika-purana (which is now available in printed forms) and Devi-p³., to Satya, and to a work named as Jyotisa". Many of these verses occur in the Nibandhas of Kamarupa and Mithila¹2 also and only a few are found quoted in Bhoja's Raja-martanda, Hemadri's Caturvarga-cintamani, etc., but the great majority of them, especially those on the rites mentioned below, is remarkably wanting in the Nibandhas of Orissa and Southern India.¹³ No Smrti-writer of Mithila draws upon the 'Nandikesvara-purana' in connection with Durga-puja. 7 See Dv (Sul.), pp. 8, 12-13, 22, 24; Dpt, pp. 8, 10, 31, 37, 44. Not a single verse on Durga-puja has been quoted from the 'Brhan-nandikesvara-purana' in the Nibandhas of Mithila. 8 See Kv, p. 511; Sp, fol. 23 b (anonymous). In Kv, p. 515 two verses have been quoted with the word 'Katyayani-slokau', but the first of these two verses has been ascribed to the 'Bhagavati-purana' in Kalasara, p. 109 and to 'Katyayana' in Vidyapati's Dbht, p. 44 and quoted anonymously in Bhojadeva's Raja-martanda, fol. 79 a. Both these verses have been ascribed to 'Katyayana' in Dv (Srin.), pp. 44-45 and Kt, fol. 63 b. * See Dv (Sul.), pp. 7, 8; Dbht, pp. 3, 37-38; St, I, pp. 72, 75; and so on. 10 Sce Kv, pp. 514-5. 11 See Dv (Sul.), pp. 9, 10, 14, 26; Dv (Srin.), p. 49 (anonymous); Dpt, pp. 1, 10, 16-17, 45; Kt, folios 63 a, 66 a; St, I, pp. 76-77, 103. 12 See, for instance, Candesvara's Kr, p. 362 (for verses which have been ascribed to the 'Bhagavati-purana' and 'Satya' in Jimutavahana's Kv, pp. 511 and 514 and to the 'Bhagavati-purana' in Vidyapati's Dbht, pp. 37-38), and Vidyapati's Dbht, pp. 37 ff. (for verses of the 'Bhagavati-purana', 'Kalika-purana', 'Bhavisya-purana', 'Devi-purana' and 'Jyotih-sastra' on Devi's bodhana, patrika-pravesa, etc.). See also Sridatta Upadhyaya's Samaya-pradipa, folios 41 b-43 a (for verses ascribed to the Devi-purana and the Brahma-purana) and Rudradhara Upadhyaya's Vrata-paddhati, fol. 28 a-b (for two verses ascribed to the 'Bhavisya-purana'). 13 In Kamalakara-bhatta's Nirnaya-sindhu, pp. 127-8 the line 'sasthyam sayam prakurvita bilva-vrkse'dhivasanam' has been derived from a work named 'Kalpataru'; but this line is not found in the section on Mahanavami-puja (the great autumnal worship of Durga) given in Laksmidhara's Krtya-kalpataru, XI (Rajadharma-kanda), pp. 191-5. As a matter of fact, Laksmidhara draws only upon the Devi-purana for his description of the Mahanavami-puja. In Bhoja's Raja-martanda, fol. 79 a the following three verses have been quoted anonymously: 'aim ravanasya vadharthaya ramasyanugrahaya ca / akale brahmana (? brahmana) bodho devyas tvam priyakrt pura (?) | | aham apy asvine tadvaj jyesthayam bodhayami te / srisaila-sikhare jata sri-phala sri-niketana // netavyo 'si samagaccha pujyo durga-svarupatah / saptamyam mula-yuktayam patrikam ca pravesya ca | l In Hemadri's Caturvarga-cintamani, II. i, pp. 906-7 these three verses and also another, viz., 'mulabhave 'pi saptamyam kevalayam pravesayet / ubhabhyam nava-bilvasya phalabhyam sakhikam tatha | |,

STUDIES IN THE UPAPURANAS (i) 'Bodhana' of Devi in a Bilva tree (generally on the Krsna-navami or Sukla-sasthi Tithi) on the eve of her worship.14 (ii) Offer of different articles (such as kesa-samskaradravya, patta-dora, darpana, etc.) to Devi on the different Tithis from Sukla-pratipad to Sukla-pancami¹³. (iii) The untying of Devi's hair (devi-kesa-vimocana) on the Sukla-caturthi Tithi¹6. have been ascribed to the 'Linga-purana' (The Caturvarga-cintamani reads 'devyas tvayi krtah pura' for the second half of the second line, 'aham apy asritah sasthyam sayahne bodhayamy atah' for the third line, '-sikhare jatah sri-phalah sri-niketanah' in the fourth line, 'maya gaccha' for 'samagaccha' in the fifth line, and 'saptamyam pratas tam sakham grham chittva pravesayet' for the sixth line. The verse 'mulabhave 'pi saptamyam' is quoted in Nirnaya-sindhu, p. 128 with the words 'tad uktam hemadrau lainge'). It is to be noted that in the first verse ('aim ravanasya vadharthaya', which is a mantra) Durga-puja has been connected with the story of Rama. As it is the Nibandhawriters of Bengal, Mithila and Kamarupa who are found to prescribe this mantra (aim ravanasya vadharthaya etc.) to be used in Devi's bodhana in the evening of the Asvina- sukla-sasthi, and as the tradition of Rama's performance of Durga-puja for killing Ravana appears to be of East Indian origin, it is probable that the above-mentioned verses of the 'Linga-purana' were taken by Bhojadeva and Hemadri from some work or works of Bengal or Mithila. 14 'sasthyam bilva-tarau bodham sayam samdhyasu karayet' (ascribed to the 'Bhavisya-purana' in Sulapani's Dv, p. 7, as well as in Dpt, pp. 6, 7, St, I, pp. 75, 102, Vk, p. 367, and Dbht, pp. 40, 62; derived from the 'Durga-bhaktitarangini' in Ns, p. 127). 'bodhayed bilva-sakhayam sasthyam devim phalesu ca' (ascribed to the 'Kalika-purana' in Sulapani's Dv, p. 7, as well as in Vk, p. 365, St, I, p. 75, Dpt, pp. 7, 10, and Dbht, pp. 41, 61; derived in Ns, p. 127 from the 'Kalika-purana as quoted in the Gauda-nibandha'; being the same as Kalika-purana 62. 7 b). 'patri-pravesat purvedyuh sayahne vindhya-vasinim / candim avahayed vidvan natra karya puraskriya | | (ascribed to the 'Nandikesvara-purana' in Sulapani's Dv, p. 7, to the 'Brahmanda-purana' and 'Nandikesvara-purana' in Dpt, p. 7 and St, I, p. 76, and to the 'Bhavisya-purana' in Dpt, p. 7; not occurring in the Nibandhas of Mithila, at least in those of comparatively early dates). 'kesa-samskara-dravyani pradadyat pratipad-dine / 15 patta-doram dvitiyayam kesa-samyama-hetave || darpanam tu trtiyayam sinduralaktakam tatha / madhuparkam caturthyam tu tilakam netra-mandanam // pancamyam amga-ragams ca saktyalamkaranani ca // (ascribed to the 'Bhavisya-purana' in Sulapani's Dv, p. 6, as well as in Dpt, p. 6, St, I, pp. 101-2, and Dbht, p. 40; derived in Ns, p. 127 from the 'Bhavisya-purana as quoted in the Durga-bhakti-tarangini'. Kamalakara reads 'pakva-tailam' for 'patta-doram' and says that the latter reading is given by the Gaudas-patta-doram iti gauda-pathah). 'sukla-pakse caturthyam tu devi-kesa-vimocanam' 16 (ascribed to the 'Bhagavati-purana' in Kv, p. 511 and Dbht, p. 37, and to the

THE SAKTA UPAPURANAS 5 (iv) 'Adhivasa' of Devi in a Bilva tree in the evening of the day previous to that of patrika-pravesa". (v) 'Nava-patrika-pravesa', i.e. the bringing of the nine plants, viz., rambha, kacci, haridra, jayanti, bilva, dadima, asoka, manaka and dhanya, into the puja-mandapa, and the worship of nine Durgas in them¹8. 'Kalika-purana' in Vk, p. 369; quoted with the words 'sistah pathanti' in Kr, p. 362; being the same as Kalika-purana 62. 18 a). "sukla-pakse caturthyam tu devi-kesa-vimoksanam' (ascribed to the 'Linga-purana' in Srinatha's Dv, p. 43 and Kt, fol. 60 a). 17 'sasthyam sayam prakurvita bilva-vrkse 'dhivasanam' (ascribed to the 'Linga-purana' in Srinatha's Dv, p. 43 and Kt, fol. 60 a; quoted anonymously in Kt, fol. 62 a and Vk, p. 371; derived in Ns, p. 127 from the 'Linga-purana as quoted in the Krtya-tattvarnava'). 'sayam sasthyam tu kartavyam parvatyas cadhivasanam / sasthy-abhave tu kartavyam saptamyam api narada | | (ascribed to 'Smrti' in Sulapani's Dv, p. 8 and in Dpt, p. 7, and quoted anonymously in St, I, p. 76). 'purvedyur adhivasyaiva bilva-vrkse tathambikam' (ascribed to the 'Nandikesvara-purana' in Dpt, p. 2). 18 saptamyam patrika-puja rambhadi-navabhir yuta / rambha kacci haridra ca jayanti bilva-dadimau / asoko manakas caiva dhanyam ca nava-patrika / / (ascribed to the 'Brhan-nandikesvara-purana' in Sulapani's Dv, pp. 12-13 and in Dpt, pp. 8 and 10). 'rambha kacci haridra ca jayanti bilva-dadimau / asoko manakas caiva dhanyam ca nava-patrikah / / (ascribed to the 'Linga-purana' in Srinatha's Dv, p. 50, and to the 'Bhavisya-purana' in Vk, p. 372; quoted anonymously in Sp, fol. 24 a, Kt, fol. 62 b, Vk, p. 401; derived in Ns, p. 129 from the 'Krtya-tattvarnava'). 'brahmani kadali-kande dadime rakta-dantika / dhanye laksmir haridrayam durga manaka-patrake | | camunda kalika kaccyam siva bilve pratisthita / asoke soka-rahita jayantyam karttiki mata / / (ascribed to the 'Bhavisya-purana' in Vk, p. 391; quoted in Dbht, p. 63 with the words 'sistah pathanti'; derived from the 'Durga-bhakti-tarangini' in Dpt, p. 12). 'purvahne nava-patrika subhakari dharmartha-siddhi-prada arogyam dhanada karoti vijayam patri-pravese subha / madhyahne jana-pidana-ksayakari samgrama-ghoravaha sayahne vadha-bandhanadi-kalaham sarpa-ksatam sarvada / / (ascribed to 'Jyotisa' in Sulapani's Dv, p. 10, Srinatha's Dv, p. 49, Kt, fol. 63 a, and St, I, p. 76). 'kadali dadimi dhanyam haridra manakam kacuh / bilvo 'soko jayanti ca vijneya nava-patrika | | (quoted anonymously in Rudradhara's Vrata-paddhati, fol. 27 b with v. 1. 'dhanya haridramalakam' and 'bilvasokau' for 'dhanyam haridra manakam' and 'bilvo 'soko' respectively and in St, I, p. 77 and II, p. 664; quoted with the words 'sistah pathanti' on p. 63 and anonymously on p. 132 of the Dbht). It is remarkable that in his Durgotsava-paddhati (fol. 15 a) Udayasimha Rupanarayana, who was a king of Goraksanagara (modern Gorakhpur) in Northern Kosala

6 STUDIES IN THE UPAPURANAS (vi) The tying of the nine plants (constituting the nava-patrika) with an Aparajita creeper¹. (vii) The performance of the Savarotsava on the Dasami Tithi 20. and preceded Vidyapati by a short period, prescribes the worship of the nine Mother Goddesses (matrkah, matarah) not in the nine plants called nava-patrikah but in an eightpetalled lotus painted on the ground with sandal-paste, etc. The present Kalika-purana (62. 51 b-52 a) also says: "The propitious nine Durgas are likewise to be worshipped in the proximity of Devi with Jayanti and other fragrant flowers, because they are the forms of Devi." For information about Udayasimha Rupanarayana and his works see our article in Our Heritage, IV, 1956, pp. 157-176). 19 'svetaparajita-baddhah sarva-sampat-samrddhidah' (ascribed to the 'Linga-purana' in Srinatha's Dv, p. 50). 20 'antapado nisa-bhage sravanasya bhaved yadi / tada sampresanam devya dasamyam savarotsavah | | (ascribed to 'Satya' in Kv, p. 514). 'sampujya presanam kuryad dasamyam savarotsavaih* (ascribed to the 'Bhavisya-purana' in Dpt. pp. 6 and 42, St, I, p. 102, and Dbht, p. 41). 'visarjanam dasamyam tu kuryad vai savarotsavaih' (ascribed to the 'Kalika-purana' in Sulapani's Dv, p. 24, as well as in Vk, p. 366, Dpt, p. 7, St, I, p. 75, and Dbht, p. 41; derived in Ns, p. 128 from the Kalika-purana as quoted in the Gauda-nibandha; being the same as Kalika-purana 63. 18 a and resembling very much Kalika-purana 62. 10 a). 'tatah sampresita devi dasamyam savarotsavaih' (ascribed to the 'Kalika-purana' in Vk, pp. 368, 374; being the same as Kalika-purana 62. 31 b). 'bhaga-lingabhidhanais ca bhaga-linga-pragitakaih / bhaga-linga-kriyabhis ca kridayeyur alam janah | | parair naksipyate yas tu yah param naksipaty api / kruddha bhagavati tasya sapam dadyat sudarunam | l (ascribed to 'Satya' in Kv, p. 514, and to the 'Kalika-purana' in Srinatha's Dv, p. 51 and Kt, fol. 65 b, in Vk, pp. 377-8, and in Dbht, p. 42; quoted anonymously in Kr, p. 362; derived in Ns, p. 130 from 'the Kalika-purana as quoted in the Krtya-tattvarnava'; being the same as Kalika-purana 63. 21 b-23 a). dhuli-kardama-viksepaih krida-kautuka-mangalaih / bhaga-lingabhidhanais ca bhaga-linga-pragitakaih / bhaga-linga-kriyabhis ca kridayeyur alajjitah | l (ascribed to the 'kalika-purana' in Sulapani's Dv. p. 24, as well as in Vk, pp. 377 and 449, Dpt, p. 7, St, I, p. 75, and Dbht, p. 42; being the same as Kalika-purana 63. 21-22 a). 'nau-yanair nara-yanair va nitva bhagavatim sivam / sroto-jale praksipeyuh krida-kautuka-mangalaih || parair naksipyate yas tu param naksipate tu yah / tasya kruddha bhagavati sapam dadyat sudarunam / / (ascribed to the 'Brhan-nandikesvara-purana' in Sulapani's Dv, p. 24 and in Dpt, p. 44, and to the 'Nandikesvara-purana' in Dpt, p. 3; the second verse being the same as Kalika-purana 63. 22 b-23 a). 'visarjanam dasamyam tu krida-kautuka-mangalaih' (ascribed to the 'Linga-purana' in Kv, p. 512 and Kt, fol. 62 a, and to the 'Kalika-purana' in Vk, p. 369; quoted anonymously in Sp, fol. 24 a; being the same as Kalika-purana 62. 20 a. According to Jimutavahana, the word 'krida-kautuka-mangalaih' in the abovementioned line means the Savarotsava.-See Kv, p. 514-tatha satyah '. . .

THE SAKTA UPAPURANAS 7 Even on 'Patrika-pravesa' (i.e. the rite of cutting, from a young Bilva tree, a small branch having two Bilva fruits and taking this branch to the puja-mandapa for worshipping Devi in it), which, as the Raja-martanda" and the Caturvarga-cintamani 22 show, was known outside Bengal, Mithila and Kamarupa, there are a number of verses, found quoted in the Smrti works of Eastern India, which do not occur in those of other provinces. As instances of such verses we may mention the following: 23 'saptamyam mula-yuktayam patrikayah pravesanam/"" 'mulabhave 'pi saptamyam kevalayam pravesayet/ tatha tithy-antare 'py evam rksesu ca phaloccayah//224 'saptamyam bilva-sakham tam ahrtya pratipujayet/'25 'saptamyam patrika-puja kartavya catha manavaih/226 'saptamyam astagayam yadi visati grham patrika sriphaladhya rajnah saptanga-rajyam jana-sukham akhilam hanti mulanurodhat/ tasmat suryodayastham narapati-subhadam saptamim prapya devin tada sampresanam devya dasamyam savarotsavah // savara-varna iva parnady- avrta-kardamadi-lipta-sariro nana-vidhasambaddha-valgita-nrtya-gita-vadyadi-paro bhutva iti savarotsava-padarthah /). ca, etc. 21 See fol. 78 b-saptamyam mula-yukte prathama-pada-gate patrika-sthapanam See also the verse 'saptamyam patrika-puja astamyam capy uposanam, etc.' quoted in fol. 79 a. In his Durgotsava-paddhati (folios 6 b-7 a) Udayasimha Rupanarayana derives the former verse from the 'Raja-martanda' and reads it as 'saptamyam mula-yoge prathama-pada-gate patrika pujaniya, etc.' 22 II. i, pp. 906-7. For the relevant lines see footnote 13 above. 23 Ascribed to the 'Devi-purana' in Sulapani's Dv, p. 8, Dpt, p. 3, and Kt, fol. 60 b, to the 'Nandikesvara-purana' in Dpt, p. 2, and to the 'Linga-purana' in Srinatha's Dv, p. 44; quoted anonymously in Sp, fol. 23 b; mentioned in Dbht, p. 46 as occurring in a 'Gaudanibandha'; derived in Ns, p. 128 from 'the Devi-purana as quoted in the Gauda-nibandha'. 24 Ascribed to the 'Nandikesvara-purana' in Vk, p. 367, and to the 'Linga-purana' in Sulapani's Dv, p. 8, Srinatha's Dv, p. 45, St, I, p. 74, and Vk, p. 368; quoted anonymously in Sp, fol. 24 a, Kt, fol. 62 b, and Krtya-cintamani, p. 28; derived from a 'Gauda-nibandha' in Dbht, p. 46. (A similar verse, agreeing in the first line but differing in the second, is found quoted from a 'Linga-purana' in Hemadri's Caturvarga-cintamani. For the text of this verse see footnote 13 above. Kamalakara-bhatta took this verse of the Linga-purana from 'Hemadri'. -See Ns, p. 128). 25 Ascribed to the 'Kalika-purana' in Sulapani's Dv, p. 13, in Dbht, p. 41, and in Dpt, p. 10; derived in Ns, p. 128 from the 'Kalika-purana (as quoted in the Gauda-nibandha).' 26 Ascribed to a 'dusprapa-kalika-puranantara' in Dpt. p. 8.

8 STUDIES IN THE UPAPURANAS bhupalo vesayet tam sakala-jana-hitam raksasarksam vihaya//27 'mula-yoge tu saptamyam patrikayah pravesanam/ yad uktam tad-dvayabhave kevalayam tithav api//228 As regards the line 'saptamyam patrika-puja astamyam capy uposanam' there is some scope for doubt. This line, which is the same as Kalika-purana 62. 19 a, has been ascribed to the 'Linga-purana' in Kalaviveka, p. 512 and Srinatha's Durgotsavaviveka, p. 43 and Krtya-tattvarnava, fol. 60 a, to the 'Bhagavati-purana' in Durga-bhakti-tarangini, p. 38, and to the 'Kalikap.' in Varsa-kaumudi, pp. 369, 371 and 375. It has been quoted anonymously in Samvatsara-pradipa, 29 fol. 24 a, Krtya-ratnakara, p. 362, and Krtya-cintamani, p. 27. On the other hand, Bhojadeva of Dhara is found to quote it in his Raja-martanda (fol. 79 a) without mentioning its source, and Udayasimha Rupanarayana, who has it in his Durgotsava-paddhati (fol. 6 b), ascribes it to 'sistas', by whom he must have meant those of Bengal (because he flourished earlier than Vidyapati, who only, among the Maithila Nibandha-writers, is found to quote the said line in his Durga-bhakti-tarangini). Besides the verses mentioned above, there are also many others which are peculiar to Bengal and in some cases to Mithila and Kamarupa also. But, as we have already said, the comparatively early Smrti-writers of Western and Southern India have no knowledge either of these verses or of the rites based on them and the later ones record some of these on the authority of the works of Bengal and Mithila. For instance, in his Nirnaya-sindhu (pp. 126-141) Kamalakara-bhatta deals with most of the above-mentioned rites and quotes many of the relevant verses, but Kamalakara's P. 77. 27 Ascribed to 'Jyotisa' in Sulapani's Dv, pp. 9-10, Srinatha's Dv, p. 49, and St, I, 28 Ascribed to the 'Nandikesvara-purana' in Dpt, p. 3. 29 This work, as occurring in Manuscripts at present, belongs to Bengal.-See Hazra in Indian Historical Quarterly (Calcutta), XXI, 1945, pp. 49-55, and D. Bhattacharya in Our Heritage, I, pp. 159-160. See also D. Bhattacharya's Introduction (pp. xl-xlii) to his edition of Halayudha's Brahmana-sarvasva. -*

THE SAKTA UPAPURANAS 9 sources are invariably the Nibandhas of Bengal and Mithila,30 the Devi-purana (a work of Bengal) on 'satru-bali' only", and the present Kalika-purana (which, we shall see hereinafter, must have been written either in Kamarupa or in that part of Bengal which was very near to it). That neither Kamalakara nor his countrymen were familiar with the rites mentioned above, is shown by the following facts. • • (i) In his Nirnaya-sindhu, pp. 120-126 Kamalakarabhatta deals elaborately with the proper time and method of performing the Navaratra-vrata without mentioning any of these rites or quoting any of the relevant verses of the 'Bhavisya-purana', 'Linga-purana', etc. mentioned above. He then begins, from p. 126, to deal with these peculiar rites with the words 'atha pratipadadisu viseso durga-bhakti-tarangi- nyam bhavisye and bases his treatment of these rites on the Nibandhas of Bengal and Mithila as well as on the Devi-purana (with regard to 'satru-bali' only) and the Kalika-purana Had these characteristic rites been in vogue in his country, he would not have dealt with them separately after describing the method of performing the Navaratravrata, nor would he have derived his materials from the present Devi-purana, Kalika-purana and the Nibandhas of Bengal and Mithila. In his Durgotsava-paddhati Udayasimha Rupanarayana also says nothing about the peculiar rites mentioned above. In this work he quotes two verses on patrika-pravesa on the authority of the 'sistas' (who, as we have already seen, must be the Smartas of Bengal) and also another verse³ on the same topic by mentioning as his source the 'Raja-martanda' 30 The only exception is the verse 'mulabhave 'pi saptamyam etc.' (on patrik a-pravesa), which Kamalakara derived from 'Hemadri'. See footnote 13 above. 31 Nirnaya-sindhu, p. 137-bali-prakaras tu devi-puranetatha 'tasyagrato nrpah snayat krtva satrum tu paistikam / khadgena ghatayitva tu dadyat skanda-visakhayoh | |* This verse is the same as Devi-purana 22. 16. 32 'saptamyam patrika-puja astamyam capy uposanam' etc. and 'mulena saphalarn bilva-sakham ahrtya pujayet' etc. (on fol. 6 b). 3ª 'saptamyam mula-yoge prathama-pada-gate patrika pujaniya' etc. (on folios 6 b-7 a).

10 STUDIES IN THE UPAPURANAS (which must have taken it from some work of Bengal or Mithila), but he prescribes the rite of satru-bali on the authority of the Devi-purana3, which is a work of Bengal. (ii) In the Bengal Nibandhas there is a verse, viz., 'rambha kacci haridra ca jayanti bilva-dadimau/ asoko manakas caiva dhanyam ca nava-patrikah//'35 which mentions the names of the nine plants constituting the nava-patrika. In treating of nava-patrika-puja Kamalakara quotes this verse with the mention of the 'Krtya-tattvarnava' as his source, but reads 'kavi' for 'kacci' and 'mana-vrksas ca' for 'manakas caiva'36. These readings 'kavi' and 'manavrksas ca' clearly show that neither Kamalakara nor his countrymen were familiar with 'kacci' and 'mana'. The learned editor of the printed text of the Nirnaya-sindhu frankly admits that he did not understand what objects were meant by the words 'mana' and 'kavi' occurring in the text of the Nirnaya-sindhu³. Now, the word 'kacci' (meaning 'kacvi' or 'kacu'-Arum Colocasia) has been used in the above verse to mean the 'kacu' plant so well known in Bengal, Mithila and Kamarupa, and its peculiar spelling with is common in the Bengal Nibandhas 38 (and also perhaps in those of Mithila) 39. The 'manaka' (popularly 34 See Durgotsava-paddhati, folios 4 b and 15 b. 35 For the names of the works quoting this verse see footnote 18 above. 36 Nirnaya-sindhu, p. 128- . . iti krtya-tattvarnava uktam / patrikas tu- 'rambha kavi haridra ca jayanti bilva-dadimau / asoko mana-vrksas ca dhanyadi navapatrikah || iti tatraivoktah | For the reading 'kavi' for ''kacci' see also Nirnaya-sindhu, p. 130- . patrikah pujayet kadalyam brahmanim dadime rakta-dantikam dhanye laksmim haridrayam durgam mane camundam kavau kalikam bilve sivam asoke soka-rahitam jayantyam karttikim cavahya sampujya durgayai balim dadyat / 37 With respect to the words 'mane' and 'kavau' occurring in Nirnaya-sindhu, p. 130 (... . . mane camundam kavau kalikam .. jayantyam karttikim .) the editor writes in a footnote: "mane kavau iti vastu-dvaya-vijnanam na bhavatiti mrgyam etat/" (See Nirnaya-sindhu, p. 130, footnote 1). The editor was not familiar with the 'Jayanti' plant also. He writes in a second footnote: 'jayanti tahakala iti amarasimha- tikayam'. 38 See Sp, fol. 24 a, Sulapani's Dv, pp. 12-13, Srinatha's Dv, p. 50, Kt, fol. 62 b, Vk, pp. 372, 401, and Dpt, pp. 8, 10. See also Vk, pp. 391, 402, 425, and Dpt, p. 12. The word 'kacu occurs in Kavi Karnapura's Krsnahnika-kaumudi, ii. 39 See, for instance, the two verses 'brahmani kadali-kande' and 'camunda kalika kaccyam' derived by Raghunandana in his Dpt, p. 12 from the Durga-bhakti-tarangini. The second verse mentions 'kacci'. But in these verses as occurring in the printed

THE SAKTA UPAPURANAS 11 called 'mana' or 'mana-kacu' in Bengal) is one of the varieties of arum. As neither 'kacci' nor 'manaka' is mentioned in Amara-kosa or any other famous lexicon, Kamalakara and his countrymen as well as the learned editor of the printed text of the Nirnaya-sindhu totally failed to understand their meanings and made the mistakes mentioned above. (iii) In two verses, of which one is derived by Kamalakara from the Durga-bhakti-tarangini with the words 'durga-bhaktitaranginyam bhavisye', and the other from the 'Kalika-purana', the reading 'saravotsavaih' is given in place of 'savarotsavaih. This wrong reading shows that Kamalakara and his countrymen were not familiar with the Savarotsava which is mentioned in all the Bengal Nibandhas 42 dealing with Durgotsava and which was widely performed, and is still performed in some form or other, in this province on the Vijaya Dasami Tithi. Bhavisya-purana IV, chap. 138, which is one of the main sources of the description of the method of Durga-puja as given in the Nibandhas of Orissa and of Western and Southern India 43, refers to none of the abovementioned rites. As a matter of fact, all these characteristic rites are of East Indian origin. Some of them have been expressly mentioned in the Bengal Nibandhas as local customs prevailing in this province. For instance, the performance , edition of the Durga-bhakti-tarangini (p. 63) the name of the plant is given as 'kacu' and not 'kacci'. In his Vrata-paddhati (fol. 27 b) Rudradhara also names it as 'kacu' in the verse containing the names of the Nava-patrika. (See footnote 18 above.) 40 Jatadharacarya's Paryaya-nanartha-kosa and Cakrapani-datta's Sabda-candrika mention 'kacvi' but do not spell it as 'kacci'. samNirnaya-sindhu, p. 126 (durga-bhakti-taranginyam bhavisye- . pujya presanam kuryad dasamyam saravotsavaih / . . . . . .) and p. 127 (kalika-purane visarjanam dasamyam tu kuryad vai saravotsavaih | | .). In his Vidhana-parijata (II, p. 630) Anantabhatta also quotes the said line of the Bhavisya-purana with the wrong reading 'saravotsavaih'. 42 Though the Samvatsara-pradipa does not make any express mention of the Savarotsava, it anonymously quotes the line 'visarjanam dasamyam tu krida-kautukamangalaih' (which is ascribed to the Linga-purana in Kv, p. 512, Srinatha's Dv, p. 43, and Kt, fol. 60 a). According to Jimutavahana the word 'krida-kautuka-mangalaih' means the Savarotsava (see Kv, p. 514). 43 See, for instance, Kalasara, pp. 103-115 (wherein verses 1-2, 10-11, 13-14, 15-16, 21-23 and 28-30 a of Bhavisya-purana IV. 138 have been quoted), and Caturvarga-cintamani, II. i, pp. 908-920 (wherein the whole of Bhavisya-purana IV. 138 has been quoted). See also Kalasara, p. 229 seyam navami bhavisyat-puranokta-durga-vratadau drastavya.

12 STUDIES IN THE UPAPURANAS of Devi's adhivasa in a Bilva tree in the evening of the day previous to that of patrika-pravesa is sanctioned by Sulapani and Raghunandana on the authority of a verse ascribed to 'Smrti'44 and by Srinatha on the authority of a metrical line 'sasthyam sayam prakurvita' ascribed to the 'Linga-purana'45 But Govindananda says that this rite is based on a local custom, that the line 'sasthyam sayam prakurvita' is not found in the Nibandhas of Maithilas and others and is, therefore, unfounded, and that if this line was taken to be 'samula', it would go against 'sistacara'.46 The tying of the Nava-patrika with an Aparajita creeper, again, is mentioned in Srinatha's Durgotsava-viveka" and Govindananda's Varsakaumudi,48 but is called a local custom in the Samvatsarapradipa and the Durga-puja-tattva 50. In his Durga-bhaktitarangini, which must have been written later than the Samvatsara-pradipa, Vidyapati also refers to this custom" but 44 For this verse see footnote 17 above. 45 For this line see footnote 17 above. 46 Varsa-kaumudi, pp. 370-371-patri-pravesa-purva-dine tv adhivasa acara-prapta eva, so'pi gandha-puspadyaih patri-bhusa-rupa eva ....... | yat tu "sayam sasthyam prakurvita bilva-vrkse 'dhivasanam" iti nama-sunya-vacanam kenacil likhitam tan maithiladi-nibandhesy adarsanan nirmulam eva | sistacaropastambhaka-samulatv- abhimane tu ... As a matter of fact the line 'sayam sasthyam prakurvita' of the 'Linga-purana' does not occur in the works of Candesvara, Vidyapati or any other Smrti-writer of Mithila. 47 P. 50-nava-patrikas cokta limgapurane-'rambha kacci haridra ca jayanti bilvadadimau / asoko manakas caiva dhanyam ca nava-patrikah // svetaparajita-baddhah sarva-sampat-samrddhidah /l 48 P. 401- . . 'rambha kacci haridra ca jayanti bilva-dadimau / asoko manakas caiva dhanyadi nava-patrikah | |' eta ekikrtya aparajita-lataya samvestya . ..... 4º Folio 24 a-acara-vasac ca bilva-sakhaya saha rambhadi-nava-patrikam aparajitalata-baddham pravesayet / 50 P. 51-tata acarad aparajita-lata-baddham nava-patrikam bilva-sakham ca sthapayitva. ☐ $1 See Durga-bhakti-tarangini, p. 63 (bilva-sakhaya saha rambhadi-nava-patrikah samacarad aparajita-lata-baddhah sthapayitva pravesayet) and p. 129 (....... samacarad aparajita-lata-baddha rambhadi-nava-patrikah sthapayet ....). It is to be noted that Candesvara or any other Smrti-writer of Mithila does not mention this custom. On the other hand, both Candesvara and Vidyapati give, without naming any authority, the procedure of worshipping the Aparajita creeper on the Dasami Tithi after the immersion of the image of Durga in water and mention it as a 'sistacara' or simply 'acara'. (See Kr, p. 365 and Dbht, pp. 208-9). But this custom was not followed in Bengal at least down to the time of Raghunandana. It is not mentioned or referred to in Jimutavahana's Kalaviveka, Sulapani's Durgotsava-viveka, Srinatha Acarya-cudamani's Durgotsava-viveka and Krtya-tattvarnava or any other pre-Raghu- ༧་༥༧ ཐ +

THE SAKTA UPAPURANAS 13 does not name or quote any authority. It is highly probable that Vidyapati derived this custom from the Nibandhas of Gauda, to which he refers on other occasions. Further, all the plants constituting the Nava-patrika are very common nandana Smrti work of Bengal. Even Govindananda, who was a contemporary of Raghunandana and deals elaborately with Durga-puja in his Varsa-kaumudi (pp. 365- 449), makes no mention of this custom. It is only in the section on Durgotsava in Raghunandana's Tithi-tattva that the custom has been mentioned in the following lines: evam ca ghatikona-dasamyam aparajita-pujanarhatvat tatpujanam purva-dine / ata eva tat-param evedam "asvine sukla-paksasya dasamyam pujayet tatha / ekadasyam na kurvita pujanam caparajitam //" iti sivarahasyoktaikadasi-yukta-dasami-nisedhaka-vacanam / tatas ca tat-purvakrtyam devi-visarjanam api tadaiva, tadantapakarsa-nyayat / vacaspati-misro 'py evam / (See Smrti-tattva, I, pp. 88-89. This is a reference to Vacaspatimisra's Krtyamaharnava, fol. 72 a). But the genuineness of these lines becomes extremely doubtful when we take into consideration the following facts: (i) In his Durga-puja-tattva Raghunandana deals elaborately with Durga-puja; but neither in the section on Pramana nor in that on Prayoga does he refer to the custom of Aparajita-puja, although this work is, on one occasion, referred to in the Tithi-tattva for detailed treatment of the method of Durga-puja (pujayam visesas tu durga-pujatative 'nusamdheyah-Tithi-tattva, p. 93) and, in commenting on the above-mentioned lines, Kasirama Vacaspati speaks of the high authority of the Durga-puja-tattva in matters of Durga-worship in the following words: tatrodaya-gaminyam muhurtanyunayam dasamyam sravana-naksatra-yuktayam kevalayam va dasamyam visarjanam iti durgarcana-tattva-likhanam / etadgrantha-pramanyam tu 'pujayam visesas tu durga-puja-tattve anusamdheya' iti tithi-tattva-likhanena drdhikrtam / (See Tithi-tattva with Kasirama Vacaspati's commentary, p. 283). (ii) In his Krtya-tattva Raghunandana deals briefly with Durgotsava (see Smrtitattva, II, pp. 444-5), but there also he makes no mention of Aparajita-puja. (iii) The above-mentioned lines of the Tithi-tattva (in which Aparajita-puja has been mentioned) do not occur in their right place after treatment of Devi's immersion (visarjana), which is to precede, and not follow, Aparajita-puja. (iv) From the statements of Kasirama Vacaspati and the Durgarcana-kaumudi (a post-Raghunandana work mentioned by Kasirama Vacaspati in his commentary on the Tithi-tattva, p. 283), it appears that even in post-Raghunandana days the custom of Aparajita-puja attained popularity in a limited circle and did not obtain the sanction of renowned Smrti-writers like Raghunandana. See Kasirama's commentary (on the said lines of the Tithi-tattva): visarjanam dasamyam iti ratry-adau visarjanam na karyam / devi-purane "patri-pravesanam ratrau visargam va karoti yah | tasya rastra-vinasah syad raja ca vikalo bhavet //" ratrav iti paryudasta-kalopalaksanam iti durgarcakaumudi // Also Durgarcana-kaumudi (folios 65 b-66 a): "atra kecit devi-visarjananantaram yadi para-dine dasami aparajita-(puja-yog)ya na prapyate tada navamiyukta-dasamyam evodaya-sambandh-abhave 'pi visarjanam karyam tasya tad-uttarakala-kartavyatvat / "ekadasyam na kurvita pujitam caparajitam" (ity a)nena ekadasiyukta-dasami-nisedha etad-visaya eveti tan na aparajita-pujanasya kalah .....

14 STUDIES IN THE UPAPURANAS in Bengal, Mithila and Kamarupa, and they do not include 'yava' or any other plant which is peculiar to Western or Southern India. So, the verses of the 'Bhavisya-purana', 'Linga-purana', Nandikesvara-purana, Brhan-nandikesvara-purana, etc. mentioned above must have been written in Eastern India, mostly in Bengal. We shall see hereinafter that the Kalika-purana was written either in Kamarupa in Assam or in that part of Bengal which was very near to it, and the Devi-purana, Nandikesvara-purana and Brhan-nandikesvara-purana were written in Bengal. 54 As regards the verses ascribed to the Bhavisya-purana and the Linga-purana, it is disappointing to find that none of them occurs in the present texts of the Bhavisya and the Linga-purana On the other hand, manuscripts are still found in Bengal of Smrti works on the 'Bhavisya-puranokta-durga-puja-paddhati' ; 52 and Kasirama Vacaspati, in his commentary on Raghunandana's Tithi-tattva, testifies to the fact that the method of Durga-puja, prescribed by the 'Linga-purana', was once followed in Bengal. 53 Regarding the verse 'vrati prapujayed devim', ascribed to the 'Bhavisya' in Raghunandana's Tithi-tattva, Gadadhara says that it was respected in 'Gauda-desa' but not in Orissa. 55 We can reasonably presume, therefore, that chapters consisting of verses on the methods of Durga-puja, as followed in Bengal, were written by the scholars of this province and inserted into the texts of the Bhavisya and the Linga-purana current among them, so that these methods, though determined mostly by the local customs of Bengal, might be regarded as based on authoritative Sastric injunctions. But all the verses ascribed to the 'Bhavisya-purana' and the 'Linga-purana' were not written at the same time. For instance, the lines 52 See, for instance, Dacca University Manuscripts Nos. 2261 and 4055, in which the work is called 'Brhan-nandikesvara-puran-anugrhita-bhavisya-puranokta-durgapujapaddhati'. 53 See Kasirama's commentary-evam ca lingapuran-anusarena cet puja kriyate tada navamyam astadasa-bhujaya bodhanam, kalikapuran-anusarena cet puja kriyate tada navamyam dasa-bhujaya bodhanam iti bodhyam /-Tithi-tattva, pp. 248-9. 5 See Smrti-tattva, I, p. 87. 55 Kalasara, pp. 106-7-yat tu 'vrati prapujayed devim saptamyadi-dina-traye / dvabhyam catur-ahobhir va hrasa-vrddhi-vasat titheh //' iti tithi-tattva-karair bhavisyavakyam udahrtam tad gauda-dese adriyate / asmad-dese tu............... nadriyata eva ata eva mahastami-paddhati-karadibhih pracinair api tad vakyam anadrtam eva.

THE SAKTA UPAPURANAS 15 "sasthyam sayam prakurvita bilva-vrkse 'dhivasanam" and "svetaparajita-baddhah sarva-sampat-samrddhidah", which record two of the most characteristic customs of Bengal and Mithila and which have been ascribed to the 'Linga-purana' by Srinatha Acarya-cudamani, were added at a comparatively late date. It may be that Srinatha himself added these two lines to those already current in his time under the name of the 'Linga-purana' We have already mentioned the names of more than half a dozen Shakta Upapuranas which are still extant, and from our examination of these works we shall see that all of them were not written at the same time or at the same place. Hence the question arises as to why the people of later ages felt encouraged to write new Puranic works or chapters on Sakti-worship, although they inherited works of similar nature from their predecessors. In the following pages we shall try to answer this question before we proceed to analyse the individual Upapuranas. We have already seen that the peculiar customs of the people of Eastern India encouraged them to compose new verses and even write new Puranic works in order to furnish these customs with a Sastric basis. People of other provinces 56 The late date of these two lines is shown by the following facts: In his Durgotsava-viveka, p. 43 Srinatha ascribes to the 'Linga-purana' nine metrical lines, in which the line 'sasthyam sayam prakurvita' occupies the fifth place. Of these nine lines, lines 3-4 have been ascribed to the 'Bhagavati-purana' in Kv, p. 512, but there is no mention of line 5 (sasthyam sayam prakurvita, etc.) in the Kalaviveka. Lines 2-4 and 6-9 agree with Kalika-purana 62.17-20, but line 5 (sasthyam sayam prakurvita etc.) is not found there. The Samvatsara-pradipa quotes lines 6-8 but not line 5. Sulapani and Raghunandana support the performance of Devi's adhivasa in a Bilva tree in the evening of the day previous to that of patrika-pravesa on the authority of a verse derived from a 'Smrti' but do not quote the line 'sasthyam sayan etc.' Govindananda informs us that this line (sasthyam sayam) was not found in the Nibandhas of Mithila and other provinces (see footnote 46 above). Hence the line 'sasthyam sayam prakurvita' must have been added later. Of the three lines (rambha kacci haridra ca, etc.), ascribed to the 'Linga-purana' in Srinatha's Durgotsava-viveka, p. 50, the third is 'svetaparajita-baddhah sarva-sampatsamrddhidah'. Although the first two lines are quoted anonymously in Samvatsarapradipa, fol. 24 a and Varsa-kaumudi, p. 401, and there is mention of the custom of tying the Nava-patrika with an Aparajita creeper in these two works as well as in the Durgabhakti-tarangini (pp. 63 and 129) and the Durga-puja-tattva (p. 51), the line 'svetaparajita-baddhah' is not found in any of these works. Hence there can be little doubt regarding the comparatively late date of this line.

16 STUDIES IN THE UPAPURANAS also must have made similar attempts for the spread of Saktiworship. But this was not the only cause of composition of new works and chapters on Sakti-worship in later days. There were other reasons, which will be evident from the following survey of the rise of Puranic Saktism in India. A perusal of the extant Puranas indicates that to the writers of these works the problem regarding Devi was not how she originated but how she was to be associated and conceived of. Yet the question of her origin played a great part in determining her nature and association. There is no denying the fact that in Devi, as we have her now, we have glimpses of Vedic deities, especially the earth-goddess Prthivi, but it must be admitted that in her present character she is pre-eminently a deity of non-Vedic origin. Modern research has made it sufficiently known that Mother-worship, in some form or other, prevailed from very ancient times among the different races and tribes, Aryan or non-Aryan, living in the vast tract of land extending from Greece to India. In Greece the only divinity certainly known to be of pre-historic origin is the Mountain-Mother, who is found figured in 'a sealimpression of late Minaon style (circa 1500 B. C.) found at Knossos'. In this impression the goddess stands on a mountain-top with a lion on either side. An ancient Greek work of about the ninth century B. C. identifies the Mother of the Gods with the Phrygian Mountain-Mother, who is the mistress of the swift-slaughtering lions. Pausanias, a Spartan of the fifth century B. C., records the existence of two ruined temples of the Mother of the Gods in Arcadia and also of two lions made of stone in the neighbourhood of one of them. In an Attic relief, dated about 400 B. C., the Mother Goddess appears as seated on a throne, with two lions couching at her feet. According to Sir John Marshall, 'female statuettes akin to those from the Indus Valley and Baluchisthan have been found in large numbers and over a wide range of countries between Persia and the AEgean, notably in Elam, Mesopotamia, Transcaspia, Asia Minor, Syria and Palestine, Cyprus, Crete, the Cyclades, the Balkans V T

THE SAKTA UPAPURANAS 17 and Egypt. The Mahabharata, 58 Harivamsa," and the Puranas tell us that in early times female deities of different forms and names were worshipped in different parts of India" by the followers of the Vedas as well as by the Savaras, Varvaras, Pulindas, Kiratas, and many other non-Aryan tribes. These female deities appear in these works generally as divine mothers 63 associated as spouses with particular male gods, but much more prominently as virgin deities sporting on mountain-tops and being the sources of origin of the divine mothers. Thus, Durgi or Durga, a great mountain-goddess associated very often with the Himalayas, is called a virgin in early sources such as the Taittiriya-aranyaka and the Mahabharata: Vindhyavasini also appears as a virgin deity in all works; and in the Markandeya, Devi and other Puranas Devi, whether identified with Uma or with Vindhyavasini, is found to kill the demons in her virgin 57 Sir John Marshall, Mohenjo-daro and the Indus Civilization, Volume I, p. 50. 58 IV (Virata-parvan), chap. 6, and VI (Bhisma-parvan), chap. 23. 59 II (Visnu-parvan), chapters 2-4 and 22. 60 Viz., Visnu-purana V. 1; Markandeya-purana, chapters 81-93; Bhavisya-purana IV. 138; Devi-purana; Kalika-purana; Varaha-purana, chapters 21-28 and 90-96; and so on. 61 Bhavisya-purana IV. 138. 9 b-rupa-bhedair nama-bhedair bhavani pujyate siva; Visnu-purana V. 1. 80-tatah sumbha-nisumbhadin hatva daityan sahasrasah / sthanair anekaih prthivim asesam mandyisyasi //; Harivamsa II. 2. 49-tatah sthana-sahasrais tvam prthivim sobhayisyasi (spoken of Vindhyavasini). 62 vasas tava mahadevi vanesupavanesu ca / savarair varvarais caiva pulindais ca supujita /| Harivansa II. 3. 6 b-7 a. saravarsavarais capi (? savarair varvarais capi) pulindais capi pujyase / vindhyavasini vasaughe amoghe ambike subhe // Devi-purana 127. 110 b-111 a. (See also Devi-purana 39. 142). sura-mamsady-upaharair japa-yajnair vina tu ya / vina mantrais tamasi sa kiratanam ca sammata // (ascribed to the 'Skanda' in Sulapani's Dv, p. 3, and to the 'Skanda' and the 'Bhavisya' in St, I, p. 68, Dpt, p. 30, and Kalasara, p. 111). 11 See also Bhavisya-purana IV. 138. 27-30 a (evam vindhyopavasinya navaratropavasinah /. ...sthane sthane pure pure / grame grame vane vane / . . . . mlecchair anyais ca 1). According to the Kalika-purana (chap. 39) Kamarupa was a seat of Devi-worship even when it was inhabited by Kiratas. The remarkable absence of Saktism among the Tamils (of Southern India) tends to show that these people were originally not associated with Sakti-worship. 63 The number of the divine mothers is generally given as eight or sixteen. 64 Taittiriya-aranyaka X (Narayanopanisad). 1; Mahabharata IV. 6.7, 14, and VI. 23.4. 2 Sp

18 STUDIES IN THE UPAPURANAS state. It is highly probable that the non-aryan tribes mentioned above had a matriarchal system of society 66 and that it was due to this system that the custom of worshipping female deities grew among them. In spite of the spread of the Aryans in India, a large section of the members of these tribes, who lived in impenetrable mountains and forests and were thus able to maintain their freedom and individuality, used to worship, as they still do, their deities in these inaccessible parts of the country 68 with the offer of wine, meat and other things. It is highly probable that on these occasions they sacrificed human beings", drank wine", 65 See, for instance, Devi-purana 35. 17 b-18 akanya devya svayam prokta kanya-rupa tu sulini / yavad aksata-yonih syat tavad devya surariha // In Devi-purana 7.39 also Devi is called 'kanya-rupa'. 66 R. P. Chanda, Indo-Aryan Races, pp. 153-156. The Vedic Aryans, being born and brought up in a patriarchal state of society, conceived of their principal deities as male. The female deities occupy a subordinate position in the Vedic literature, no matter whether they appear as wives of great gods or not. In comparatively late periods, however, the mother came, in some Aryan families, to be regarded as more important than the father, and this is shown by the fact that the male members of these families were named after their mothers. 67 It is not known definitely whether these aboriginal tribes originally dwelt in mountains and forests, or they were driven to these places by the invading Aryans. In the Atharvaveda the Kiratas have been mentioned as living in mountains. (See Atharvaveda X. 4. 14-kairatika kumarika saka khanati bhesajam / hiranmayibhir abhribhir girinam upa sanusu //). 68 Devi is said to be living in mountains and forests in Mahabharata VI. 23. 11 b-14 (... ... ... ... ... ... durge kantara-vasini / kantara-bhaya-durgesu bhaktanam calayesu ca/ nityam vasasi patale //), Harivamsa II. 3.6 (parvatagresu ghoresu nadisu ca guhasu ca / vasas tava mahadevi vanesupavanesu ca //), Devi-purana 17.17 (hemakute mahendre himadrau mahidharini vindhya-sahyalaye srigirau samsthite) and 17.23 (saila- srngesu tungesu vata kantara-vasini), and so on. Cf. also Mahabharata IV. 6.20 b-21 (kantaresv avasannanam ll ...... ..tvam gatih kantaresv atavisu ca //); Harivarsa II. 2.54 (kantaresv avasannanam tvam gatih 11). 69 Human sacrifice is allowed even by the present Kalika-purana (chap. 71). Reference is made in the Dasa-kumara-carita (Purva-pithika, Ucchvasa I) and other works to the sacrifice of human beings to deities by Kiratas, Savaras and other tribes. 70 For references to the use of wine and meat in Devi-worship as well as to Devi's liking for these things see Harivamsa II. 22. 53 b-54 a (arcyate dasyubhir ghorair mahabali-pasu-priya | sura-pisita-purnabhyam kumbhabhyam upasobhita //), Visnu-purana V. 1.84 a (sura-mams-opaharais tu bhaksya-bhojyais ca pujita), and Bhavisya-purana IV. 138. 98 a (surasavabhrtaih kumbhais tarpayet paramesvarim). It is to be noted that chap. 138 of Bhavisya-purana IV deals with the method of Devi-worship during the Navaratra-vrata. See also Mahabharata IV. 6. 17 b (sidhu-mamsa-pasu-priye), Harivamsa II. 2. 52, and Q 3 £<<& (R) 3 ws¢$ £ (R) ང

THE SAKTA UPAPURANAS 19 indulged in frantic revels, and practised sexual promiscuity". The aboriginal tribes seem to have worshipped their female deities mainly for protection against ferocious wild beasts and ghosts 73. They also appear to have believed 3. 12 (sura-mamsa-bali-priya), Devi-purana 9. 56 (rudhira-mamsa-madya-priye), etc. as well as the verse 'sura-mamsady-upaharaih, etc.' quoted in footnote 62 above. 71 In Devi-purana 39. 142 Savaras, Pulindas and other aboriginal tribes are said to have worshipped Devi according to the 'Vamacara-marga.' Note also the Savarotsava mentioned in the Kalika-purana (62.31 and 63.18), as well as in the verses ascribed to Satya, Bhavisya-purana, Linga-purana, Brhan-nandikesvara-purana, ctc. in the Bengal Nibandhas. (See footnote 20 above). 72 Harivamsa II. 3.7-8- savarair varvarais caiva pulindais ca supujita / mayura-piccha-dhvajini lokan kramasi sarvasah // kukkutais chagalair mesaih simhair vyaghraih samakula / ghanta-ninada-bahula vindhya-vasiny abhisruta // Devi-purana 17.26-mrgendrair grhitam gajendrair vibhinnam viluptam bhujangais ca dastam vane capi mudham rane hiyamanam mahagraha-grastam mateva samraksase putravan nityam. Markandeya-purana 92.25-29 aranye prantare vapi dasyubhir va vrtah sunye grhito vapi satrubhih // simha-vyaghr-anuyato va vane va vana-hastibhih | smaran mamaitac caritam naro mucyeta samkatat // mama prabhavat simhadya dasyavo vairinas tatha / durad eva palayante smaratas caritam mama // And so on. ..... khagendrair The story of Devi's killing of the demon Mahisa (who had the form of a buffalo) in a previous Kalpa, and the tradition that whenever Devi killed the demons, she had a lion as her mount (vahana), seem to be based on the aboriginal conception of Devi as a spirit controlling wild beasts. The bell (ghanta), which is said to be carried by Devi, might have been originally meant for scaring away wild beasts. (For references to Devi's carrying a 'ghanta' see Mahabharata IV. 6.10, Harivamsa II. 3.8, Markandeya-purana 82.21 and 84.23, Devi-purana 17.23, and so on). In Southern Bengal a deity popularly called 'Daksin-ray' (Daksina-raya) is worshipped by villagers for getting rid of the havoc created by Royal Bengal tigers of the Sundarvans. This deity has an ugly face very much like that of a tiger but its body is that of a human being. 73 In Mahabharata IV. 6.18 and Harivamsa II. 2.52 Devi is said to be followed by ghosts (krtanuyatra bhutaih). In commenting on the Mahabharata Nilakantha takes the word 'bhuta' (occurring in Mahabharata IV. 6.18) to mean 'Brahma and others' (bhutair brahmadyaih pracinaih). But Nilakantha's interpretation is hardly acceptable. See also Harivarsa II. 2.46 (kirna bhuta-ganair ghoraih), and 22.53 (bhutasamgha-nisevita); Bhavisya-purana IV. 138.1-2 (punya mahanavamy asti ../ sanustheya 11 .. .... bhuta-preta-pisacanam prity-artham cotsavaya ca //); Devi-purana 31.15- 17 a (dadyac ca dig-balim sakra sarva-diksu samanvitah / bhuta-vetala-samghasya mantrenanena suvrata // jaya tvam kali bhutesi sarva-bhuta-samavrte / raksa mam nija-bhutebhyo balim grhna siva-priye //) and 17.17 (pramatha-laksair vrte

20 20 STUDIES IN THE UPAPURANAS that these deities were able to protect them in all kinds of danger and to confer on them wealth and prosperity as well as victory in war. As a matter of fact, the occasions for Devi's appearance on the Himalayas and the Vindhya, as given in the Markandeya-purana, Devi-purana, etc., show that she is conceived of in these works primarily as a war-goddess, with whom other female deities were identified. The very name of Sakambhari as well as her story given in Markandeyap. 92.42-46, and the fact that Durga is called 'kantaravasini' and is said to protect those who fall in trouble (avasanna) in deep forests, tend to show that there were some female deities who were looked upon by the aborigines as vegetation spirits. It should be mentioned here that the character of Devi as a war-goddess explains why she is to be worshipped in autumn and spring, which are highly favourable for military expeditions. We do not know the number and names of the female deities originally worshipped by these aboriginal tribes, and among the Sakta deities of the Puranas and Tantras there are certainly some who owe their origin to the deification of abstract ideas; but it admits of little doubt that many of the Sakta deities of the Puranas and Tantras, viz., Uma, Kausiki, Vindhyavasini, Durga, Candi, Kali, Kalika, Camunda, Kamakhya, Sakambhari and others, were modelled on the popular ones, especially those associated with mountains, viz., Himalaya and Vindhya. It is remarkable that the comparatively early Vedic works do not betray in any way their recognition, or even knowledge, of these Shakta deities, nor do they speak of the lion as a mount of any of their candike); Markandeya-purana 92.18 (rakso-bhuta-pisacanam pathanad eva nasanam); and so on. Bhavisya-purana IV. 136 describes a festival known as Bhutamatrutsava (the festival of the mother of evil spirits), which is most probably borrowed from the Savaras and in which an ugly and hideous-looking female deity called Bhuta-mata, is worshipped for protection of children from the pernicious influence of evil spirits. This Bhuta-mata, who is clearly of non-Aryan origin, is said to be born of Parvati's urine and is thus called a form of Bhagavati herself. ** The conception of Devi as a war-goddess is certainly not of very late origin. In Mahabharata VI. 23.4 and 8 Durga is called 'siddhasenani (generaless of the Siddhas) and 'rana-priya' (fond of war). See also Markandeya-purana, chapters 81 ff., Bhavisya-purana IV. 138, Devi-purana, and so on.

THE SAKTA UPAPURANAS 21 goddesses. On the other hand, in the Harivamsa Devi is said to be worshipped as Vindhyavasini by Savaras, Varvaras and Pulindas, to be crowded by cocks, goats, sheep, lions and tigers, and to make a high and continuous sound with her bell. So, we find in Vindhyavasini an interesting parallel not only of the Mountain-Mother of Crete and Greece who is the mistress of lions but also of the Mother of the Gods, who is addressed in Homeric prelude 'as rejoicing in the clash of cymbals, the beating of drums, the blare of pipes, and the roar of wolves and lions.' But the worship of neither the Puranic or Tantric Devi in India nor her parallels in Crete and Greece can be said to be a development from the vague conception of the Mother Earth. Of whatever nature the female deities of the aboriginal tribes may have been, they were not allowed an easy access into the Vedic pantheon. The military spirit of the Vedic Aryans as well as their deep-rooted feeling of antagonism towards the natives stood in the way of their having any regard for the deities of the latter. As a matter of fact, none of the names of the different forms of Durga is mentioned in the Vedic Samhitas and the Brahmanas. It was only at a much later date that the aboriginal deities began to be admitted by a very slow process of assimilation. In the different texts of the Yajurveda as well as in its Brahmanas there are, of course, passages in which Ambika has been mentioned and often called Rudra's sister" and once his 75 For the relevant verses of the Harivamsa see footnote 72 above. 76 See Taittiriya-samhita I. 8.6.1-esa te rudra bhagah saha svasrambikaya tam jusasva....../ Kapisthala-katha-samhita VIII. 10-same as in the Taittiriya-samhita quoted above. Kathaka-samhita IX. 7-esa te rudra bhagah saha svasrambikaya tam jusasva svaha; and XXXVI. 14-esa te rudra bhagah saha svasrambikaya tam jusasva svaheti sarad vai rudrasya svasambika tam eso 'nvavacarati tasmad esa saradi bhuyistham hanti .............. / Maitrayani Samhita I. 10.4 esa te rudra bhagas tam jusasva saha svasrambikaya svaha; and I. 10.20-esa te rudra bhagas tam jusasva saha svasrambikaya svaheti sarad vai rudrasya yonih svasambikaitam va eso 'nvabhyavacarati tasmac saradi bhuyistham hanti tayaivainam saha niravadayate 1

22 STUDIES IN THE UPAPURANAS mother as well", but in these passages Ambika, unlike Durga and other Sakta deities 78 of the Puranas, Tantras, etc., has been allowed a share of the sacrificial offerings. She, therefore, does not seem to have been derived from a mother goddess of the non-Aryans. According to the Kathakasamhita, Maitrayani Samhita and Taittiriya-brahmana this Ambika, sister of Rudra, was the autumn season (sarat) personified. The two Samhitas further tell us that as Rudra followed his sister Ambika, who was identical with autumn (sarat), his work of killing reached its climax in this season. The Taittiriya-brahmana, on the other hand, says that it was with his sister Ambika appearing as Sarat that Rudra carried out his work of killing. So, commenting respectively on Taittiriya-brahmana I.6.10.4 and Vajasaneyi Samhita 3.57 Sayana and Mahidhara describe Ambika as a cruel deity (himsika, krura-devata) like Rudra and say that Ambika helped Rudra in his slaughter by appearing as Sarat and creating fever and other diseases 80. As a matter of fact, Vajasaneyi Samhita 3.57-same as Kathaka-samhita IX. 7 quoted above. Taittiriya-brahmana I. 6.10.4 esa te rudra bhagah saha svasrambikayetyaha | sarad va asyambika svasa | taya va esa hinasti / yam hinasti tayaivainam saha samayati / See also Satapatha-brahmana II. 6.2.9-esa te rudra bhagah saha svasrambikaya tam jusasva svahetyambika ha vai namasya svasa tayasyaisa saha bhagah and II. 6.2.13-sa rudrasya svasambika nama sa ha vai bhagasyeste ; 1 77 See Maitrayani Samhita I. 10.20 quoted in the immediately preceding footnote. 78 The Hiranyakesi-grhya-sutra (II. 8.7) prescribes sacrificial offerings 'to god Bhava's wife' (Bhavani ?)-bhavasya devasya patnyai. But this Grhya-sutra comes from a comparatively late date. 7 Note the words 'sahaya-bhutaya' and 'sadhana-bhutaya' used by Sayana and Mahidhara respectively with regard to Ambika in commenting on Taittiriya-brahmana I. 6.10.4 and Vajasaneyi Samhita 3.57. For extracts from the commentaries see the immediately following footnote. 80 In commenting on Taittiriya-samhita I. 8.6.1 Sayana quotes Taittiriya-brahmana I. 6.10.4 and explains it as follows: sarat-kalo hi pinasa-jvar-ady-utpadanena himsakas tadvad iyam ambika himsika / tatah sarad ity ucyate / esa rudras tayaiva sahaya-bhutaya praninam hinasti / atas taya saha purodasa-bhaga-sevaya tustaya tayaiva sahaivainam rudram samayati himsa-rahitam karoti // Mahidhara explains Vajasaneyi Samhita 3.57 thus: ambikaya rudra-bhaginitvam srutyoktam / ambika ha vai namasya svasa tayasyaisa saha bhaga iti yo 'yam rudrakhyah kruro devas tasya virodhinam hantum iccha bhavati tadanaya bhaginya krura-devataya sadhana-bhutaya tam hinasti / sa cambika sarad-rupam prapya jvaradikam utpadya tam virodhinam hanti / rudrambikayor ugratvam anena havisa santam bhavati / tatha ca tittirih-esa te rudra bhagah saha svasrambikayetyaha sarad va asyambika sa bhiya esa

THE SAKTA UPAPURANAS 23 the autumn season (sarat) has been looked upon from the early Vedic period as the most dangerous part of the year, so much so that a person who could pass this season safely was taken to be out of danger for the remaining part, that often the age of a person was expressed and the years counted in terms of the autumn seasons passed, and that this season was mentioned in blessing one with, or aspiring for, a long life¹. So, the identification of Ambika with Sarat can by no means be taken to indicate that she was originally a harvest deity and that her identification with Durga gave rise to the widespread custom of the latter's worship in autumn. On the other hand, it appears that the custom of worshipping Durga mainly as a war-goddess in autumn owed its origin much more to her identification with the cruel and destructive deity Ambika than to the suitability of this season for war expeditions, because it was quite natural for the warriors to believe that by propitiating Durga with worship and animal sacrifice they could escape death and also create havoc on their enemies. The connection of Durga with harvest, though coming from an early date, decidedly forms a comparatively late phase in her character as a deity. But very different is the case with Uma, who, for her peculiar name, her association with a mountain, and her mount (vahana), a lion, seems to be originally the same as the Babylonian Ummu or Umma, the Arcadian Ummi, the Dravidian Umma, and the Scythian Ommo, 82 which are all mother goddesses. This Uma is neither mentioned in any of the Vedic Samhitas and Brahmanas nor allowed any share of the sacrificial offerings. Even in the story of the destruc- (? -ka svasa taya va esa) hinasti yam hinasti tayaivainam saha samayatiti (Taittiriya-brahmana I. 6.10.4). ... 81 See, for instance, Rg-veda I. 72.3 (tisro .... saradah), 89.9 (satam innu sarado), II. 27.10 (satam no rasva sarado ..), III. 36.10 (asme satam sarado jivase), etc.; Atharva-veda I. 10.2 (satam jivati saradah), II. 13.3 (satam ca jiva saradah), III. 11.4 (satam jiva sarado vardhamanah), 12.6 (satam jivema saradah), etc.; Vajasaneyi Samhita 25.22 (=Rg-veda I. 89.9), 35.15 (satam jivantu saradah), 36.24 (...... pasyema saradah satam jivema saradah satam srnuyama saradah satam, etc.); and so on. See also Raghu-vamia 10.1, Malavikagnimitra 1.15, etc. 8 The goddess appears as 'Ommo' in Huviska's coin, **

24 STUDIES IN THE UPAPURANAS tion of Daksa's sacrifice, as given in Mahabharata XII. 283-4, it is only Siva, and not his consort Parvati, who is promised a share in the sacrifice. The earliest works mentioning Uma are the Taittiriya-aranyaka and the Kena-upanisad. In the former work Uma as well as Ambika is associated with Rudra, who is there called Ambika-pati (Ambika's husband) and Uma-pati (Uma's husband), 83 and in the latter she is called Haimavati 84 (daughter of, or belonging to, Himavat). The Taittiriya-aranyaka mentions Durgi (i.e. Durga) also, but calls her 'kanya-kumari' (virgin). So, according to the Taittiriya- aranyaka Uma and Durgi (i.e. Durga) were different. It is highly probable that originally this Durgi was, like Uma, a mountain goddess connected with the Himalaya. In the Mundakopanisad (1.2.4.) 'Kali' and 'Karali' are given as the names of two of the seven tongues of Agni. These references indicate that towards the end of the Vedic period, some female deities of non-Vedic origin began to be associated with the Vedic gods in various ways, and that Uma was the first non-Aryan deity to be regarded as the wife of a Vedic god, whose character had already been much modified by the assimilation of alien characteristics. But during this period the relation between Rudra and Uma was of a very ordinary nature, like that between a husband and his wife, and it was not characterised by the relation between Purusa and Prakrti of the Samkhya system or Brahma and Maya of the Vedanta. Towards the end of the Vedic period the high spirit of domination by military power and of maintenance of racial and cultural distinction of the Vedic Aryans began to calm down to a very great extent, and there was a growing tendency towards cultural assimilation. Now, the work of assimilation was mainly taken up by the authors of the Mahabharata and the Puranic works, who believed deeply in 83 Taittiriya-aranyaka X (Narayanopanisad). 1-katyayanaya vidmahe kanyakumari dhimahi / tanno durgih pracodayat || And X. 18-namo ambika-pataye uma-pataye namo namah. 84 Kena-upanisad 3.12-sa tasminn evakase striyam ajagama bahu-sobhamanam umam haimavatim tam hovaca .11 1

THE SAKTA UPAPURANAS 25 the theory of rebirth, gave a very high place to the Samkhya and Vedanta systems of philosophy, and explained the nature of the gods and goddesses and the universe through a fusion of the principles of these two systems. These authors had much broader views than the orthodox followers of the Vedic religion and, unlike the Vedic Aryans, tried to bring the antagonistic non-Aryans into the Aryan fold by making a wise compromise between the Vedic and non-Vedic ideas. But as no female deity could be given a pre-eminent position in the recognised pantheon without going against the Vedic tradition and the principles of Samkhya and Vedanta, they connected some of their female deities with Siva and some with Visnu and explained the relation of these gods and goddesses with the principles of Samkhya and Vedanta. We have already mentioned that in the Taittiriya-aranyaka Uma, an originally Himalayan deity, was connected in a very general way with Rudra. So, Uma was already recognised as a Hindu deity. This well-established position of Uma tempted · the Saiva authors of later ages to regard her as the Prakrti and Maya of Siva and to look upon the other goddesses as the different forms of Uma. The Vaisnava authors, on the other hand, picked up Vindhyavasini, the most prominent among the female deities of the Vindhya, and connected her with Visnu by taking her to be an incarnation of Visnu's Yoga-nidra (or Yoga-maya). With this Yoga-nidra the Vaisnavas seem to have identified a deity of the Abhiras also. It is probable that like the people of Punic Africa, Egypt, Phoenicia, Asia Minor, and Greece of early times, the Abhiras also worshipped a maiden ('unwed') goddess with a young subordinate god. When the Abhiras came to India, this goddess was no longer 'made first the mother of her companion by immaculate conception, and then of the gods and all life by the embrace of her own son', because such an idea was repulsive to the people of India, but the young subordinate god was identified with Visnu and the goddess with the maiden Vindhyavasini, who also was connected with Visnu as his Yoga-maya. So, through the story of Krsna's birth, the votaries of Visnu managed to include both the

26 STUDIES IN THE UPAPURANAS Abhiras and the worshippers of Vindhyavasini into their fold 85. In course of time, however, people conceived of a central Devi to represent Prakrti and Maya of Samkhya and Vedanta respectively and all the goddesses (including Uma, Vindhyavasini, etc.) to be her different forms assumed at different times. But being encouraged by their firm faith in the non-duality of God, as well as by their spirit of sectarian rivalry, the Saivas claimed that this central Devi was none but Uma and that Vindhyavasini, Kali, Camunda and other goddesses were Uma's incarnations. The Vaisnavas, on the other hand, said that it was Visnu's Yog-maya who manifested herself as Uma, Sati, Kali, etc. for the good of the world. These contending sectaries fabricated various kinds of stories often in conformity with philosophical principles to support and popularise their respective views. How conflicting these stories were, will be evident from the following Puranic accounts of Devi's original nature and her assumption of various forms under different circumstances. In Markandeya-purana, chapters 81-93 it is said that Visnu's Yoga-nidra (also called Mahamaya) assumed a Tamasa form at the time of rousing Visnu from his sleep. She next manifested herself on the Himalaya 86 as a collection of energy issuing from the bodies of Brahma, Visnu, Siva and other gods and was called Ambika, Candika, Bhadrakali and Durga. She had a thousand hands and a lion as her mount (vahana) and killed Mahisasura. Next, being eulogised by the gods on the Himalaya for killing Sumbha and Nisumbha, Visnu-maya (called Siva and Ambika) came out as Kausiki from the body of Parvati when she was going to bathe in the Ganges. Parvati's complexion turned dark, and she became known as Kalika and continued to live in the Himacala. 85 Cf. Bhagavata-purana II. 4.18- kirata-hunandhra-pulinda-pukvasa abhira-suhma yavanah khasadayah | ye'nye ca papa yad-upasrayasrayah sudhyanti tasmai prabhavisnave namah // 86 Cf. Markandeya-purana 82.28 adadat himavan vahanam simham ratnani vividhani ca // It is to be noted that among the mountains, only Himavat is mentioned as making presents to Devi.

THE SAKTA UPAPURANAS 27 Candika (also called Ambika and Durga), in the extremely beautiful form of Kausiki, lived on the Himalaya and was asked by Sumbha to marry himself or his brother. As she refused to do so on the pretext of a pledge, war broke out between Candika (i.e., Kausiki) and Sumbha and Nisumbha. After issuing from the forehead of Kausiki (Ambika), Kali killed Canda and Munda (sent by Sumbha to capture her) and was known as Camunda. Sumbha next came to the battle-field, and Candika (i.e., Kausiki) was helped by Kali (Camunda), the lion, and the Saktis of Brahma and others (viz., Brahmani, Mahesvari, Kaumari, Vaisnavi, Varahi, Narasimhi and Aindri). 'Then, being surrounded by these Deva-saktis Isana said to Candika, "Soon kill the Asuras for my pleasure." Then, from Devi's body came out the extremely fierce and dreadful Candika-sakti, who was attended with the howlings of hundreds of jackals'. 87 This Candika-sakti sent Siva to Sumbha and Nisumbha as a messenger and was known as Siva-duti. Now the war began with Kali (i.e., Camunda), Siva-duti, Brahmani, Mahesvari, Kaumari, Vaisnavi, Varahi, Narasimhi and Aindri, who are called Matrs 88. Brahmani and other Deva-saktis fought with the demon Raktabija, but Candika killed this demon with the help of Camunda who drank the demon's blood before its fall to the ground. Candika next killed Nisumbha with the help of Kali, Siva-duti and Brahmani and other Deva-saktis. Being censured for her assistants Devi Candika absorbed into herself all her vibhutis (viz., Brahmani and others), fought alone with Sumbha, and killed him. Being eulogised by the gods for killing Sumbha and Nisumbha, Devi (who is described as 'parama maya') said that in the Vaivasvata Manvantara two other demons named Sumbha and Nisumbha would be born, and that they would be killed by Devi born of Yasoda in the house of Nanda-gopa and living on the Vindhya (vindhyacala-nivasini). She further assured the gods that in future she would assume the following incarnations. (a) She would eat up the Vaipracitta Danavas 87 Markandeya-purana 88. 21-22. 88 Ibid., 88. 38-39 and 49. $3

28 STUDIES IN THE UPAPURANAS in such a way that her teeth would become as red as pomegranate-flowers, and thus she would be known as Raktadantika. (b) Being eulogised by sages during a drought lasting for a hundred years she would have an immaculate origin and be known as Sataksi and Sakambhari, respectively for looking at the sages with a hundred eyes and for maintaining, until rainfall, the whole creation by means of vegetables originating from her own body. During that drought she would kill a powerful demon named Durgama and be known as 'Durga-devi'. (c) She would again appear in a terrible form (bhima-rupa) on the Himalaya, exterminate the Raksasas, and be famous under the name of 'Bhima Devi'. (d) She would kill the demon Aruna by taking the form of a bhramara (bee) and be known as Bhramari. According to Devi-purana, chapters 4-9 and 13-20, a demon named Dundubhi saw Uma occupying the left half of Siva's body and got enamoured of her. Consequently, Siva reduced him to ashes by an angry glance. From these ashes rose another demon who also had a strong desire for Uma. So, Uma cursed this dreadful (ghora) demon to go down to the earth. Finding that Uma thus spared the life of this demon, Siva reproached her for her foolishness. Uma became angry with Siva and blessed the demon to have his residence in Kusa-dvipa, become the lord of the three worlds and be invincible to the gods. Siva also got annoyed at Uma's conduct and cursed her saying, "You will go down to the world of mortals, and there this vile creature will want to become your husband." Uma also said that if this dark-complexioned demon would do so, she would kill him sportively by mounting a lion (pancanana-vyavasthita). Now, in course of time, Dundubhi was reborn as Ghora and began to trouble the three worlds. At the prayer of the gods tyrannised by Ghora, Uma (who is described as the Adya Sakti and the Yoga-maya) came down to earth, sported in the Vindhyas as a virgin girl by mounting a lion constituted of all the gods and goddesses, and became known as Vindhyavasini. Being informed by Narada of this Vindhyavasini, Ghora went to the Vindhyas to capture

THE SAKTA UPAPURANAS 29 her. With the help of the gods and Matrs come down to the Vindhyas as her attendants, Vindhyavasini gave a tough fight and at last killed Ghora who had assumed the form of a buffalo (mahisa) before death. Thus Vindhyavasini became known as the killer of Mahisa. In the story given above, the virgin Vindhyavasini, who is said to be none but Adya Sakti Yoga-nidra incarnate, is identified with Uma come down to earth for killing Ghora. Thus, according to the Devi-purana, Uma is originally the same as Yoga-nidra, and it is Uma who is said to exist in different forms, viz., Durga, Sakambhari, Vindhyavasini, Kausiki, Camunda, Candi, Parvati, Daksayani, Vaisnavi Yoganidra, Ambika, etc. According to Bhavisya-purana IV, chap. 138 (which deals with the Puranic method 89 of the worship of Durga on the Mahastami and Mahanavami Tithis of Asvina), it is Bhavani (wife of Siva) who is worshipped by gods, men, Gandharvas, Kimnaras, Raksasas and others 'in different forms and under different names'.90 This Purana adds: "Bhavani is worshipped by the gods in all her repeated incarnations. She always kills demons by incarnating herself on earth and protects all creation in heaven, earth, and the nether world. This great goddess (mahadevi) was again born from the womb of Yasoda and killed the demon Kamsa by placing her foot on his head (?). From that time, people on earth have installed this 'giver-of-joy-to-Yasoda' on the Vindhya mountain and reintroduced her worship"."1 Thus Bhavisya-purana IV (i.e. the Bhavisyottara), though taking Bhavani to be the original Devi, clearly indicates that the deity, now worshipped on the Mahastami and Mahanavami Tithis of Asvina, is none but the virgin Vindhyavasini,"² who is said to have been born of Yasoda. From what has been said above it is evident that serious ritualistic and doctrinal differences were sometimes found 8 Cf. Bhavisya-purana IV. 138.41 b-puja-mantran pravaksyami puranoktan aham tava. so Ibid., IV. 138. 9-10 rupa-bhedair nama-bhedair bhavani pujyate siva. 91 Ibid., IV. 138. 22-25. ** Ibid., IV. 138. 18, 24-25, 27 and 91.

30 STUDIES IN THE UPAPURANAS among people of different climes and times, and it was often due to these differences that they felt it necessary to write new Puranic works. There were also other causes of such enterprise. With the spread of Saktism Kamarupa attained importance as a seat of the Sakta cult and greatly influenced the religious life and practice of the people of Eastern India, so much so that works like the Kalika-purana, Mahabhagavata-purana, Brhaddharma-purana, etc. highly glorify Kali (or Kalika), the principal deity of Kamarupa.93 Besides this Kali (or Kalika), many other local deities" grew into prominence, and new Sakta goddesses were invented. In order to give these new goddesses a position in the Sakta pantheon, they had to be connected either with Uma or with Vindhyavasini (or Yogamaya). Thus, in the Kalika-purana, Mahabhagavata-purana, Brhaddharma-purana, etc. Kali (or Kalika) has been taken to be the central goddess, of whom Durga, Ganga and others have been said to be the different forms; and in Devi-bhagavata IX. 1.136-7 the village goddesses have been regarded as parts of Devi (Mula-prakrti). There were also changes in the conception of these deities. But no less important for the compilation of new Puranic works was the call for social discipline (as understood by the followers of the Vedas) which was brought about by unfavourable circumstances not rarely due to political reasons. A good example in this direction will be found in the early history of Kamarupa, where, as the Mahabharata and the Puranas (including the present Kalika), the Harsa-carita of Banabhatta, and the inscriptions of Bhaskara-varman and others inform us, the dynasty of Naraka *³ Mahabhag. 77. 3 a (kamakhya kalika devi svayam adya sanatani), 4 a (...... kali kamakhya-rupini .), 5-6 (dhyayatam paramesanim kamakhyam kalikam param / rakta-vastra-paridhanam ghora-netra-trayojjvalam // caturbhujam bhima-damstram yuganta-jalada-dyutim / mani-simhasane nyastam simha-pretambuja-sthitam //), and so on. 94 Devi-bhagavata IX. 1.136-7--bahvah satyah kalas caiva prakrter eva bharate / ya yas ca grama-devyah syus tah sarvah prakrteh kalah // Ibid., IX. 1.158 pujita grama-devyas ca grame ca nagare mune. These lines are the same as Brahma-vaivarta-purana II. 1.138 (v. 1. 'santi' for 'satyah', and 'tah sarvas ca' for 'syus tah sarvah') and 157.

THE SAKTA UPAPURANAS 31 was ruling from very early times. Though mostly following Saivism 95 (which in its Kapalika form must have been the religious faith of the soil)," the rulers of this dynasty encouraged varnasrama-dharma with much enthusiasm. The Kalika-purana (chap. 39) tells us that after Naraka had defeated and killed Ghataka", king of the Kiratas, and driven out his faithful followers from Kamarupa, Veda-knowing Brahmins as well as sages and other members of the four castes were settled there and the country was made fit for Vedic study, rites and donations. Even Naraka himself has been said to have been born in Janaka's sacrificial ground in Mithila. Although there is much scope for doubt about the amount of truth contained in this story, it can hardly be denied that migration of Maithilas into Kamarupa, presumably for political and economic reasons, began quite early and there was a very close relation between Mithila and Kamarupa even before the time of composition of the present Kalika-purana From the testimony of the Padma-purana (Srsti-khanda), the Kalika-purana, and the Dharma-purana98 it appears that the Maithila Brahmins were generally full of respect for the Vedas, faithfully followed the varnasrama-dharma, and tried to maintain Smarta discipline in society. But with the passing of the political power to the 'Mleccha' (or Mech) dynasty set up by 95 The early epigraphic records of Kamarupa show that all the kings from Bhaskaravarman (if not from his earliest predecessor) down to Harsapala (father of Dharmapala of the Brahmapala dynasty) were devotees of Siva. Even Dharmapala, whose Puspabhadra Copper-plate Inscription (circa 1100 A.D.) testifies to his Vaisnava faith, was a Saiva in the earlier part of his life. 96 The Kalika-purana amply testifies to the prevalence of Kapalika Saivism in Kamarupa even before Naraka's arrival there. It also records a story in which Naraka is mentioned as lying, immediately after his birth, outside the sacrificial ground of Janaka with a human skull under his head. 97 In the Vangavasi Press (Calcutta) ed. (38. 99, 101) of the Kalika-purana the name of this king has been given as 'Ghataka'. 98 This Upapurana, though being an independent work and still occurring independently in Manuscripts, has been included in the printed editions of the Padma-purana, Srsti-khanda (of which it forms chapters 44-82 in the Poona edition, chapters 47-82 in the Calcutta edition, and chapters 49-86 in the Bombay edition). We shall see afterwards that it was written in Kamarupa between 1250 and 1325 A.D. It is remarkable that not a single of the Bengal Manuscripts of the Srsti-khanda includes the Dharma-purana, and of the Devanagari Manuscripts, so far known, it is only one which excludes it. For this Devanagri Manuscript see Eggeling, Julius Eggeling, A Descriptive Catalogue, VI, p. 1214, No. 3380 (Manuscript No. 215).

32 STUDIES IN THE UPAPURANAS Salastambha from about 655 A.D., Kapalika Saivism, which, in all ages, has been indifferent to the Vedic way of life, attained great popularity, Mleccha tendencies were encouraged, and the Maithila Brahmins of Kamarupa experienced a great social and economic set-back on account of their Vedic inclination. The Kalika-purana contains frequent references to the bad effects of the spread of Saivism, especially in its Kapalika form, and the Dharma-purana gives a dismal picture of the Hindu society, greatly consequent upon the rule of the Mleccha dynasty, in which neglect of the Vedas, Puranas, etc. and wilful breach of caste rules in every sphere of life were the orders of the day, people lost all sense of morality and earned their livelihood by learning the Mleccha-dialect (called Paisaciki Bhasa) and serving the beef-eating and immoral Mlecchas, and Brahmins, having social relation with Candalas and even marrying their girls, were seldom required to practise any severe penance or subjected to social stigma. In some verses of the Padma-purana (Srsti-khanda), which must have been added by the affected Maithilas, the local Brahmins of Kamarupa have been mentioned under the distinctive name of 'Parvatiya' and said to have universally followed a peculiar custom in which a piece of bone from the forehead of a deceased person (asthi lalatajam) was finely powdered, mixed up in a (silver) cup with Madhuparka and milk and curd, and offered to a Brahmin couple, who drank the mixture up before accepting the bed of the deceased person. 100 As this custom of bone-eating on particular 99 The name 'Parvatiya' for the people of Kamarupa is a very old one, being used on several occasions in the Mahabharata. (See, for instance, Drona-parvan 19.11 and 20). This name (Parvatiya) may be a Sanskrit equivalent for an Austric formation like 'Pagar-juh (jo)-tic' (meaning a region of extensive high hills), which seems to be the basis of the name 'Prag-jyotisa'. See B. K. Kakati, The Mother Goddess of Kamarupa, p. 6, and B. K. Barua, A Cultural History of Assam, Volume I, p. 11. 100 See Padma-purana, Srsti-khanda, 10. 13 b-18 a (which include the line 'esa eva vidhir drstah parvatiyair dvijottamaih'). The Venkatesvara Press (Bombay) and Vangavasi Press (Calcutta) editions read 'rajatasya' (for 'arghyam dattva'), and 'lalatikam' (for 'lalatajam'). All these verses have been quoted by Aniruddha-bhatta, Hemadri and Pitambara Siddhanta-vagisa in their Haralata (p. 199), Caturvarga-cintamani (III. i. pp. 1613 and 1614), and Preta-kaumudi (pp. 138-9) respectively, and the first takes the 'Parvatiyas' to be 'the inhabitants of Kamarupa, etc.'-parvatiyanam kamarupadivasinam. (It may

THE SAKTA UPAPURANAS 33 occasions is of non-Aryan origin, being found among the hilltribes of Assam, the Buddhist monks of Tibet, and the local Hindus of Mymensingh and the neighbouring districts of East Bengal and Assam, 101 the Parvatiya Brahmins were considered censurable, so much so that in a verse of the Padma-purana, Srsti-khanda, they have been declared unfit for invitation in a Sraddha ceremony. 102 Yet they enjoyed much greater state support; and in the Padma-purana, we hear of the clamour, made undoubtedly by the Maithila Brahmins of Kamarupa, about the enviable prosperity enjoyed by the Mlecchas and the Parvatiyas. This clamour appears in the form of Savitri's curse to Laksmi, the goddess of fortune, in the following verses: "naikatra vaso laksmyas tu bhavisyati kadacana/ ksudra sa cala-citta va murkhesu ca vasisyati/ mlecchesu parvatiyesu kutsite kutsite tatha/ murkhesu cavaliptesu abhisapte duratmani/ evamvidhe nare tubhyam vasatih sapa-karita//"'103 "Never shall Laksmi's residence be at one place; mean and be that by the word 'kamarupadivasinam' Aniruddha-bhatta meant 'the original inhabitants of Kamarupa', who were distinct from the Maithila Brahmins). According to Pitambara Siddhanta-vagisa, the piece of bone was powdered together with a diamond, and the custom was followed only by the Parvatiyas (. . . . lalatasambandhy-asthi grhitva hiraka-sahitam curnayitva rajata-patre nidhaya bhojayed ityarthah/ ..... parvatiyanam evayam sayya-dana-vidhih. The above verses do not occur in the Bengal Manuscripts of the Srsti-khanda. ). 101 On the occasion of Sri Jawahar Lal Nehru's visit to the eastern frontier of India in March-April, 1953, the chieftain of Tsawlaw, leader of a head-hunting tribe, and his lieutenants touched a tiger-tooth and chewed a bit of their ancestor's bone in an oath never to wage war again. Among the Tibetan Buddhists there is a custom that after the death of Dalai Lama his bones are powdered and eaten up by the monks, who believe that by doing so they make the Lama live among them even after his death. In Mymensingh and the neighbouring districts the custom of offering powdered bone of a deceased person to Brahmins still prevails, and such Brahmins are called 'Had-gila' (bone-eating) Brahmins. 102 Padma-purana, Srsti-khanda 27.67- ye canye parvatiyas ca sraddhe narhanti kecana. For this verse see Vangavasi Press (Calcutta) and Venkatesvara Press (Bombay) eds. 32.66-v. 1. 'ketanam' (for 'kecana') in both. This verse occurs in the Bengal Manuscripts of the Srsti-khanda. 108 Padma-purana, Srsti-khanda 17. 176-178 a (=Venkatesvara Press (Bombay) ed. 17. 176-178 a=Vanga.ed. 17. 169-171 a. V.1. in both these editions-"ca" for "va", "kutsite'kutsite" for "kutsite kutsite", and "syat te" for "tubhyam"). 3

34 STUDIES IN THE UPAPURANAS fickle-minded, shall she live in fools as well as in the Mlecchas and the Parvatiyas, and the vile and the despised. In blockheads and the proud, in the cursed and the wicked-in a person of this type [shall be] your residence caused by [my] curse.' " Side by side with popular Saivism, which was being guided by the Tantras, 104 including those of the Vamacaras, from a fairly early period, there grew up into prominence the Tantric worship of a mother goddess, variously named as Kamakhya, Kamaksi, Kali and Kalika, who was conceived of as the wife and Sakti of Siva. Being connected with popular Saivism for a long time and practised often in the Vamacara-marga, this Tantric Saktism attained great popularity in Kamarupa and became an additional, but no less potent, factor for social indiscipline and moral degradation of the people. We shall see hereinafter that Kalika-purana 84. 1-28 a narrate a story of the spread of Vamacara Tantricism in Kamarupa as a result of Vasistha's curse to Ugratara, Siva's Ganas and Siva himself for turning him out of that place and that Kalika-purana 84.28 b-86.39 give the description of a devastating flood which was caused by Brahma's waterbodied son (Lauhitya) procreated on Santanu's wife Amogha and brought down by Jamadagnya and which flowed into the southern sea after obscuring all the sacred pools and holy places in Kamarupa. Although the entire portion of the Kalika-purana describing the said flood must be taken as spurious for reasons to be stated later in their proper place, there is little scope for doubt that the rule of the Mleccha dynasty for more than three hundred years, the ruin brought by the devastating flood of the Lauhitya on the followers of the varnasrama-dharma who lived mostly in the lower plains, and the spread of Tantricism among the Saivas as well as the Shaktas, seriously affected the social, religious and moral 104 In the Gauhati Copper-plate Inscription of Indrapala-varma-deva (circa 1038 A.D.) 'Tantra' has been mentioned among the subjects studied by him. The study of "Tantra' by Indrapala-varma-deva himself shows how popular it was in those days, and its recognition by the king testifies to an earlier beginning of its study in Kamarupa.

THE SAKTA UPAPURANAS 35 life of the people of Kamarupa. Consequently, the present Kalika-purana was composed, most probably by the Maithila Brahmins of Kamarupa, to popularise the Brahmanical ideas of life and society by re-vitalising the dying Vaisnava faith by associating it with Saktism. Thus, in the present Kalikap. goddess Kali or Kalika has been described as Visnu's Yoga-nidra or Maya, Visnu has been presented as superior to Siva, and people have been advised to be Vaisnavas before they set themselves to Devi-worship. We shall now examine the individual Upapuranas.