1 A rather frivolous example of the process that is the subject of this paper

2 I am deeply grateful to mr.PeterBee not only for arranging the facilities at the National Library, but also for assisting me throughout the photography sessions. I am most grateful to mr.E.H.S.Simmonds, Reader in Tai Languages and Literatures in the University of London and the mrs. Judith M.jacob, Lecturer in Cambodian at the School of Oriental and African Studies for all their help.

3 I found that it is catalogued XVII 37(i).

4 The further work in this field of Mr.Singaravelu is awaited with great interest. From the South Indian standpoint, it is of interest that chants based, as will be shown, on Tevaáram, should be uttered by brahman priests. Presumably their connexion with Indian is of greater weight than their language. The chanting of Teváram, Popularized recently in Tamilnad, is the Prerpgative of the Otuvár, non–brahman reciters. The Ta. verb otu &recite' can be used in Vedic contexts too. There seems no doubt that this is a Drav. root; see Burrow and Emeneau, Dravidian etymological dictionary,p.77 886.Against this may be set the tradition thaone of the Teváram hymnists, Naņacampantar, was a brahman from Sirkáli; see Cekkilãir, Periyapuránam, Tiruňáa; esp.vv 1 and 3, and Avvai Turaicami Pillai, Caivailakkiya varaláru p.35. It is Possoble that brahmanical association with this material, long since forgotten in Tamilnad, is Preserved in this way in Thailand.

5 u.infra, p.303.

6 u.infra, p.304.

7 Facing p.55. It is of a corrupt Skt.text.As is usual in these Grantha MSS, Thati symbols for beginning and ending are included.

8 Facing pp.56,7.

9 Quaritch Wales’s Plate ii in fact depicts the first spread of MS XVII37(i), minus its title. The same MS includes the corrupt versions of four Tamil Teváram hymns that will herein be discussed; u.infra, p,303. He does not mention that there were two forms of the script but mentions that Dr.L.D. Barnett was of the opinion that‘ they are Pandyan, and may be ascribed to a period not later than the middle of the thirteenth century A.D’ op.cit., 56. While this may be true of the epigraphy of many of the characters the MSS copies are comparatively molden, as the Thai date affixed to one of them shows; u.infra,p.303.

10 op.cit., 56. He states that the originals were in Siamese characters

11 u.infra,p.303.

12 The Principal of which are:C– voiceless;–CC–voiceless tense;–nasal+.C–voiced; –C–poiced lax.

13 Save in centain dialects that feature aspiration, such as Ceylon Tamil.

14 Moreover, these Grantha characters are used to represent Ta.uallina r, a Point of some interest.

15 Mr.Singaravelu also kindly allowed me the privilege of hearing a recording he had made of a Thai brahman‘s; recitation read, he informed me, from a Thai–script MS. The hymns therein had been identifield by Fr. F. X. Thani Nayagam with three Teváram hymns and with Manikka–vácakar‘s Tiruvempávai, he told me. As will be shown, the three Tev. humns are contained in MS XVII 37(i) also, but Tiruvempávai is not.

16 See plates I and II.

17 Line 41.

18 I am indebted to Professor J.C Wright for this suggestion.

19 Lines 7,8.

20 of.this with Thai 3 visanjaini, While: or: have been in use in South India at least since the early colas, see SII, III, p.264, some Gupta scripts had two dashes.

21 For the meaning of a, v. infra, p.292.

22 See plate I and Plate III,1.4.

23 Abbreviated SIG, this will be used herein to denote those forms largely of palm–leaf origin that culminate in printed Grantha used in Tamilnad nowadays in rather limited contexts. A few of its characters are used in Tamil.

24 See T. N. Cuppiramaniyan, Pantai ltamil eluttukkal, p.44, tab.3, row xv, and p.49.SIG,á–in common with Tamil and Cambodian, joins the lengthener to the character and is based on plam–leaf forms.

25 Sivaramamurti, C., Indian epigraphy and South Indian scripts, Madras, 1952, 46.Google Scholar

26 I have not observed it elsewhere in the DG material examined.

27 Sivaramurti, op.cit., does not notice it at all, but Cuppiramaniyan, op. cit., has one instance; see tab.3, col.xvi. He gives the source as a c.p.g. of Varatunkaráma Pántiyan.

28 op.cit., tab.3, col. xvi.

29 Cuppiramaniyan, op. cit., tab.3,col.XI.

30 Remarks on the Indo-Chinese alphabets; JRAS, NS, III, 1868, art.ii, tab. and pp.65–80. The alphabet of the Cambodian brahmans is the 21st in his list; he also lists one from brahmans from Siam but this does not appear in the tab. I am greatly indebted to Mrs.Judith Jacob for this reference.

31 The term is Dani’s. See A.H.Dani, Indian palaeography, Oxford,1963,278.

32 u.infra, P.321, notes (29) and (30).

33 u.infra,p.315. DG has only the Grantha form, to which Bastian's script and Cambodian seem to adhere.

34 Sivaramamurti, op.cit., p.97,fig.37.

35 Tev., Tiru. Vii, Pat.I,u.infra.

36 U.infra,p.313,text(b),1.3.

37 See plate II.

38 See Sivaramamurti, op. cit., p.103, fig.40.

39 u. supra, p.286. The Tamil phonemes are set out by Dr.K.Zvelebil in Archiv Orientáni; XXXI,2, 1963,225 ff., &The vowels of colloquial Tamil ' (see esp.p.226, sec.1.1). Use of OG gh medially could be in part explained by Tamil realization of intervocalic lax –g–

40 CuppiramaCuppiramaṇiyaṉ, op.cit., tab. 3, row xv, and p.49; Sivaramamurti, op. cit., 104–

41 On close examination I am certain that what appears as a short vertical stroke beneath the character is an error or blemish. But resemblance to the stroke beneath Telugu aspirates such as Ά or ф was compelling. Cf.also SIG conjunct ddh,æ

42 Bastian, art.cit., tab.; Cuppiramaṇiyaṉ op. cit., rows VIII– XI.

43 The DG ṇ eschews this, and so looks older.

44 JRAS, NS, III, 1868, art, ii. This is one of several points of contact between the DG and Bastian ’s script.

45 Sivaramamurti, op.cit., p.117, fig.47.

46 ibid.

47 Bastian has for this, which bears a striking resem blance to the OG character for ȡjust considered. Again, there is a link here between Cambodian brahman and DG, as will be seen; u.infra, p.298.

48 Sivaramamurti, op.cit., p.120, fig.49; Dani, IP,p.284 and plate XVIIIa, row 9.

49 See Tha Myat, The history of Mon–Burmese alphabet, Rangoon, n.d.,p.28,tab.

50 Sivaramamurti, op. cit., p. 129, fig.53.

51 ibid., p.131, fig.54.

52 Sivaramamurti, op. cit., p. 133, fig. 55.

53 ibid., p.138, fig.57; there are also good parallerls in c.p.g: e.g. the Velvikkuti and Velankuti grants of Pantiya Jatilavarmaṉ (eighth century). See Cuppiramaṇiyaṉ op.cit., p. 46 and p.45, fig.3, rows VIII, IX.

54 Though not the ‘closed’ from ஸ also in use, to which Mal.സ may be compared.

55 Cuppiramaṇiyaṉ op.cit.,p.45, tap.3, row XIV, and p.49.

56 See Bastian, art.cit., tab. He lists the former as no.9 and the latter as no.17,see p.80.

57 u.infra, p.313.

58 e.g.ll.6and 10,ś and l.35,ş It appears separated from k– in an apparent rendering of kṣa as kaṣa at the end of the alliterative litany. See plate I.

59 cf.ஈ

60 Sivaramamurti, op. cit., p.145, fig.60.

61 See plates I and III, and infra, pp.313,314.

62 Where it stands for દદ see M.Haas, The Thai system of writing, Washington, D.C.1956, 15.

63 Sivaramamurti, op. cit., p.72, fig.27.

64 ll.35 and 40 respectively.

65 ‘Verse’ in the context of the Thai chants is used in quotation marks as there is no evide that the units are metrical; they correspond approximately to the verses of the Tamil orig.

66 The date at the end of XVII 37(i) = A.D.1875, u.infra,p.303.

67 The phon man and khomūt.

68 u.infra, p.301. It coexists with a supported i– glosed as such.

69 Cuppiramaṇiyaṉ op.cit., tab.3.

70 fol.252, ll.6 and 7: īsva: ramproktum and īsvaramproktum.

71 op.cit., 66.

72 fol. 272,1.3.

73 u.infra,p.301, and plate II. It may serve to show both forms.

74 cf.110. The process is seen partially in Devanāgari; cf.0–আ and au– আ

75 u.infra, p.302.

76 fol.261,1.8.

77 fol.262,1.1.

78 fol. 264, 1.3. The same orthograph is seen in the DG version of Tiruvěmpāvai, representing the –o– in the refrain:–melorěmpāvāy, as will be seen; v, infra, p.308. The ambiguity above in translating the unnecessarily–subscribed final nasal is realized.

79 fol. 241,title–line.

80 I was rejected because of its strict implication of utterance that could not be substantiated from a written context.

81 u.infra, p.312,text,1.3.

82 A certainty; the word is Viṣ ṇ fol.244, 1.5.

83 fol. 243,1.5.

84 fol.245, 1.7. These three are in the Skt. portion.

85 Transliterated –ie.

86 Tev., Tiru. I, pat.I.

87 fol.264, 1.2.

88 Tev., Tiru.I, pat.1.2.1.3.

89 op.cit.,tab.

90 See esp.fol. 266, the whole of‘v.’5.

91 e.g. last group of repeating-unit of chant starting at fol. 259;u. infra, transcript, p. 314.

92 fols. 243, 1.4, 253, 1.1. In the former it is at the end of a group for Skt. enclitic -ca ‘and’.

93 It is seen here over gha, and precedes initial a-, a common collocation.

94 See fols. 262, 1.2, and 264, 1.3. u. supra, P. 294.

95 e.g. fol. 247, 1.7, and fols. 259 ff.

96 See plate II.

97 um, im, am, a:. The reason for showing anusvāra with all three sort vowels, especially as it is written in lino with them SIG fashion,is obscure. Cambodian customarily shows it with the three vowels u, a and ā, the anusvāra of course being superseribed.

98 art. cit., tab.

99 u. supra, p. 287.

100 Sivaramamurti, op. cit., p. 103, fig. 40.

101 JRAS, NS, III, 1868, 69.

102 In common with the prathamavarga ṭ (dentally) and p.

103 u. supra, p. 288.

104 e.g. MS XVII 37(v), fol. 128, 1.5: vilaiyade for Ta. vilaiyati; fol. 131, 1.6: bādivaṉd Aṉkum… for Ta. pātivantorkkum, and interestingly,fol. 132, 1. 3: mandaṉ for Ta⃜mutěṉrěllo⃜

105 JRAS, NS, III, 1868, art. ii, tab.

106 The similarity of this to OG and SIG ḷ is interesting.

107 u. supra, pp. 288, 259. Here, as elsewhere, the inherent-a is omitted for convenience.

108 Though the loop is at the right-hand and in SIG: .

109 of. for instance prachum silāčhāru‘ k phāk thĭ, III, 1965, 6, wherein ṭesam transliterates an inscriptional) tesam.

110 Wherein a circle denotes th, u. supra, p. 288.

111 JRAS, NS, III, 1868, art. ii, tab.

112 e.g. Dakkhina-ThŨpa inscription of Dathopatissa, seventh century, in Ceylon.

113 Sivaramamurti, op. eit., p. 114, fig. 46.

114 The Myat, op. eit., 26.

115 The secondary forms are perhaps written for reference, being somewhat dissimilar to thir primaries.

116 JRAS, NS, III, 1868, art. ii, tab.

117 And, obscurely, Telugu/Kannada.

118 It is interesting to note that, for Tamil characters ண, ற, ன, the wheel has turned full circle. Initial a– is அ, á– ஆ, and ná, ๟á, and ṉá are respectively ண, ௫, ை, All other are written with medial –á ச. The special grapheme for lengthening initial a– is used to denote –á medially rather than the other way a bout ! In SIG the graphemes are distinct; ை is a–, ௲ is d– and all consonants take medial –á ா ṇá is ணா and iá is தா.

119 See Sivaramaurti, op. cit., figs. 20, 21, 26, pp. 57, 59, and 70.

120 Sivaramamurti, op. cit., 61.

121 It Will be recalled that OG in common with SIG has medial –e Prefixed to initial e– to give ai., a structure reflected in the medial SIG ൈ and Thai แ.

122 There are one or two exceptions in DG MSS as noted in the transcripts. This may be due to Thai or Cambodian influence.

123 e.g. XVII 37(V), fol. 131, 1. 6.

124 e.g. ibid., 1. 7 for Ta. u– fol. 133, 1. 8, for i– and e–

125 e.g. banadi(ch)aiyyaguráay for Ta. poṉaticaipakaráy, fol. 131, 1. 3: nariṉ(ch)án A for Ta.ṉiruǹcīraṉ, fol. 132, 1. 1.

126 v. supra, p. 287.

127 This in fact appears at the foot of fol. 270, with khomūt after it, so it may be a closing rubric, not a title. But PitpratūṢiválaiy in DG appears as the title of XVII 37(v), fols. 179-85, and Quaritch Wales has a similar title in his MSS list, op. cit. 63.

128 I am much indebted to Mr. E. H. s. Simmonds for reading this for me.

129 It was these three hymns that figured on Mr. singaravelu's recording and were identified by Fr. Thani Nayagam.

130 v. infra, p. 314.

131 v. infra, p. 310.

132 One is called Vástupuruṡ

133 The term chant is used to describe these texts.

134 Tev., Tiru. II, 47, v. 11, 1. 4.

135 Ta. moṉai and ětukai.

136 Strengthening Quatich Wales’s argument that the chants were not understood: op.cit.,55.u.infra,p.306,n.140.

137 Save v. ll, common to l.4 of all verses: kāṇāte potiyo pŨmpāvāy. There is a sub–refrain that shifts between ll.2 and 3: Kāpālīccuram amarntāṉ In the Bangkok version these are …ghānātbholiyobhuāvāyya and gabhālai camadāṉ respectively, with minor variations; see transcript.

138 Text source: Tiruṅ āṉacampantar cuvāmikal tevāram, Kasi Mutt ed., 1950,661-3.

139 PeetpratŨbra:tamnak k⁛⁛wkrailād ȘOpening the doors of the crystal court of Kaiāśaș I am much indebted to Mr.Peter Bee for this reading.

140 Again strengthening Quartich Wales's argument;u.supra, p.304, n.136. He states further, op. cit., 84:‘This use of the three lauguages is interesting: the corrupt mantras, by reason of their being unintelligible together with their venerable antiquity, are of sociological value in surrounding the ceremony with an air of mystery ’.

141 fol. 260, 1.4.

142 Tamil text sources: Tiruāṉacampanta cuvāmikal., Tevāra ttiruppatikankal, mutal tirumuṟai, Tarumapuram, 1953. See text (b),p.313.

143 Nāṉacampantar's hymns being in Tiru. I-III, Appar's in IV–VI, and Cuntarar's in VII. But the preceding chant in the MS is of a stray one from Tiru.II.

144 Text pub.at Srivainkuṇtam,1958.

145 The DG has: oᭅkāṉyammahatdarŨprattitrattum.Ȧ after a title Brarājabidijāhā(na). Moreover, both MSS seem to incorporate herein a very corrupt reading of 1.4 of Appar's Tev., Tiru. Iv, pat. 1. v. 5: tutakki mutakki yita ārttārpiṉalār atikaikkětilavirattānattuṟai ammāṉe.of fol.242, 1.6, and DG, fol.165, 1.3: gandidamudakkimudakgayittavā gunnati(gri) gassaviṉaturo harihavaro.

146 op. eit., 63.

147 But this suggests that the eaption was in This soript, recalling noted for OG MS XVII 37(!) Where it appears at the end of fol. 270.

148 That results in the omission of lines 5–7½ of v.5 altogether.

149 1.5: maṉṉavare and kanava.

150 u. infra, p. 317–18, fol. 130, 1. 6-fol 131, 1. 8. A small portion of v. 4, 1.2 appears instead of the ‘jump line’ of v. 6.

151 The is seen again in the sequel, pŨjāmuraiypotakalām wherein b Atti stands for Ta. poṟṟi at the beginning of eaach of each line of Tirucěm., v. 20.

152 u. supra, p. 304.

153 Mr. Peter Bee kindly suggested that this is probably representuing the Thai sŨat ‘to recite a mantra right through’

154 XVII 37(V), fol. 143, II. 1–4.

155 ibid., 143, 1. 4, fol. 144, II. 1–3.

156 fol. 144, 1. 4 includes:…naṉra ttulaṉnāgā naratabhota…, suggestive of ta. nāṟṟattuḷāymuṭ nārāyaṉ…poṟṟa… Tiruppāvai, v. 10, 1.3; fol. 144, 1.6 has nautagobhan… which recalls Ta. nantagopaṟ, ibid., v. 16. 1.1.

157 Quaritch Wales, op. cit., 247 ff.

158 The popular name is Lo Jin Jā.

159 cf. Aymonier, , Le Cambodge, III, 548, and Quaritch Wales, op. cit., 59.Google Scholar