Elements of South-Indian Palæography, from the Fourth to the Seventeenth Century A.D. c5.s4: § 4. The South-Indian Nāgarī Alphabets
Elements of South-Indian Palæography, from the Fourth to the Seventeenth Century A.D. c5.s4
- Language
- English
§ 4. The South-Indian Nāgarī Alphabets
c5.s4.1
The South-Indian form of the Nāgarī character as current in modern times, usually goes by the name of Nandināgarī, a name it is quite as difficult to account for, as Devanāgarī². The Nandināgarī is directly derived from the N. Indian Devanāgarī of about the eleventh century, but it is from the type that prevailed at Benares and in the West, and not from the Gauḍī or Bengālī. This last is chiefly distinguished from the other types by the way of marking the secondary **e** and **o**, which is done by a perpendicular stroke before the consonant in the case of **e**, and by a similar stroke before and another after the consonant in the case of **o**, and this is, very nearly, the actual Bengālī system. The other type marks these vowels in the same way as is done by the ordinary Nāgarī alphabet. Thus the S. Indian Nandināgarī is derived from the Siddhamātraka character, used, according to Albīrūnī (1031 A.D.) in Benares, the Madhyadeça and Cashmere¹. It now differs greatly from that type or from the N. Indian Devanāgarī, and is remarkably illegible; but this deterioration took place very slowly, and is unquestionable owing to the practice of writing on palm-leaves. The Nāgarī inscriptions in S. India are all, with one exception, subsequent to the tenth century; this exception is at Seven Pagodas in the temple of Atiraṇacaṇḍeçvara near Šaluvaṅkuppam, and is in nearly the same character as a dated inscription of the seventh century found near Nāgpur and published in the *Bombay Journal*². As this inscription is given in two different characters, this must have been done for the benefit of pilgrims from the North. It has already been published³.
c5.s4.2
² The word ‘Nāga(rī)’ first occurs, it seems, as the name of an alphabet in the Lalitavistara, a life of Buddha that is in its original form perhaps two thousand years old; but as it exists in Sanskrit and Tibetan, it would be very unsafe to put it at an earlier date than about the seventh century A. D. The Tibetan version (of which Profr. Foucaux has published a most excellent edition and translation) was made in the ninth century by three natives of India named Jinamitra, Dānaçila and Munivarmā with the assistance of a Tibetan Lotsava named Bande Ye-šes-sdes; this fact is stated in the Tibetan index to the great collection called Bkah-hgyur (Kandjur) in the description of the work in question (Rgya-tcher-rol-pa *i. e.* Lalitavistara), and is to be found on p. 16 (No. 95) of this index as reprinted at St. Petersburg. Nāga(rī) occurs as the name of an alphabet in ch. x. (v. p. 113 of vol. i. of Profr. Foucaux’s edition) which describes how the young prince, afterwards known as Buddha, was taken to a school and completely posed the pedagogue. Sixty-four alphabets are mentioned some of which are, no doubt, mythical, but others are real (*e. g.* Drāviḍa, Aṅga and Baṅga), though it is against all the evidence of the inscriptions that they existed as distinct alphabets before the ninth or tenth century A. D. If therefore the framework of the Lalitavistara be old, this passage is certainly an interpolation, though very valuable evidence regarding the ninth century A. D. But this Tibetan version by no means bears out the meaning usually assigned to the word Devanāgarī—“nāgarī of the Gods or Brahman”, nāgarī being usually referred to nagara and being supposed to mean ‘writing used in cities’. The Tibetan text has here the ordinary name (in that language) of the Nāgarī character—“klu-’i yi-ge” (as a translation of the Sanskrit ‘nāga-lipi’) and this is also literally “writing of the nāgas”. (*Cfr.* Jaeschke’s W. B. p. 7.) It is evident, therefore, what the natives of India understood nāgalipi or nāgarī to mean in the ninth century A. D., and it only remains to be seen if this derivation is possible. I think this question must be answered in the affirmative, as not only Prakṛit but also Sanskrit words exist which are formed in the same way. (Bopp’s Comp. Grammar, ed. Bréal, § 940.) There is yet another possible explanation of ‘nāgarī’—that it means the writing of the Nāgara or Gujarat Brahmans. (*Cfr.* ‘nāgara’ in Molesworth’s Mahr. Dictionary.) Albīrūnī (Reinaud, “Mémoire”, p. 298) mentions the ‘nāgara’ character as used in Malva, and the ‘arda-nāgarī’ (*i. e.* ardhanāgarī) as used in Scinde.
c5.s4.3
¹ He (see Reinaud's "Mémoire", p. 297) says: "les traits sont horizontaux et ne débordent ni au-dessus ni au-dessous de la ligne; chaque lettre est surmontée d'une ligne horizontale au-dessous de laquelle elle se developpe. On compte plusieurs écritures dans l'Inde. La plus répandue est celle qui porte le nom de siddha-mātraca ou substance parfaite; elle est usitée dans le Cachemire et à Benarès, qui sont maintenant les deux principaux foyers scientifiques du pays."
c5.s4.4
² Vol. I. pp. 148 ffg.
c5.s4.5
³ "Transactions of the R. As. Society", II., pl. 15 (in Dr. Babington's Paper on Seven Pagodas). For the position of the place see the map in *Madras Journal*, xiii., and in Major Carr's reprint of papers on this subject. I had this little temple cleared of sand in 1867, and took copies of the inscriptions which I gave Major Carr.
c5.s4.6
A few inscriptions in a variety of this character have been found near Jayapura (in the Ganjam district), they are of the tenth century⁴. This character appears, there, to have been the origin of the Oriya alphabet.
c5.s4.7
⁴ The dynasty to which these belong seems to have been established by fugitives of the Veṅgi family in the 7th century. (See p. 22 *n.*) During the anarchy from 977 to 1004 A.D. these kings again rose to power for a time, and appear to have resided at Kalinganagara. The succession is: Jayavarmadeva
c5.s4.8
Inscriptions in the same character, both'Hindu and Buddhist, occur in considerable numbers in Java. Grants⁵, explanatory remarks⁶, inscriptions on rings and Buddhist
c5.s4.9
⁵ There is a stele with a long grant (?) in this character in the Museum at Batavia; it is, unfortunately, very illegible.
c5.s4.10
⁶ On statues in the Batavia Museum. These have been partly published by Friederich "Over Inscriptiën van Java en Sumatra" pl. i. Some such are visible on sculptures figured in Raffle's "Java".
c5.s4.11
Anantavarmadeva (in 985 A. D.)
c5.s4.12
Rājendravarmadeva confessions¹) of faith have all been found in this character; it is, therefore, of some importance, and I, accordingly, now give the alphabet. (Pl. xxii.)
c5.s4.13
¹ See plate xxii. From a bronze statuette in the Batavia Museum.
c5.s4.14
¹ Faria y Sousa ("Asia Portuguesa" ii. pp. 189-190) gives a list of the kings down to 1545, but with much confusion in parts. The European writers and travellers of the 16th and 17th centuries, however, give much valuable information. See especially: Couto, Decada vi., 5, 5 (f. 92 of the first edition, 1614). There is much confusion in all the accounts of the earlier times after Vīra and down to Krishṇarāja, and inscriptions fail for this period. This is owing to the first conquest of Vijayanagara (about the end of the 15th century) by the Muhammadans.
c5.s4.15
It is thus plain that the examples which occur of this character in S. India and Java must be due to emigrants from the North who saw fit to leave their own country in considerable numbers. It may be not impossible to discover the causes of this emigration, which, in later times, is probably to be attributed to the Muhammadan conquest. In earlier times, religious disputes may have been the cause.
c5.s4.16
There is little trace of development of this character.
c5.s4.17
In the Deccan, Nāgarī inscriptions begin to commonly appear during the temporary fall of the Kalyāṇa Cālukyas²), and this character appears to have been much used by the revolted feudatories³). On the revival of the original dynasty the
c5.s4.18
² For a specimen see the grant under Akālavarsha *d.* ç, 867 (=945 A. D.) in the *Indian Antiquary*, i., pp. 205 ffg.
c5.s4.19
³ The chief of these feudatories (often independent) are as follows: i. Rāshtrakūta, or Raṭṭa. General remarks on, and genealogy of this dynasty occur in *Bombay Journal*, i., p. 211 and iii., p. 98; "Indian Antiquary, i., pp. 207-9. do. iv., p. 274 and 279. For inscriptions see *As. J. v.*, (*d.* 973 A. D.) *Bombay Journal*, i., pp. 209-224 (*d.* ç. 930=1008 A. D. in Nāgarī); ii., p. 272, *n.* pp. 371-6 (*d.* ç. 675=753 A. D. also Nāg. !?); iv., p. 104 (*d.* ç 855=A. D. 933 also in Nāg.) Rāshtra seems to be merely a brahmanical perversion of the Telugu "Reḍḍi". Mr. Fleet has succeeded in restoring the complete genealogy. See *Indian Antiquary*, iv., 280. ii. Kālabhuri. (Kālacuri seems to be erroneous). *Madras Journal* ("Hindu Inscriptions" by Sir W. Elliot) vii., pp. 197, 211-221, and 224-225. *Indian Antiquary*, iv., 274; v., 45. The most important of the three kings whose names occur is Vijjaladeva the first; he conquered Tailapa ii. (of Kalyāṇapura), and during his reign (1156-1165) the revolt of Basava and the Liṅgāyats broke out which cost him eventually his throne and life. iii. Kadamba (neighbourhood of Goa). Probably an old branch of the Cālukyas. "Notes on Sanskrit Copper-plates found in the Belgaum Collectorate" by J. F. Fleet (*Bombay Journal*, ix., pp. 231-246). "Some further Inscriptions relating to the Kadamba Kings of Goa" by the same (do. pp. 262-309). See also Sir W. Elliot's article in *Madras Journal*, vii., pp. 226-9. *Indian Antiquary*, iv., 208. v., p. 15. 356. vi., pp. 22-32. iv. Sindavamça. For the genealogy as established by Mr. Fleet, see *Indian Antiquary*, v., p. 174. The new dynasties which replaced the older Calukyas in the Deccan from the 13th to the 14th centuries are: i. Devagiri Yādavas. See Lassen (I. A.—K. IV. pp. 945-6). ii. Dvārasamudra Yādavas. (do. IV. pp. 972-3). iii. The Kākateyas of Orukkallu (Ōruṅgal). From the twelfth century to 1311. I have not been able to find any inscriptions of this dynasty. iv. The Rāyas of Vijayanagara; from about 1320 to 1565. The following is the list as I have been able to correct it from several sources (see my "Vaṁçabrāhmaṇa," p. xvi.); the dates, however, are only approximate!). Saṅgama of the Yādava family and Lunar race!!
c5.s4.20
Hariyappa (1336-1350)
c5.s4.21
use of this character continued, as the sovereigns betrayed a great partiality to N. Indian literary men. There is not, apparently, the least trace of any patronage
c5.s4.23
The earlier kings of this dynasty had conquered all S. India before the end of the 14th century; but they left many of the original kings (*e. g.* the last Pāṇḍyas) undisturbed for a time; in the 16th century they had their deputies (called Nāyak) at Madura (from about 1540), Tanjore and Gingee (Šinji). In the 17th century these Nāyaks acted as independent sovereigns; the last Nāyak of Tanjore Vīrarāghava (*e. g.*) granted Negapatam to the Dutch by a grant on a silver plate, now in the Museum at Batavia. These predatory chiefs and the rabble they brought with them are the "Badagas" of whom the early Portuguese Missionaries complain so much. They did not reach the extreme South till about 1544. (Lucena, "Vida do Padre F. Xavier", p. 115 *b*; Sousa, "Oriente Conq." i., p. 231.) There is a good account of the condition and relations of these Nāyaks at the end of the 16th century by Pimenta (a 'Visitor' of the Jesuits) who was on the Coromandel Coast in 1599. Purchas (vol. ii., pp. 1744-1750) gives an abridgement of it; as also does Jarric ("Thesaurus", i. pp. 625-690). Veṅkaṭapati was then at war with the Nāyak of Madura.
c5.s4.24
¹ According to Ferishta, Rāmarāja was ruling in 1535.
c5.s4.25
² This genealogy rests on the Villapākkam grant of 1601 and similar documents.
c5.s4.26
bestowed by them or by their successors the Yādavas of Devagiri¹ on vernacular culture.
c5.s4.27
¹ The well known law-book the Mitāxarā was composed in the reign of Vikramāditya v. (1076-1127); but it is not known of what country the author was a native (*Bombay Journal*, ix., pp. 134-8). The Vidyāpati of this king was a Cashmere Brahman named Bilhaṇa. (See letter from Dr. Bühler in *Indian Antiquary*, iii., p. 89, and his edition of Bilhaṇa’s “Vikramāṅkadevacarita”).
c5.s4.28
The Muhammadan invasion of the Deccan in 1311, and the destruction of the old kingdoms, brought about the establishment of the Vijayanagara dynasty, under which not only the Sanskrit, but also the vernacular literatures were much cultivated. The early inscriptions of this dynasty are either in the Haḷa-kannaḍa or Nandināgarī character; the latest (of the 15th and 16th centuries) are almost exclusively in the last. In the South, Grantha was occasionally used. They constitute by far the largest class of S. Indian inscriptions, for the sovereigns of this dynasty at the end of the 15th and beginning of the 16th century repaired or endowed most of the large temples in the South².
c5.s4.29
² Many examples are already published. “Bengal As. Soc. Transactions,” iii., pp. 39 ffg.; also in vol. xx. Colebrooke’s “Essays”, ii., pp. 254-267. “*Indian Antiquary*”, ii., p. 371, and following vols.
c5.s4.30
The S. Indian Nandināgarī alphabet calls for very little remark, as from the earliest examples of the fourteenth century up to 1600 A. D. there is scarcely any development. It is certainly one of the most illegible characters in use in all India.
c5.s4.31
MSS. in this character are not uncommon, as it is the favorite alphabet of the Mādhva sect, which counts an immense number of adherents in S. India, especially in Mysore, the neighbourhood of Conjeveram, and Tanjore. All members of this sect are Brahmans, and all learn more or less of the books on their dogmas written by Ānandatīrtha (Madhvācārya) and his successors. The Nandināgarī is used nearly exclusively for writing on palm-leaves; for writing on paper, the ordinary Mahraṭha hand of Devanāgarī is used, and the writing is often exceedingly minute. All the inscriptions on copper-plates, and MSS. on palm-leaves that I have seen are numbered with the ordinary Telugu-Canarese numerals. This character was evidently at the beginning of the 16th century the official character of the Vijayanagara kingdom, for in it is written the name of Kṛishṇarāja on the coins which gave rise to the name “pagoda”³.
c5.s4.32
³ See the Glossary of Anglo-Indian Terms, by Col. Yule and myself, now in the press. *s. v.* ‘Pagoda’.
c5.s4.33
The modern Nāgarī (or Bālbodh) character was introduced into S. India by the Mahraṭha conquest of Tanjore in the latter part of the seventeenth century⁴, and was chiefly used in Tanjore, where it is still current among the numerous descendants of the Deccan Brahmans attracted there by the liberality of the Mahrāṭha princes.
c5.s4.34
⁴ The date of the final conquest of Tanjore by Ekoji, and the end of the Nāyak (Telugu) princes is far from certain. Orme in the last century could not be sure about the date, though he had all the Madras Government records at his disposal. Anquetil Duperron (“Recherches sur l’ Inde”, I. pp. 1-64) has gone into the question very elaborately, and puts the date at 1674-5, which appears to be as near as can be expected.
Cites (1)
- Lalitavistarā — Haribhadra Sūri, Sanskrit