Pramāṇa · Search · About · API

Indian Palaeography from about B.C. 350 to about A.D. 1300 c7.s15: § 20. — The precursors of the southern alphabets

Indian Palaeography from about B.C. 350 to about A.D. 1300 c7.s15

Language
English

Contents

§ 20. — The precursors of the southern alphabets

A. — The alphabet of the Kṣatrapas of Mālva and Gujarāt; Plate III

c7.s15.t1.1

While the inscriptions of Northern India thus show in the first and second centuries A. D. the beginning of the development of a new local variety of the Brāhmī, we find in the documents from Western and Central India, as well as from the Dekhan, the first steps leading up to the later southern alphabets. The inscriptions and coins of the Kṣatrapa dynasty of Mālva and Gujarāt, descended from Caṣṭana or Tiastanes, illustrate the western writing, and col. VI, taken from the Girnār Praśasti of the reign of Rudradāman (about A. D. 160)⁵ gives a specimen of it. This script agrees with the later southern alphabets (§ 27, below) in the following characteristic points: — (1) in the curves at the ends of *A* and *Ā* (1, 2), *ka* (7), *ña* (15), *ra* (32), and of medial *u* and *ū* (not in the plate); (2) in the round-backed *ḍa* (18); (3) in the *ba* (28), notched on the left; (4) in the *la* (33) with the vertical bent to the left; and (5) in the medial *ṛ* (see *sṛ*, 37), which is difficult to distinguish from *ra*. Its other letters, for instance, *śa* (35) and the tripartite subscribed *ya* of *lya* (42), partly agree with those of the inscriptions of Soḍāsa, and partly, — for instance, *kha* (8), *na* (25) with the bent base-line, *pa* (26) with the notch in the left vertical, *ya* (31) with the curve on the left, and the frequently rounded *va* (34), — with the types of the Kuṣana period. Peculiar is its *ṭa* (16). Its cursive medial *ū*, which is used only in *nū* (25) and in *rū* (compare pl. VII, 33, III), and the *au* in *yau* (31), besides which the older form of pl. II, 28, XVIII, is used, appear here for the first time.

c7.s15.t1.2

⁵ Bhaṇḍārkar, Early Hist. of the Dekkan², 26 ff., C.CMI, 3—5; Bhagvānkāl, J.RAS. 1890, 642; Bühler, Die ind. Inschr. u. das. Alter d. ind. Kunstpoesie, 46 ff.

c7.s15.t1.3

The letters on the somewhat older coins¹ of Rudradāman's grandfather Caṣṭana and of his father Jayadāman, which probably were struck in Ujjain, exhibit no material differences. Among the later Kṣatrapa inscriptions,² that from Junāgaḍh, incised during the reign of Rudradāman's son Rudrasiṁha, fully agrees with the Girnār Praśasti. The Gunda inscription of the same prince from the year 103 (or, according to the usual assumption, from A. D. 180), and the Jasdan inscription of Rudrasiṁha's son Rudrasena from the year 127 (?) or A. D. 204-205, show a few more advanced characters. Both these documents offer the bipartite subscribed *ya*; and the second has several times the northern *ma* of the Gupta period (pl. IV, 31, I ff), as well as the *e* standing above the line (compare, for instance, *ne*, pl. VII, 27, V). The same *ma*, or a similar sign with a straight base-stroke, appears also frequently on the coins of the later Kṣatrapas.³ Its occurrence probably indicates a northern influence, perhaps that a northern alphabet was used at the same time; compare § 28 below, A.

c7.s15.t1.4

¹ C.CMI. pl. 1; J.RAS. 1890, pl. at p. 638; B.ASRWI. 2, pl. 7.

c7.s15.t1.5

² Compare facsimiles in B.ASRWI. 2, pl. 20; J.BBRAS. 8, 234; Sanskr. and Prākr. Inscrṣ. Bhaunagar, pl. 17—19 (unreliable).

c7.s15.t1.6

³ See the plates cited in note 1 above.

B. — The alphabets of the cave-inscriptions of the western Dekhan and the Koṅkaṇ; Plate III

c7.s15.t2.1

[43] The writing of the western Dekhan and the Koṅkaṇ in the caves of Nāsik, Junnar, Kārle, Kaṇheri, Kuḍā, &c., shows three varieties, an "archaistic" or retrograde type, a more advanced one with mostly faint traces of southern peculiarities, and an ornamental one. The first two appear in the oldest dated inscriptions of the Śaka Uṣavadāta or Usabhadāta (Ṛṣabhadatta),⁴ the son-in-law of the Kṣaharāta king and Kṣatrapa Nahapāna from the years 41 to 45 of, according to the usual assumption, the Śaka era,⁵ or from A. D. 118 to 122. The Kārle inscription No. 19 (col. VII) offers the "archaistic" or retrograde type, among the letters of which *gha* (10), *ja* (13), *da* (23), *bha* (29), *ya* (31), *la* (33), *sa* (37) and *ha* (38) come close to the forms in the older alphabets of pl. II, especially to those of the oldest Andhra inscriptions in cols. XXIII, XXIV. The same variety is found in some other, partly older, inscriptions of the same caves,⁶ and must be regarded as a direct development from the ancient Andhra type. It shows only very faint traces of the southern peculiarities enumerated above. The curves at the ends of the verticals are only rudimentary. The vertical of *la* is curved, but to the right. The triangular *dha* (24), which appears here for the first time, is found also in other alphabets of this plate (see col. XI ff.); the abnormal *kha* (8) is confined to Kārle No. 19.

c7.s15.t2.2

⁴ Usabhadata only in Kārle No. 19, B.ASRWI. 4, pl. 51.

c7.s15.t2.3

⁵ Thus Bhaṇḍarkar, Early Hist. of the Dekkan², 25, and Bhagvānlal, J.RAS, 1890, 642; see also Bühler, Die ind. Inschr. u. das. Alter der ind. Kunstpoesie, 57 f.; while Cunningham, CMI. 3 f., refers Nahapāna's dates to the Mālava era of B. C. 57-36, and Oldenberg, IA. 10, 227, places Nahapāna between A. D. 55 and 100.

c7.s15.t2.4

⁶ Kārle, Nos. 1—14, B.ASRWI. 4, pl. 47, 48; Nāsik, No. 4, op. cit., pl. 51.

c7.s15.t2.5

Against this rather clumsy alphabet, we find in Uṣavadāta's inscriptions from Nāsik (cols. VIII, IX) very neatly made letters, the ductus of which resembles that of Soḍāsa's inscriptions (col. I) and of the Girnār Praśasti (col. VI). They show no trace of archaic forms, and the traces of the southern peculiarities are faint or entirely wanting. Only the southern *da* (18) is distinct and constant. Noteworthy are *śa* (35, 42, VIII), which agrees with that of col. VI, the final *m* in *ddham* (41, VIII), and the tripartite subscribed *ya* in *bhyaḥ* (41, IX).

c7.s15.t2.6

Very similar to this script is that of the Nāsik inscriptions (No. 11, *a*, *b*, = col. X) of the Andhra king Gotamiputa Sātakaṇi, who destroyed the Kṣaharāta dynasty, — possibly just Nahapāna and Uṣavadāta, — and of his son Siri-Puḷumāyi, Puḷumāi or Pulimāvi (Nāsik No. 14, = col. XI), who is mentioned by Ptolemy as Siri-Polemaios or Polemios.¹ The only material difference occurs in the triangular *dha* (24, XI; compare col. VII), which however is by no means constant. Nearly of the same type are the alphabets shewn in col. XII. from the Nāsik inscription of the somewhat later Andhra king Gotamiputa Siriyaña Sātakaṇi, in col. XIII. from the undated inscription Nāsik No. 20, and in col. XIV. from Nāsik No. 12, incised during the reign of the Ābhīra king Īśvarasena.² In col. XII, however, we have a peculiar form of *ta* (21) developed from a looped form, a looped *na* (25) somewhat differing from the northern form in col. IV, a *ra* (32) with a stronger curve, and a *la* (33) with the vertical bent towards the left; further, in col. XIII. a looped *ta* (21), and in col. XIV. a *ta* (21) and a *na* (25) derived from looped forms, a *ya* (31) with a curve on the left, a *la* (33) bent towards the left, a cursive subscribed *ña* in *jñaḥ* (40), and a peculiar, *r*-like, medial *u* in *du* (23), which reappears in later southern inscriptions; compare, for instance, *bhu*, pl. VII, 30, XII, and the *ū* in *tū*, pl. III, 21, XVII, XIX.

c7.s15.t2.7

¹ See the works quoted in note 5 on page 42 above.

c7.s15.t2.8

² According to BHAGVANLĀL'S estimate, J.RAS. 1894, 657, "somewhat later than Nahapāna."

c7.s15.t2.9

Cols. XV, XVI, give two somewhat differing specimens of the ornamental variety of this period according to the undated inscriptions of Kuḍā (Nos. 1—6, 11, 20) and of Junnar (No. 3). Both agree in the ornamental treatment of medial *i* and *ī*. But the Kuḍā inscriptions extend it to the curves at the ends of all verticals, and show notches in the left [44] strokes of *pa* (26) and *ba* (28; compare col. VI). In col. XVI. there are two other noteworthy signs, the bipartite subscript *ya* in *yya* (40), and the *śa* with the horizontal bar in *śri* (41; compare 35, III—V). Ornamental forms, resembling those of cols. XV, XVI, are found also in the approximately datable inscriptions of Puḷumāyi in Kārle Nos. 20, 22, and of the minister of the queen of his successor Vāsiṭhīputa Sātakaṇi in Kaṇheri No. 11. The first two of these documents show a looped *ta* and a *ṇa* like that of col. XVII; the third exhibits the neat characters of Western Kṣatrapa inscriptions. It is, therefore, certain that during the 2nd century A. D. all these three varieties were used promiscuously in the western Dekhan and the Koṅkaṇ,³ and the inscriptions from the Amarāvatī Stūpa⁴ prove that they occurred also on the eastern coast of India. The contemporaneous employment of more advanced types and of more archaic ones with an admixture of more modern signs will have to be explained in this, as in other cases, by a desire to select archaic and monumental forms for epigraphic purposes and a failure to completely carry out this intention.

c7.s15.t2.10

³ Compare facsimiles in B.ASRWI. Vol. 4, pl. 45, Kuḍā Nos. 12—18; pl. 46, Kuḍā Nos. 22—26; Mahād Nos. 1—4; Kol Nos. 3, 5; pl. 47, Beḍsā Nos. 1—3; pl. 48, Kārle Nos. 15—18; Śailarvāḍī No. 19; Junnar Nos. 1, 2; pl. 49—51, Junnar Nos. 4—34; pl. 52, Nāsik No. 6 a; pl. 54, Junnar No. 32; Kārle No. 20; pl. 55, Nāsik Nos. 17—19, 21—24: and Vol. 5, pl. 51, Kaṇheri Nos. 2—5, 10, 12—14.

c7.s15.t2.11

⁴ B.ASRSI. 1, pl 53, 57; pl. 58, Nos. 28—34, 37; pl. 59, Nos. 39, 43; pl. 60, Nos. 44, 45, 47—50; pl. 61, Nos. 51—53, 55, 56; and the autotypes of the Andhra coins, C.CAI. pl. 12, and J.BBRAS. 13, pl. 3.

C. — The alphabet of the Jaggayyapeta inscriptions; Plate III

c7.s15.t3.1

In the Kistna districts of the eastern coast, a still more ornamental alphabet, found in the Jaggayyapeṭa inscriptions from the time of the Ikṣvāku king Sirivīra Purisadatta (cols. XVII, XVIII), as well as in some Amarāvatī inscriptions,⁵ was developed out of the ornamental variety just discussed, probably somewhat later, in the 3rd century A. D. One of its most prominent characteristics is the very considerable elongation of the verticals of *A*, *Ā*, *ka*, *ña*, *ra* and *la*, as well as of the medial *i*, *ī* and *u*. To a later time point the cursive forms of *tha* and *ha*, which latter agrees with the northern Gupta form (pl. IV, 39, I, VI), and the medial *e* of *me* (30), which, with its downward curve, agrees with the *e* of the later southern inscriptions (compare 30, XIX, XX, and pl. VII, 35, XII), and the medial *ū* in *tū* (21; compare col. XIX, and pl. VII, 30, XX). The medial *ū* of *tū* (40), in which the stroke expressing the length of the vowel has been attached to the head of the consonant, is entirely abnormal.

c7.s15.t3.2

⁵ B.ASRSI. 1, pl. 58, Nos. 35, 36; pl. 59, Nos. 38, 40—42; pl. 60, No. 46; pl. 61, No. 54; pl. 62.

D. — The alphabet of the Pallava Prākrit land-grants; Plate III

c7.s15.t4.1

The highly cursive writing of the Prākrit land-grants of the Pallava kings Vijayabuddhavarman and Śivaskandavarman from Kāñcī (Conieveram) in the Tamil districts,¹ shows in its *ductus* a certain relationship to the Jaggayyapeta inscriptions. But it is not doubtful that these documents are much later, though it is for the present impossible to fix their dates exactly. The use of Prākrit for official purposes perhaps indicates that they are not later than the first half of the 4th century A. D. The broad *E* (5, XX) with the rudimentary vertical to the right (compare pl. VII, 6, XI ff.), the *ḍa* with a tail in *ṇḍam* (40, XX ; compare pl. VII, 19, IV f.), the subscribed *tha* open on the right in *ttha* (41, XIX ; compare pl. VII, 45, XX), and the constantly looped *o* in *lo* (33, XX ; compare pl. VII, 34, III f., XIII, XVII) point to the later period.

c7.s15.t4.2

¹ Compare facsimiles in IA. 9, 100; EI. 1, 1 ff.

c7.s15.t4.3

IV. THE NORTHERN ALPHABETS FROM ABOUT A. D. 350.²

c7.s15.t4.4

² Preparation of Plates IV, V, and VI : —

Linked texts (1)