Pramāṇa · Search · About · API

Indian Palaeography c15.s1: Main Divisions

Indian Palaeography c15.s1

Language
English

Main Divisions

c15.s1.1

In order to understand the interrelationship of the proto-regional scripts they are grouped here under four main heads:

c15.s1.2

A. *Scripts of North India.* These show a uniform pattern of development both in the body of the letters and in the application of the medial vowels. The changes in ornamentation and technical adaptation affect all the scripts equally.

c15.s1.3

B. *Scripts of Gujarat, Rajasthan, and Central India.* There is a medley of scripts in these areas. The earlier influences of the south fade away and the northern influences come in, but at the same time we notice some independent developments.

c15.s1.4

C. *The Deccani scripts.* This name applies to those characters which have the square head-mark. It was widely adopted from Malwa to the southern kingdoms. The head-mark brings them under one group, but they are differentiated from one another by following the general styles of the regions where they were used.

c15.s1.5

D. *The scripts of South India.* There is also a general unity of development in these scripts. The shapes of the letters as well as the medial vowels follow a common pattern.

c15.s1.6

Under these four broad heads fall nine main geographical divisions which clearly emerge in the writing styles of this period, though there is some overlapping owing to political and other factors:

c15.s1.7

A I. *Middle Ganges Valley.* This region includes the greater portion of modern Bihar and Uttar Pradesh but excludes the westerly districts which border on Mathura. Its influence reached as far south as Orissa and the forest regions of the Vindhyan belt. Under the rule of the Imperial Guptas the style of this region was in great favour and it spread into Bengal in the east and Nepal in the north, and exerted considerable influence on the scripts of the westerly regions. The Mehrauli iron pillar inscription of Chandra¹ is entirely in this style, as also is the Udayagiri cave inscription (Fleet,² no. 6) of Chandra Gupta II. In the post-Gupta period the western styles overran this region and ultimately paved the way for the *Nāgarī* script.

c15.s1.8

¹ See below, p. 144.

c15.s1.9

² All references to Fleet are to his *C.I.I.* iii.

c15.s1.10

A II. *Eastern India.* This region properly includes Bengal, Assam and Orissa, but Orissa was already linked with the Middle Ganges Valley. Assam had no separate script of its own. The late inscriptions found there followed the style of Bengal. With Bengal is grouped Nepal, as the writing style of both these regions is derived from the Middle Ganges Valley and there are many remarkable links between the scripts of these two areas. Each of these will be treated as a sub-class—II (a) Bengal and II (b) Nepal. With the rise in the eighth century a.d. of the Pālas, who ruled over the delta as well as part of the Middle Ganges Valley, the regional style was modified under extraneous influences. The growth of proto-Bengali did not take place here until two centuries later. But certain tendencies that originated in this period led on to the growth of the later regional script.

c15.s1.11

A III. *Mathura and the north-western region.* Mathura had been the centre of a line of the Northern Kshatrapas and later of the Kushāṇas. Their important contribution to the development of Indian writing has been noted in the previous chapters. This Kushāṇa style of Mathura continued to influence the region in the Gupta period. Its influence is also marked in the Karamdanda inscription of Kumāra Gupta I and the Bhitari pillar inscription of Skanda Gupta.³ In the time of Harshavardhana this style merged with the Rajasthani style of writing and is represented in this region by the well-known Lakhamandal inscription on the Upper Jamuna.⁴ The ornamental features of this style appear to have originated from the eastern region, but it spread in the whole of the Gangetic Valley, where it led on to proto-Nāgarī, but in the hill states of the north-west, as evidenced from Kashmir and Chamba, the Mathura style continued, leading to the Śāradā script.

c15.s1.12

³ See below, pp. 122–3.

c15.s1.13

⁴ *Epigraphia Indica*, i. 10–15.

c15.s1.14

B IV. *The Rajasthani style.* With Rajasthan was intimately associated Malwa, which, because of its peculiar geographical position, was open to influences from the north as well as the south. The curled tails of the verticals and the side notches continued in the Gupta period, but towards the close of the sixth century A.D. a distinctive style developed which, along with the Mathura style, underwent a further change to become the precursor of the Nāgarī script. This region was the original homeland of Nāgarī, and it was from this source that the Rāshṭrakūṭas derived their proto-Nāgarī writing.

c15.s1.15

B V. *The Kathiawadi style.* This is a continuation of the writing style known from the inscriptions of the Western Kshatrapas. It is easily traceable in the Maitraka records and is intimately connected with the writing known from the inscriptions of the Traikūṭakas and the Gurjaras of Broach.

c15.s1.16

C VI. *The Deccani style.* This is found for the first time in the inscriptions of the Vākāṭakas, in the Eran pillar inscription of Samudra Gupta, and in the Udayagiri cave inscription of Chandra Gupta II (Fleet, no. 3). In the fifth century A.D. its influence was felt in the inscriptions of the Kadambas, the Gurjaras of Broach, the Western Chālukyas, and the Pallavas and the Madharas in Andhra. It survived in the inscriptions of the kings of Śarabhapura and those of the Eastern Gaṅgas to the end of the seventh century A.D. Its influence is also noticeable in the Bilsad inscription of Kumāra Gupta I. This style did not leave any trace in the later scripts.

c15.s1.17

D VII. *Mysore and Mahārāshṭra.* This style is a continuation of that known from later Śātavāhana records. Its most fully developed form is found in the inscriptions of the Chālukyas and the Rāshṭrakūṭas, by which time it had attained a distinctive character, to which the name of proto-Kannadi can be given.

c15.s1.18

D VIII. *Andhra.* The style known from the Ikshvāku records died an immediate death after their collapse. Only faintly is its character recognizable in a few inscriptions of the Bṛihatphalāyanas and the early Pallavas. Side by side with the Deccani influence we notice here a close similarity to the style known from the Chālukya inscriptions, especially after the establishment of the Eastern Chālukyan dynasty. This close similarity was maintained even after the regional scripts, Telugu and Kannada, were later evolved, but in this period the style had some relation to the writings of the extreme south.

c15.s1.19

D IX. *South India.* Not much evidence is available from the extreme south. The early inscriptions of the Pallavas show the long verticals of the Ikshvāku style and also give evidence of a technical tradition seen earlier in the Bhattiprolu inscriptions. The Sanskrit records are again divisible into two styles, one mainly found in inscriptions from the northern part of the region, showing influence from the Vākāṭaka and Chālukyan records, and the other occurring in southern inscriptions from Kāñchī, Trichinopoly, Annamalai, Mamallapuram, &c., which are all very late, and which display a new and special style showing great draftsmanship in the formation of the letters. This is the Grantha style of the seventh century a.d., but so far its earlier counterpart has not been found. In a few letters the influence of the north is traceable and in others we can discern the influence of the Deccani style. But the Grantha style must have had its precursor in this very region. It cannot be wholly derived from the north or the Deccan.

c15.s1.20

This division agrees almost entirely with what we learn from the account given by Alberuni,¹ provided we bear in mind that he describes conditions obtaining two centuries later. Out of the eleven scripts mentioned by him three, viz. Ardhanāgarī, Malwārī, and Saindhava, which were current in Bhatiya and Sindh, are hardly known to us, as no record from this region has so far been published. Three inscribed potsherds, which were recently discovered in the Bhambhor excavation in Sindh,² show a style approximating to Ardhanāgarī. One style, which Alberuni calls Bhaikshukī, used in Uduṇpur in Pūrvadeśa (possibly Audaṇḍapura, modern Bihar Sharif—a seat of an important Buddhist monastery), was probably a special writing used by the Buddhists, as he himself says. The remaining scripts, Siddhamātṛikā, Nāgara, Gaurī, Lārī (or Lāṭī, i.e. Kathiawadi), Karṇāṭa, Andhrī, and Dirwārī (i.e. Drāviḍī), are not difficult to recognize. Of these a little confusion arises between Siddhamātṛikā and Nāgara. The latter term is definitely derived from Varṇanāga,³ applied to a style of writing widely current in Malwa, as Alberuni rightly notes. Siddhamātṛikā, according to him, was current in Kashmir, Madhyadeśa (i.e. 'the region round Kanauj') and Vārāṇasī (Banaras), and must refer to the style termed by Fleet '*kuṭila*' and Bühler 'acute-angled', but by about tenth century A.D.¹ it was displaced by Nāgarī in the Ganges Valley and Śāradā in Kashmir. The use of the term 'Siddhamātṛikā' for the *kuṭila* writing is also confirmed by Ziogon's note on the script of the Horiuzi Palm-leaf MS.²

c15.s1.21

¹ *Alberuni’s India*, Eng. tr. by E. C. Sachau, London, 1910, i. 173.

c15.s1.22

² The potsherds were sent to me by the Director of Archæology, Pakistan, for decipherment.

c15.s1.23

³ Actually the phrase is ‘Varṇa-nāga-kṛipāṇikā’, because an extra *kṛipāṇa* (dagger) is placed on the *nāga* (serpent); see *Epigraphia Indica*, xxxi. 28.

c15.s1.24

¹ See Vogel, *Antiquities of Chamba*, for the presence of *kuṭila* character in Chamba.

c15.s1.25

² Max Müller and B. Nanjo, *The Ancient Palm Leaves* (Horiuzi Palm-leaf MS.); *Anecdota Oxoniensia*, vol. i, pt. iii, Oxford, 1884, pp. 14 and 16.

c15.s1.26

*New techniques*

c15.s1.27

This development of the alphabets reached its culmination later, when the regional languages developed and they became associated with one or the other script. That stage was a long way ahead. This chapter is concerned only with the process of change that the letters were undergoing at this period. This will be analysed, as in the last chapter, on the basis of three noticeable stylistic tendencies. The first was the result of a technical development, either in the use of new tools or in the new manner of using the old tools; the second was due to the taste for better or ornamental forms; and the third was dictated by a desire for speed and simplification, which is the real motive in cursive writing.

c15.s1.28

(i) *Technical*. Two different types of tools were used for writing and these had a marked influence on the shape of the letters. As we have seen, the pen was used throughout North India. The use of the pen gave three important characteristics to the letter-forms: (*a*) a definite head-mark to the letters of the north. It assumed the shape of a solid triangle, called by earlier palaeographers 'wedge' or 'nail-head'. (*b*) From the fifth century A.D. onwards the pen leaves behind a blot at the foot of the verticals, which gradually begins to grow and, as shown in Fig. 11, new forms of the letters are evolved. In some cases this blot develops a tail as in *da* and *ra*, and in some, as in *bha* and *sa*, it grows into a triangle and later opens its mouth. This blot will be henceforward referred to as the foot-mark. (*c*) From the sixth century there is noticeable a peculiar twist of the pen which makes an interplay of thick and thin lines in the ornate medial vowels (Fig. 12). The same tendency led to the origin of the *kuṭila* letters (Fig. 13). The use of the pen in the north was mainly responsible for the new shapes which so fundamentally differed from the southern characters.¹

c15.s1.29

¹ See Pattadakal pillar inscription of the Chālukyan king Kīrtivarman II, *c.* A.D. 747–57, which gives the same inscription in the two characters, northern and southern; *Epigraphia Indica*, iii. 4–6.

c15.s1.30

About this time writing in the south shows the growing influence of the stylus, as a result of which the letters become more rounded

c15.s1.31

[Figure: EVOLUTION BY THE PEN TECHNIQUE FIG. 11]

c15.s1.32

and have waves on their arms, as shown in Fig. 14. These waves are probably due to the nature of the palm-leaf on which stylus was used—a style developing from this source affecting the inscriptions as well.

c15.s1.33

(ii) *Taste for ornamentation.* Different kinds of ornamentation are noticeable in different regions. Their details will be discussed when the individual palaeography is taken up. Generally speaking, the embellishment is seen in the shapes of the letters (Figs. 13, 15) and in the formation of the medial vowels (Figs. 12, 16). This development is again seen in the formation of the lines in the respective scripts, as dictated by the pen and the stylus. Both in the north and the south the medial vowels are the first to show flourishes. The best examples from the north are seen in the Bilsad inscription and the Mehrauli pillar inscription, both of the fifth century (Fig. 12). In the sixth century, as illustrated in the inscriptions of Yaśodharman and Mahānāman, the medial vowels grow to their full extent. In the seventh century the flourish is further embellished by floral decoration (not shown in the plate). From

c15.s1.34

[Figure: MEDIAL VOWELS IN NORTH INDIA FIG. 12]

c15.s1.35

the last quarter of the sixth century A.D. another tendency is marked in the North Indian inscriptions. The right vertical limbs of the letters become slightly bent inwards (*kuṭila*), and hence Fleet's term *kuṭila* alphabet. At the same time, because of this bending, this vertical makes an acute angle with the base line, and hence Bühler's term 'acute-angled' alphabet. Bühler realized correctly that this acute angle is found much earlier than the *kuṭila* tendency and hence he traced this type of writing back to the fifth century A.D. But such a tendency can be seen in letters such as *sa* even earlier. Thus it is very difficult to adhere to Bühler's term. Fleet's term, though derived from a late inscription,¹ if understood to imply the style of writing which shows ornamental forms with

c15.s1.36

¹ Dewal inscription of (Vikrama) year 1049. See Cunningham, *A.S.R.*, i. 355, pl. LI.

c15.s1.37

[Figure: ORNAMENTAL EVOLUTION IN THE NORTH FIG. 13]

c15.s1.38

their right limbs slightly bent, may be retained, as it conveys a chronological sense. This style survives the end of the period dealt with in this chapter.

c15.s1.39

In South India embellishment is seen in the upper (in the case of *la*) and lower (in the case of *a*, *ka*, *ta*, and *ra*) curves of the letters (Fig. 15). In the earliest Pallava, Bṛihatphalāyana, and Kadamba² records of the fourth century a.d. the verticals are unusually prolonged downwards, probably from the influence of the Ikshvāku records, but later the normal length is restored. In the seventh century A.D. the curves grow either almost into a circle, as in the

c15.s1.40

² Chandravalli inscription of Mayūraśarman, *An. Rep. of Mysore Archaeological Survey*, 1929, p. 50, pl. XI.

c15.s1.41

[Figure: EVOLUTION BY THE STYLUS TECHNIQUE FIG. 14]

c15.s1.42

case of *a* and *la*, or into double parallel lines, as in the case of *ka* and *ra*. The medial vowels in the south follow also a different course of development (Fig. 16). Fundamental differences from the north are seen in the case of *au* and *ṛi*. The medial vowel *o* in *lo* is cursive. The right limb of *ā* is prolonged downwards as is *e* on the left side of the letters. The most important development is seen in the application of the medial vowels *e* and *ai*. Unlike the North Indian fashion, where they are applied above the letter, in South India the signs hang downwards, in some cases starting from the middle of the letter. This tendency comes down from the practice seen in the letters *mā* and *me* in the Girnar version of the Aśokan R.E.¹ The medial *i* grows into a full circle, but optionally in the extreme south it retains the form of a broad half circle. The medial *ī* has an additional line, curl, or circle to distinguish it from the medial *i*. The curl of the medial *u* grows until in the seventh century A.D. it is doubled. For medial *ū* an additional mark is given either on the right or on the left of the sign for the medial *u*. Another rounded form, as seen in *bhu* or *bhū*, also develops in the south.

c15.s1.43

¹ See above, p. 47.

c15.s1.44

[Figure: (a) ORNAMENTAL EVOLUTION IN THE SOUTH
(b) FORMS OF OM FIG. 15]

c15.s1.45

(iii) *Tendency to simplification.* This tendency is seen throughout and writers adopt it whenever they prefer and whatever their writing material. It would be wrong to say that the simplified style was usually adopted in the case of copper plates or manuscripts. The two copper plates of Harshavardhana (Fig. 12) are in two different styles—the Banskhera plate in the ornate style and the Madhuban plate in the simplified style. The Sonpat seal of Harshavardhana (Fleet, no. 52) is in the ornate style, but the Asirgadh seal of the Maukhari Śarvavarman (Fleet, no. 47) is in the simplified style. The two Bodhgaya inscriptions again illustrate this point, which was realized even by Rudolf Hoernle when he wrote that some parts of the Bower MS. are in an ornate style and others are written in a slovenly manner.¹ Here the writing material cannot have been responsible for differences in the style of writing. A tendency to simplification is to be found in the hands of some writers but not of others. Wherever it is traceable, it can be seen to follow the ornate or complicated way of writing, or at least the accepted form already achieved by a particular technical process,

c15.s1.46

¹ *Indian Antiquary*, xxi. 30.

c15.s1.47

FIG. 16

c15.s1.48

the desire being to reduce the time taken in producing the complex form without sacrificing its individuality. This description applies only to simplified as distinguished from cursive writing, and I confine myself to the former because I have so far found no true example of early Indian cursive writing,¹ be it on copper plate, stone, palm-leaf, or birch-bark. It is not possible to understand the whole process of simplification as no cursive writing is available for comparison. What I am trying to do is to determine the adoption of the particular simplified forms in the usual course of writing and to find as far as possible the time when they became current. From the available material we can get only a few clues to elucidate

c15.s1.49

¹ I reserve the term ‘cursive writing’ for quick hand-written documents in which the letters of the same words grow tails and are connected together. This is not the case in any of the manuscripts discovered so far. Hence we are not faced with problems connected with cursive writing though we have, of course, cursive forms as a result of simplification.

c15.s1.50

[Figure: EVOLUTION BY SIMPLIFICATION FIG. 17]

c15.s1.51

the way in which simplified writing was gradually developing. This is best illustrated in the development of the forms *a* and *la* in South India (Fig. 17). This simplification much affects the conjuncts, in which either the first (as in the case of *ṇḍa*), or the second (as in the case of *jña*), letter is reduced. The same is the case when the second element of the conjunct is *ya* or *ra*. In these conjuncts the writer neglected the rule of grammar, which requires the first element of the conjunct to lose its inherent short vowel *a*, and hence become constricted, in order to combine with the second element. He was influenced solely by the motive of facile writing. It is this same tendency which reduces the triangular head-mark to a simple horizontal line, as in the Indor plate of Skanda Gupta and in the series of Damodarpur copper plates. Another feature is noticeable in the inscriptions of the north-western region, where the triangular foot-mark of *ra* gives place to a simple slant. The simplification also affects the medial vowels, in which the ornate forms give place to straightforward lines. Later still the bent (*kuṭila*) right limbs of the letters in the north become straightened but leave a tail behind—a feature of most of the North Indian letters. This tendency is to be found in its final stage in this period. It is fully marked in the development of the regional scripts of the next period. In fact the preference for simplification in one or another way is one of the important factors contributing to the differentiation in the regional scripts. We are not concerned with this process in this chapter, which deals mainly with (i) the writing style that a particular region inherited in the fourth century a.d., (ii) inter-regional influences due to the setting-up of new kingdoms or the movement of the scribes, (iii) the change due to the technical development in the ordinary forms of the letters and in ornamental tastes, and (iv) the full ornamental letters or their simplified forms.

c15.s1.52

*New symbols*

c15.s1.53

The most conspicuous symbol in the copper plates of this period is that which stands for the auspicious sound *oṁ*, the various forms of which are given in Fig. 15*b*. It will be seen that in North India it generally assumes the form of a curled head with a tail either to the left, right, or bottom. In South India this tail is generally spiral. Another development is seen in the case of a final consonant used without any vowel. The usual practice was to write the final consonant in smaller size slightly below the usual line. In this period we see the use of an arc-like mark over the consonant, or a slanting mark attached to the right foot of the consonant in order to indicate that the vowel is absent. This last practice is the real source of the modern *halanta* (i.e. consonant without vowel). The use of the *jihvāmūlīya* and the *upadhmānīya* has been noted in the last chapter, and the forms given there were usually adopted in North India. In the south slight variations are seen, and these will be discussed in the relevant sections. The cross-form of *ṛi* noticed in the last chapter curves the arms of the cross in one type, while in the other type the left curves are separated from those of the right and joined by a bar, and the upper left curve optionally has a line head-mark. These forms are found in the Bower MS. In the Horiuzi MS. only the simplified type of the second form is seen. In this case the right curve is straightened, and the form comes closer to that of the Nāgarī letter. The initial *ṛi* is formed by adding a slant to the foot of the right limb. A new letter *ḷi* is also met in the last manuscript. It consists of a fish-hook upside down with the hook turned to the left. The initial *l̥ī* is formed by adding a curved slant to the vertical of the first.

Linked texts (6)