Indian Palaeography c11: 3. Origin of Brāhmī

Indian Palaeography c11

Language: English

3. Origin of Brāhmī

c11.2

SINCE Bühler wrote *On the Origin of the Indian Brāhma Alphabet*, no new material has been found which could throw further light on this vexed question. Ojha¹ critically analysed the derivations proposed by Bühler and showed that the differences in the letter-forms, except in a few cases, were so fundamental that the Brāhmī forms could not be derived from those of North Semitic characters. The discovery of the Indus Valley writing has led some scholars to guess that Brāhmī was locally developed out of the Indus Valley system. But no one has been able to demonstrate how this evolution could take place.² In fact, the long gap between the time when the Indus Valley writing was in use and that when Brāhmī is first attested throws doubt on such a supposition. However, in recent years some further points have emerged from the discussion on the origin of the alphabet,³ which enable us to review the origin of Brāhmī afresh.

c11.3

¹ *Bhāratīya Prāchīna Lipimālā*, pp. 17–31.

c11.4

² See above, pp. 19–21.

c11.5

³ See Diringer, *The Alphabet*, London, 1949, pp. 195–221, and his bibliography; I. J. Gelb, *A Study of Writing*, London, 1953.

c11.6

Before we tackle the problem, it is necessary to disabuse our minds and remember that the development of Brāhmī was not an isolated phenomenon. Its origin and growth are rooted in the civilization of the Ancient World of which India was a member. The discovery of the Indus pictographs or ideographs, or whatever that system of writing was, provides ample evidence of direct contact between India and the cultural centres of Western Asia. The appearance of the Indo-Aryans in history further strengthened this contact. Whether the Indian Aryans were literates or not is a different question, but it may be conceded that they were in contact with peoples who knew the art of writing. However, we have no definite evidence of the Indian Aryans practising this art. Ojha⁴ has given an analysis of the literary evidence and traced the concept of the letters (*akshara*), the vowels and the consonants, and the grouping of the last into *ghosha* consonants, *mūrdhanya* (cerebral), *dantya* (dental) letters, &c., in the *Upanishads* and the *Āraṇyakas*. In the *Aitareya Brāhmaṇa* (5. 32) he finds *oṁ* stated to be formed by the combination of *akāra*, *ukāra*, and *makāra*. Later still in the works of the grammarians we find minute descriptions of the letters, the rules for their combination, and the changes that they undergo. Yet we have no concrete evidence of alphabetic writing in India before the time of Aśoka. How then are we to explain these concepts evidenced in the early Sanskrit literature? Some may maintain that the Brahmins, being fastidious in the pronunciation of the Vedic *mantras*, were led to this phonetic analysis which found perfection in the hands of the grammarians, while others would surely find parallels in the contemporary attempts made in Western Asia at the evolution of an alphabet. It is in Western Asia alone that the different stages in the evolution of the alphabet are more or less clear, but in Sanskrit literature, until we come to the time of the grammarians, acquaintance with the vowels and the consonants is superficial. Diringer¹ rightly maintains the alphabet to be originally an invention possibly in one region; it is no wonder that the principle underlying alphabetic writing soon spread with trade and commerce, and it may be suggested that as the Brahmins found this principle a great help in their Vedic recitals, they developed it further in their grammatical treatises.

c11.7

⁴ Op. cit., pp. 1–16.

c11.8

¹ *The Alphabet*, London, 1949, p. 216.

c11.9

The precedence of grammar over the Indian system of writing is indubitably proved by a critical analysis of the earliest examples of writing known in India.² The phonetic system of Indian grammar, though it conceives of vowels and consonants, still does not have pure consonants. Here we have a system which I. J. Gelb describes as being in the syllabic stage, but this syllabary has been further standardized. Each consonant, when not followed by an explicit vowel, inherently implies only one vowel, i.e. *a*, and no other. This peculiar characteristic of the Indian system is traced by Diringer³ ‘to the influence of the Aramaic language, in which the final *aleph* predominated’. But the necessity of this inherent *a* is also felt in the Sanskrit language, in which the consonants are required to be pronounced with some definite vowel unless they are explicitly characterized as vowelless. It is therefore not necessary to look further afield for this feature. The peculiar character of the Indian alphabet is the contribution of the Indian grammarians, and, whatever script has been used in India to express this language, this notion of the alphabet has remained constant. It is essential that the nature of this alphabet is understood before the question of the origin of any of the scripts is taken up.

c11.10

² See also, pp. 46–47.

c11.11

³ *The Alphabet*, London, 1949, p. 336.

c11.12

In the arrangement of the letters the vowels come first. These are arranged here in the order of short, long, *guṇa* and *vṛiddhi*:

c11.13

| *Short* | *Long* | *Guṇa* | *Vṛiddhi* |
| (1) a | (2) ā | .. | .. |
| (3) i | (4) ī | (10) e | (11) ai |
| (5) u | (6) ū | (12) o | (13) au |
| (7) ṛi | (8) ṝi | .. | .. |
| (9) ḷi | .. | .. | .. |

c11.14

Then follow the consonants with the inherent *a*, divided into twenty-five mutes falling into five classes, four semi-vowels, three sibilants, one aspiration, one pure nasal, and three voiceless spirants. According to some the pure nasal and the three voiceless spirants are to be taken as vowels. Each class of the mutes consists of hard and soft consonants, their aspirates, and a corresponding nasal:

c11.15

| *Hard* | *Aspirate* | *Soft* | *Aspirate* | *Nasal* |
| ka | kha | ga | gha | ṅa |
| cha | chha | ja | jha | ña |
| ṭa | ṭha | ḍa | ḍha | ṇa |
| ta | tha | da | dha | na |
| pa | pha | ba | bha | ma |

c11.16

The four semi-vowels are *ya*, *ra*, *la*, and *va*, and the three sibilants are *śa*, *sha*, and *sa*. The aspiration is *ha*. The pure nasal is *anusvāra* (*ṁ*), and the three voiceless spirants are *ḥ* (*visarga*), *ḫ* (*jihvāmūlīya*), and *ẖ* (*upadhmānīya*).

c11.17

As the consonants contain an inherent *a* there arose the concept of the *halanta* or the vowelless consonant. This concept led to the formation of the conjuncts, which have their own regulations as provided in the rules of *sandhi*.

c11.18

This alphabetic system is maintained in India, with minor additions or omissions, down to the present day, though it is not phonetically suited to the various provincial languages in India.

c11.19

It is on the basis of this alphabet that we have to understand the origin and the development of Brāhmī.

c11.20

There is no exact correlation between the Brāhmī character and the alphabet given above—a point which decidedly proves that neither was Brāhmī the pure invention of the Indian grammarians nor was the above alphabet developed out of Brāhmī. As shown in Fig. 1, the exceptional formation of *kha* and *tha*, which

c11.21

[Figure: DERIVED LETTERS IN BRĀHMĪ UNASPIRATES ASPIRATES EXCEPTION ANOTHER EXCEPTION ANOTHER EXCEPTION FIG. 1]

c11.22

have no connexion at all with *ka* and *ta*, is a sufficient proof of the former, and the derivation of the cerebral letters from *tha* supports the latter. However, the influence of phonetics on the formation of the Brāhmī script cannot be altogether brushed aside. Despite exceptions, we find some regularity in the evolution of the signs. As shown in Fig. 1, the aspirated letters have regular additions with the exception of the two. Each class of mute has been provided with a nasal, as required by the grammar, and further a dot is invented for the *anusvāra* (Fig. 2). Similarly, *guṇa* and *vṛiddhi* have also regular additions. But if these forms, which are made by additional strokes, are taken out, we actually get three

c11.23

FIG. 2

c11.24

vowels—*a*, *i*, and *u*, and nineteen consonants—*ka*, *kha*, *ga*, *cha*, *ja*, *ta*, *tha*, *da*, *na*, *pa*, *ba*, *ma*, *ya*, *ra*, *la*, *va*, *śa*, *sa*, and *ha*—a total of twenty-two letters, exactly the same number as found in North Semitic or Aramaic. The cognate sounds can be correlated as follows:

c11.25

| ʾaleph | *a* | kaph | *ka* | gimel | *ga* |
| ʿayin | *i* or *e* | qoph | *kha* | ṣad | *cha* |

c11.26

| zayin | *ja* | pe | *pa* | lamed | *la* |
| taw | *ta* | beth | *ba* | waw | *va* |
| ṭeth | *tha* | mem | *ma* | shin | *śa* |
| daleth | *da* | yod | *ya* | samekh | *sa* |
| nun | *na* | resh | *ra* | he | *ha* |

c11.27

The only letter left unconnected is *kheth* in North Semitic and *u* or *o* in the Brāhmī.

c11.28

Of these letters the similarity is clearly seen in the following:

c11.29

[Figure: NORTH SEMITIC & EARLY PHOENICIAN BRĀHMĪ Semitic forms are adapted from Diringer's The Alphabet FIG. 3]

c11.30

This similarity hardly leaves any doubt that Indian Brāhmī was created on the basis of the North Semitic letters. But Brāhmī is not a slavish adoption of the North Semitic signs. There is considerable local genius visible in its formation. The most fundamental change introduced by the Indians is in giving a definite direction to the letters. The way in which *'aleph* is changed into *a*, *lamed* into *la*, &c., suggests that Indian Brāhmī was written from left to right from the very beginning.

c11.31

Attempts have been made in the past to show that Brāhmī was originally written from right to left and that only later was its direction changed. Bühler was the first to propound this hypothesis on the basis of a faulty coin from Eran¹ in which some letters are found to be reversed. Though many others have clearly demonstrated Bühler's mistake, Diringer once again reasserts the claim and quotes as another example the Yerragudi M.R.I. of Aśoka.¹ He concludes: 'As Mr. Sahni points out, this inscription leaves no doubt that the *boustrophedon* style was known in the time of Aśoka. There is, thus, sufficient evidence of the existence of an earlier Brāhmī script written from right to left, followed . . . as in the development of the early Greek script by a transitional system of writing in *boustrophedon* style.'² If further examples of this type prove the point, I could cite many more from Ceylon.³ But mere number is not conclusive. The argument must be based on technical analysis of the evidence. The Yerragudi inscription is a barbarous example of writing, in which the direction can hardly be said to be *boustrophedon*, and the Ceylonese inscriptions, which are all very late, betray the writers' ignorance of the true system rather than preserve an old practice. Such mistakes are found in several other inscriptions. In the Bhattiprolu inscriptions⁴ there are some letters which are similarly reversed. In one of the Chālukyan inscriptions⁵ the writing goes from below upwards. These stray examples cannot in any way change the style of the main character, which from the very beginning, when Brāhmī was created on the basis of the North Semitic writing, had determined its direction from left to right.

c11.32

¹ See below, p. 61.

c11.33

¹ *An. Rep. A.S.I.*, 1928–9, pp. 161–7.

c11.34

² Diringer, op. cit., p. 339.

c11.35

³ See below, p. 217.

c11.36

⁴ See below, p. 70.

c11.37

⁵ See below, p. 211, no. 4. III.

c11.38

There has also been a controversy over the question of the particular branch of the Semitic alphabet from which Brāhmī is 'derived'. Opinions have varied from the South Semitic to North Semitic, or particularized to Phoenician or to Aramaic. As has been shown before, the similarity with the North Semitic letters is the closest. Within this broad division Diringer⁶ prefers the 'Aramaic alphabet as the prototype of the Brāhmī script', since Aramaic is known to have been used right up to the borders of India. But the choice need not be limited to Aramaic alone, as the Brāhmī letters are not literally 'derived' from the Semitic as is commonly understood, but are only based on them. The idea may have been brought from further afield. This supposition is supported by the fact that the two basic forms of *a* in Aśokan Brāhmī have their counterparts only in North Semitic, but not in Aramaic.

c11.39

⁶ Op. cit., p. 336.

c11.40

Whatever may be the particular source of inspiration, Brāhmī is a creation of the Indian Pandits. The script has been evolved to suit the local grammar, and hence the semi-syllabic character of Brāhmī and the invention of the strokes for the medial vowels. The only case where the script conflicts with the grammar is in the formation of the conjuncts.¹ Here we find two or three letters placed one above the other, disregarding the presence of the inherent *a* in the consonants. The same is the case when the medial strokes are applied to the consonants. In these examples they are taken to be pure consonants, at least in the early writings known from India. Later on the first letter of the conjunct came to be mutilated, suggesting the loss of the inherent *a*. But this is a later practice. The original construction as seen in Aśokan Brāhmī still preserves the true alphabetic nature of the writing, and thus, despite innovations, Brāhmī falls within the general class influenced or inspired by Semitic.

Linked texts (1)

Aitareyabrāhmaṇa — Mahidāsa Aitareya, Sanskrit