Indian Palaeography c9: 2. The Indus Script

Indian Palaeography c9

Language: English

2. The Indus Script

Definition

c9.s1.1

THE earliest script of the Indo-Pakistan subcontinent, known from discoveries of seals first at Harappa in the Montgomery district of Punjab (Pakistan), and later at Mohenjodaro in the Lakarna district of Sind, is termed here the ‘Indus Script’, as it was mainly in use, if it did not actually originate, in the Indus Valley zone. These discoveries received greater importance and attention after 1922, when the excavations conducted at these two sites under the direction of Sir John Marshall brought to light a new civilization, ranking in content and time with the civilizations of the Ancient Orient. The Indus Valley found a place in the Ancient World of Bronze Age cultures, in a period when the art of writing had already developed.

Origin

c9.s2.1

We still do not know where the art of writing was first invented, and with this question is related the origin of the Indus Script. There is nothing more to add to what Dr. G. R. Hunter wrote in 1934:

c9.s2.2

A cursory examination of the script of Mohenjodaro and Harappa will reveal that it is distinctive. It is neither Sumerian, nor any other known script, though it bears certain resemblances to several. Some of these are doubtless coincidental, since in the very nature of pictographic writing it is hardly possible to avoid some similarity in depicting the same object. A closer examination will establish that it is precisely the commoner signs of our texts that are the most distinctive—e.g. [Figure:] At the same time it would be rash, in the present state of our knowledge on the subject, to rule out of court the hypothesis of a common descent from some remote ancestor for the script of Harappa and any other pictographic script. We know so little, after all, of the ultimate pictographic ancestry of any script, even Sumerian.

c9.s2.3

*Attempted decipherments*

c9.s2.4

But while the Sumerian and the hieroglyphic writings have been successfully read, and the meanings of the inscriptions written in these characters have opened to us the minds of the builders of those civilizations, the Indus Script remains undeciphered. Only the material remains have acquainted us with the day-to-day life of the people who formed this civilization. Their language also remains a secret. The material remains have been interpreted in the light of the contemporary world or, more commonly, against the background of the succeeding historical cultures in the subcontinent, and various conclusions about the people and their language have been drawn. These are, of course, hypothetical. It is on these hypotheses that attempts at deciphering the script have been made. The first in the field was L. A. Waddell, who believed in the identity of the Sumerians and the Aryans, tried to read the inscriptions in Sumerian on the basis of a few apparent links, and found in them names of Vedic and Epic heroes. Dr. Pran Nath followed the same line, but supposed the language to be some form of old Sanskrit or Prakrit and utilized the Brāhmī script of a much later date to ascertain what he believed to be the correct sound value of the symbols. Another suggestion of Pran Nath’s, that the significance of the symbols might be arrived at on the basis of Tantric signs, was independently developed by Śankarananda and also by Dr. B. M. Barua, both assuming that the writing is alphabetic. Professor B. Hrozny took the people to be proto-Hittites, an earlier branch of the Indo-Europeans, and tried to interpret the inscriptions with the help of the Hittite language. As opposed to these suggestions, Father H. Heras assumed that the people were Dravidians and tried to reconstruct a proto-Dravidian language for them, and for his interpretation compared the Indus signs with Sumerian, hieroglyphic, and proto-Chinese symbols. Finding no clue to the people and their language, Sir Flinders Petrie tried to interpret the Indus symbols as ideographs, on the analogy of those of Egypt and on the further assumption that the Indus seals contain only titles of the officials. Later, Meriggi realized Petrie’s mistake in taking all the symbols as ideographs. He believed the Indus Script to be an ideo-phonographic system of writing and assumed some symbols to be ideograms and others phonemes. For determining their meanings he often referred to the Hittite vocabulary and to Hittite hieroglyphs. G. de Hévesy rejected all these suppositions and drew attention to the similarities between the Indus Valley Script and the script of Easter Island, maintaining that the latter was the progenitor of the former. But apart from a certain apparent similarity there is nothing to authenticate the antiquity of the script of Easter Island.

c9.s2.5

There is very little external evidence to prove or disprove the readings suggested by the different scholars mentioned above. The only effective argument against each of them is the fact that the criticism levied by one against the other is equally applicable to the critic’s own method. These difficulties were fully realized by Gadd, Sydney Smith, Langdon, and Hunter, and they all chose to confine themselves to listing the signs, mechanically counting their use in the different inscriptions, and suggesting their possible significance. Beyond this mechanical process Gadd has ventured to discover in them ‘an ancient Indo-Aryan language’ and actually suggested the reading *putra* (son) for [Figure:] [Figure:] [Figure:]. Langdon and, following him, Hunter have tried to see some connexion between the Aśokan Brāhmī and the Indus Script, and the latter especially has ventured to suggest vowel signs on this very basis. Both of them argue for a syllabic system of writing. Sydney Smith alone has refrained from falling into any such temptation. He has commented on the mechanical nature of the writing, and has attempted to discover the determinatives, end-signs, and beginning-signs. Beyond this stage it is hardly possible to go with the materials at present available. Professor J. Friedrich rightly points out the main difficulty:

c9.s2.6

The decipherment of an unknown language in a known script, such as the cuneiform Hittite, Etruscan, or Urartean; or of a known language in an unknown script, such as the Greek in the Cyprian syllabic script, is much more hopeful than that of an unknown script and unknown language, as the case is with the hieroglyphic Hittite or Cretan languages. For such works certain points of contact are necessary. Nothing can be made out of nothing.

c9.s2.7

Another great difficulty is the total absence of long inscriptions. We have so far found only short epigraphs on seals, or metal tablets, or rare impressions on pottery.

A New Approach

c9.s3.1

As the purpose of this book is to study the script in particular, and not the inscriptions as such, the method adopted here is different from that followed in the works of Marshall and Hunter. But they remain the basic groundwork for all those who wish to decipher the Indus Script. Here we shall simply attempt a morphological study of the script in order to familiarize ourselves with the nature of the writing followed by the Indus people.

c9.s3.2

*Method followed here*

c9.s3.3

In the plates (I and II) illustrating the script an attempt is made to arrive as far as possible at the correct form by comparing the symbols in various publications and copying the most common type. No tracing of the symbol from any particular seal is therefore to be expected. The sign-lists given by Hunter and in Marshall's volumes (pp. 434–52 and pls. CXIX–CXXIX) have been compared with the original photographs of the seals and readjustment made. Some further symbols from Mackay's *Further Excavations at Mohenjodaro* have been added.

c9.s3.4

A new principle of classification has been adopted here. Pl. I*a*, group A, illustrates stylized living beings; group B of this plate and Pl. I*b* show objects which can be easily recognized. In Pl. II *a*, *b* are shown those symbols which take geometrical forms, such as ovals, triangles, quadrangles, angles, crosses, &c. A few of these can also be construed as highly stylized forms, e.g. the oval (Pl. II*a*. 1. 1)¹ does not appear to be different from the objects hanging from either end of the pole carried by the man in Pl. I*a*. IV. 2–11. Again in Pl. II*a* the two complicated symbols in columns 14 and 15 are obviously magical emblems—one of crosses and the other of the mysterious so-called taurines and a square. Pl. II*b* shows in the last five lines strokes—horizontal, vertical, or slanting—often in combination as they have been actually found. In Pl. II*b*. XII. 10–15 are given the symbols, termed brackets, as they are found bracketing other symbols in the seals (see reading no. 2). In Pl. II*a* seven readings have been given.

c9.s3.5

¹ The first number represents plate, the second the line or row in which the particular symbol appears, and the third the column in which it lies.

c9.s3.6

The symbols are copies from the photographs of the seal impressions as they have been usually published. Probably this is the correct way of looking at them, rather than as on the seals themselves.

Direction of Writing

c9.s4.1

On the analogy of the Egyptian hieroglyphs, as well as from the fact that the animals mostly face right, it has been suggested that the direction of writing is from right to left, i.e. it begins from the side of the animal's head and goes to its tail. Another argument adduced is from the symbol, Pl. II*b* XII. 3, which has four strokes in the first row and three in the second beginning from the right. But the symbol, Pl. II*b*. XII. 2, suggests just the reverse. Further evidence is taken from the seal given in reading no. 7, where it is suggested that the writing begins from the right but owing to shortage of space some symbols are put on the left below the line. On the other hand, Marshall argues that the writing which begins from the right is actually *boustrophedon*. He cites the evidence of the seals, readings nos. 2 and 3, and points out that the second line possibly begins from the left; reading no. 6, as he says, shows the reverse of the last three symbols in no. 5. But a careful examination of the readings 2 and 3 in the original seals shows that the presence of the animal's head with horn resulted in the second line being placed on the left half. The proof can be seen in reading no. 4, where the second line is in the middle. Therefore the arguments produced are not sufficient to confirm either suggestion. However, if pictographs are predominant in this writing, and if it has any resemblance to the Sumerian pictographic and Egyptian hieroglyphic scripts, the writing may be *boustrophedon*.

Nature of the Script

c9.s5.1

In all, 537 symbols have been listed here. Out of these the living beings, which are of 8 categories, include 99 symbols; the stylized objects, which are 19 in all, have 132 symbols; and the remainder, 306, are grouped under other categories. There can be no finality in the correctness of the number as long as the script remains undeciphered. Probably symbols as in Pl. I*a*. II. 6, 7, or in Pl. I*a*. VII. 5, 6, may represent the same sense. Such possible corrections must be borne in mind.

c9.s5.2

Quite obviously a system with such a large number of symbols must be very different from any known alphabetic scripts. On the other hand, the compound symbols, such as Pl. I*a*. III. 1 (= Pls. I*a*. I. 1 and II*b*. IX. 5), Pl. I*a*. X. 6 (= Pls. I*a*. X. 1 and I*b*. VII. 10), Pl. II*a*. IV. 4 (= Pls. II*a*. III. 7, Pl. I*b*. IV. 3), suggest the same principle as underlies the conjuncts in the later Indian scripts; e.g. म्ह = म + ह; च्य = च् + य; क्ख = क् + ख. Probably this feature led Langdon to propose the syllabic nature of the script, and Hunter further argued that this very nature is preserved right down in modern Indian scripts, which are not truly alphabetic except in the case of the initial vowels (e.g. क is actually क् + अ). This can hardly be dogmatically asserted when the recognizable objects easily suggest pictographic or ideographic meanings. On the other hand, the way in which strokes—horizontal, vertical, or slanting—are added to the original object symbols (e.g. fish sign or jar sign) militates against taking them to be simple pictographs or ideographs. The addition of strokes to the original sign is one of the chief characteristics of the Indus Script. They cannot all be taken to be *guṇa* additions. If they have any other meaning, they suggest different values. But such additional strokes are not to be found at all in homo-signs and some others. Obviously in these cases the pictographic character is preserved. Hence it seems that the nature of the Indus Script is picto-phonographic or ideo-phonographic—included by David Diringer among 'transitional scripts'.

c9.s5.3

*Special features*

c9.s5.4

The homo-signs depict a man as a simple vertical line with slanting strokes, two for arms and two for legs. He is single as in Pl. I*a*. I. 1, 2, or double as in Pl. I*a*. I. 3, but more often in action, suggested by the addition of recognizable objects. Symbols, Pl. I*a*. I. 7, 8, have two additional upward strokes, one on either side of the head. They are usually taken as horns, as these play an important role in the Indus civilization, but Father Heras takes them to be hands and suggests a four-handed deity, on the analogy of the later Indian iconography, probably because the strokes start from the shoulder and not from the head. Whatever they are—horns, additional hands, or a simple aureole—they indicate a superman or divine being, and hence the symbol in Pl. I*a*. I. 10 with only a single additional stroke on the left may be a semi-divine being. Symbols, Pl. I*a*. IV. 8–12, probably show two men carrying a load, although two heads with a single body are depicted. The sub-classes show objects carried in the hand, at the waist, over the head, at the leg or feet, and on the shoulders.

c9.s5.5

In the bird-signs one can see a creature in flight (probably a bat), Pl. *Ia*. VI. 14; a bird in a cage or enclosure of some kind, Pl. *Ia*. VI. 12; and a bird with a cross in the belly, Pl. *Ia*. VI. 13. One can also possibly recognize a hen, Pl. *Ia*. VI. 10; a crow, Pl. *Ia*. VI. 9; and a peacock, Pl. *Ia*. VI. 6; but probably these different species are immaterial. What is more important is the addition of strokes, 1–3 in the belly and at the tail.

c9.s5.6

The strokes are clearer still in the fish-signs: (i) one or three downward slanting strokes on either side of the fish; (ii) one vertical or horizontal stroke, or three slanting strokes, in the belly; (iii) two upward slanting strokes as in the case of the divine man; and (iv) an arch-like sign above.

c9.s5.7

In the case of the scorpion one or two horns are discernible, and these can be compared with Pl. *Ib*. I. 11–14. The quadruped-sign, Pl. *Ia*. VIII. 11–12, is taken by Hunter to be a development from the one in Pl. *IIb*. VII. 8. But here again the internal and external strokes are well marked.

c9.s5.8

In the jar and container signs the strokes range from 1 to 3, but in the latter a fourth stroke and also a fifth stroke (Pl. *Ia*. X. 12) can sometimes be seen. The container is also found in combination with others, like wheel, cross, &c. The horn (Pl. *Ib*. I. 15) has significant dots at the mouth, suggesting the historical cornucopiae. Some of the *pipal* leaves (Pl. *Ib*. II. 4–5) are also horned. Father Heras's suggestion that they are double leaves is hardly tenable

c9.s5.9

The wheel has four, six, or eight spokes, and sometimes shows what appears to be the axle hole (Pl. *Ib*. IV. 1). It is also found in combination, the most significant example of which is the wheeled vehicle (Pl. *Ib*. IV. 6). The bow has its arrow pointed upward, and it is often found in combination with a man. Another compound containing this symbol is Pl. *IIb*. VII. 6. To the water-sign is significantly attached a horn (Pl. *Ib*. IV. 15). The shrimping net, besides having strokes from 1 to 4, is found in combination with a man (Pl. *Ia*. III. 11). The hill-sign has significantly two or three peaks, or two or three internal strokes. The plant has 1–5 strokes, besides the arch-like sign; in Pl. *Ib*. VIII. 6, 7 there are ten strokes and in 8 we have eighteen.

c9.s5.10

The same system of compounds and stroke addition is followed in the case of ovals and their parts; but in quadrangles, triangles, angles, and crosses, the addition of strokes from 1 to 6 is most common. The only compounds noticed are with the cross, taurine, wheel, or sun within rhombus (Pl. II*b*. II. 8–9, 15), or double tree attached to rhombus (Pl. II*b*. III. 6), or as in Pl. II*b*. III. 2 and VII. 6–7. The arch-like sign is flattened in Pl. II*a*. VI. 9–12 and VIII. 1. The four-armed cross has ovals or crosses at the ends (Pl. II*a*. VIII. 2, 15). It sometimes develops into a six-armed symbol, Pl. II*a*. VIII. 3.

c9.s5.11

A fuller idea about the strokes can be obtained from Pl. II*b*, group F, where we see them ranging from 1 to 12. In one case (Pl. II*b*. IX. 12) we can count 15. Whether they all signify numerals cannot be dogmatically asserted, but some of them obviously suggest it. In the ninth row over Ts we have 2, 4, 5, 6, 9, 10, 11, 12, and 15 strokes. In the tenth row we have only 4 in the first group; 3 and 4 in the second group; and 4, 5, 6, 7, and 8 in the third group. The long strokes in the line occur in groups of 1, 2, 3, 4, and 5. The smaller strokes show 1, 2, 3, 4, 5, 6, 7, 8, 9, and 12. It is difficult to say whether these are different systems of enumeration, or whether some represent tens, hundreds, &c.

c9.s5.12

*Conclusion*

c9.s5.13

If we leave aside the additional strokes and compounds, the actual number of signs in the Indus Script, according to the present reckoning, is 27 objects and 27 geometrical forms, besides numerals, brackets, and two magic emblems. The number is increased to 474 (the remainder 63 being numerals, &c.) by the following two principles:
(i) by combination with other signs;
(ii) by the addition of strokes (*a*) internal, (*b*) external, from 1 to 6, besides the arch-like sign.

c9.s5.14

The combinations are always with some recognizable objects, and strokes are found added to all but the homo-, snake-, wheel-, vehicle-, and bow-and-arrow-signs.

c9.s5.15

Can we draw any conclusion regarding their significance? Suggestions can be made, and have already been made, but nothing definite can be asserted in the present state of the materials. One thing is certain, that the Indus Script, in its ultimate analysis, is not very complicated. Probably the two fundamental principles of combinations and stroke-additions, if they survived, influenced the formation of the conjuncts and the open syllables in the later historical Indian scripts.

c9.s5.16

Barua, B. M., ‘Indus Script and Tantric Code’, *Indo-Iranica*, Calcutta, vol. i, 1946. Diringer, D., *The Alphabet*, London, 1947. Gadd, C. J., ‘Seals of Ancient Indian Style found at Ur’, *Proceedings of the British Academy*, London, xviii, 1932. Heras, H., *Studies in Proto-Indo-Mediterranean Culture*, Bombay, 1953. Hevesy, G de, ‘Sur une ecriture océanique paraissant d’origine néolithique’, *Bulletin de la Société préhistorique française*, Paris, xxx, 1933. Hrozny, B., *Die älteste Völkerwanderung und die proto-indische Zivilisation*, Prague, 1939. —— ‘Inschriften und Kultur der Proto-Inder von Mohendscho-Daro und Harappa. Ein Entzifferungsversuch’, *Archiv Orientalni*, Prague, xii, 1941. —— *Les Inscriptions proto-indiennes. Essai de dechiffrement, I.* Monograph, vol. xiv, Prague. —— *Ancient History of Western Asia, India and Crete*, Prague, 1943. Hunter G. R., *The Script of Harappa and Mohenjodaro and its Connection with other Scripts*, London, 1934. Mackay, E. J. H., *Further Excavations at Mohenjodaro*, Delhi, 1937–8. Marshall, J., *Mohenjodaro and the Indus Civilisation*, London, 1931 (containing contributions from C. J. Gadd, Sydney Smith, and Professor Langdon). Meriggi, P., ‘Zur Indus Schrift’, *Zeitschrift der Deutschen Morgenländischen Gesellschaft*, Leipzig, xii, 1934. Petrie, W. F., ‘Mohenjodaro’, *Ancient Egypt*, 1932. Pran Nath, ‘The Script of the Indus Valley’, *J.R.A.S.*, London, 1931. —— ‘The Scripts of the Indus Valley Seals’, *I.H.Q.*, Calcutta, 1931. Quintana Vives, J., *Apportaciones a la interpretacion de la escritura protoindica*, Madrid–Barcelona, 1946. Ross, Allan S. C.: ‘The Numeral-signs of the Mohenjodaro Script’, *M.A.S.I.*, no. 57. Sankarananda, Swami, *The Rigvedic Culture of the Prehistoric Indus*, Calcutta, 1943–4. Vats, M. S., *Excavations at Harappa*, Calcutta, 1940. Waddell, L. A., *The Indo-Sumerian Seals Deciphered*, London, 1925.