DECIPHERMENT OF THE INDUS SCRIPT, A PERSONAL ACCOUNT
The decipherment of the Indus script is the most important advance in the study of ancient India after the discovery of the Harappan Civilization. This article provides a personal account of the breakthrough by the author who has been intimately involved in the effort. Jha and Rajaram have just completed the book The Deciphered Indus Script: Methodology, Readings, Interpretations, to be published in the second half of this year. A preview of the book is given later in this issue.
Indus script decipherment and its ramifications The decipherment of the famed Indus script by Natwar Jha, ably assisted by Birendra Kumar Jha, marks a watershed in the study of ancient history. While its immediate impact will be on the history of ancient India, before long it will be seen to have wider ramifications, affecting all of world history. In particular, it goes to show that the current model of history, which holds civilization to have begun in the river valleys of Mesopotamia in the late fourth millennium, is no longer tenable. In the specific context of India, it shows that a very ancient and largely indigenous civilization rooted in the Vedas existed at least a thousand years before the rise of the civilizations of Dynastic Egypt, Mesopotamia and the Indus Valley. (See the news item on Worlds oldest writing in this issue.)
A re-examination of the archaeology of Prehistoric Europe in the light of what has been called the 'Second Radiocarbon revolution' tells a similar story: there existed an earlier civilization in Europe also, before the rise of Egypt and Mesopotamia. It is known that Prehistoric Europe and Vedic India were connected by language, religious thought and mythology. The Celtic Druids of Western Europe traced their origins to Asia in the fourth millennium; ancient Indian chronicles tell us that the people known as Druhyus were forced to leave India in the fifth millennium due to a combination of ecological and political disruptions. There are other recorded movements also which may account for the similarities that we find between the languages and myths of India and pre-Christian Europe. (Note 1; numbers refer to notes at the end of this article.)
The Indus script decipherment may also provide a partial answer to the puzzle of the sophisticated intellectual achievements of Ancient Egypt and Vedic India. It may lend support to the claims of several modern scientists and historians of science that Ancient Egypt and Vedic India represent the culmination of an earlier civilization rather than the beginning. It is possible that the decipherment will shed light on this question by allowing fresh interpretations of Indian, and possibly, other ancient literature.
Eventually the decipherment will have to be viewed in the global context of this evolving, changing picture of the ancient world, and not in isolation as a technical solution to an academic problem. It has the potential to change the whole picture of the world and how man sees himself in it. While there are bound to be points of disagreement and debate on the details of the readings, it is clear that the writing of history will be profoundly affected by the decipherment.
While all these are issues of undoubted importance, the thrust of the present article lies elsewhere. Circumstances combined to place this writer at the center of events during the days and months following the decipherment, and place him also at the center of the controversy that was perhaps inevitable. It so happened that this writer was the first international worker to take note of Jha's decipherment, and independently read a large number of Indus seals based on it. He was also the first to make a systematic analysis of the decipherment by comparing it with other ancient scripts, and in the process bring it to the attention of both the public and the scholarly world through articles, lectures and presentations to the media. This has allowed him to observe events from a unique perspective.
One of the goals of the present article is to provide a first-hand account of the course of events from the time of his involvement in the decipherment to the present, as seen from this unusual vantage position. Such an account, it is felt, might be of some historical value considering the importance of Jha's breakthrough.
The present article therefore has a two-fold purpose. First, it presents a personal account of the decipherment and its immediate aftermath from the perspective of one who has been at the center of events as they unfolded; next, it presents also a brief summary of the decipherment itself including a few interesting readings. Before we get to the description of the script and the story of its decipherment, it is useful to have an understanding of the historical background to the civilization of which the Indus script is part. (Note 2)
Vedic Harappans: the historical backdrop
The famous Indus Valley Civilization, known also as the Harappan Civilization, was discovered by Indian and British archaeologists beginning about 1921. The first important sites found were Harappa and Mohenjo-Daro, now in Pakistan. Following their discovery, over a thousand sites have been found on both sides of the border. Among the more intriguing finds are more than 3000 seals unearthed by archaeologists. These famous 'Harappan seals' containing both pictures and writing have been objects of intensive study for over seventy years. The decipherment of the script and the language of the seals has been one of the great unsolved problems of our time.
Part of the problem has been the confused historical picture resulting from a surfeit of theories and conjectures. Prior to the discovery of Harappa and Mohenjo-Daro, scholars believed that civilization was brought to India by bands of nomadic invaders from Central Asia known as the Aryans. These invaders were said to have composed the Vedas after coming to India sometime after 1500 BC. This is the famous Aryan invasion theory that in one form or another is still found in history books. According to this theory, there was no civilization in India prior to the coming of the Aryans in 1500 BC.
The discovery of the Indus Valley Civilization (c. 3100 - 1900 BC) dramatically changed the picture. It showed the existence of a great civilization in India at least a thousand years before the supposed arrival of the Aryans. (On the basis of more recent finds like Mehrgarh it is now possible to push this date back to 7000 BC; Koldihwa in Central India dating to the same period may be regarded its southern counterpart.) To account for this changed scene, historians claimed that the inhabitants of these cities were Dravidians said to be a totally different linguistic and cultural group who had been displaced and driven south by the invading Aryans. While this allowed them to keep the invasion theory more or less intact, it added to the confusion in an already complex picture.
The writings on the seals show that the Harappan people were literate. Since the Aryan invasion theory held the inhabitants to have been Dravidians displaced by the invading Aryans, it was now claimed that the language of the seals had to be some form of ancient Dravidian unrelated to Sanskrit. But no one has been able to decipher the script and read the language of the seals under such an assumption. For over seventy years, the decipherment of the Indus script has resisted all efforts by scholars.
As noted at the beginning of the article, there has recently been a dramatic breakthrough. Dr. Natwar Jha, a 58 year-old (in 1996) Vedic scholar and paleographer from West Bengal has at last found the solution to the great problem. In a slim volume of sixty pages titled Vedic Glossary on Indus Seals, Natwar Jha, assisted by his son Dr. Birendra Kumar Jha, has provided both the key to the ancient script as well as a large number of readings. After a careful examination of his work, American Vedic scholar David Frawley and this writer found his decipherment to be substantially correct. This was announced at a public meeting in Bangalore on January 4, 1997.
Further study following that announcement has only strengthened our conviction about the script and its decipherment. We can now independently read a large number of seals using Jha's decipherment. What follows next is a brief account of the days and weeks following our announcement of Jha's decipherment. The account is necessarily a personal one; this writer was probably the first worker (other than N. Jha and B.K. Jha) to recognize the decipherment, and the first also to use it in producing a large number of independent readings and bring it to the attention of the world. What is given in the next section is an account from that vantage position; the following sections provide a brief description of the script and language including some readings. (Note 3)
The decipherment: a personal account
The October 29, 1996 issue of the Indian newspaper The Hindu carried a review of a small book titled Vedic Glossary on Indus Seals. The book, authored by Dr. N. Jha, principal of the Kendriya Vidyalaya in Farakka, West Bengal, was published by Ganga-Kaveri Publishing House of Varanasi. The reviewer spoke glowingly of Jha's originality and scholarship. The decipherment about which the reviewer was non-committal seemed to indicate that the language of the seals was Vedic Sanskrit.
Speaking from a personal point of view, the situation in October 1996 was as follows: I knew the language of the Indus script had to be Sanskrit of the late Vedic period, or some closely related language. A careful study of Harappan archaeology as well as a comparative analysis of the mathematics of Old-Babylonia (c. 1900 - 1750 BC) and the Egyptian Middle Kingdom (c. 2100 - 1850 BC) had convinced me that they both owed much to mathematical texts of the late Vedic period. These are famous as the Sulbasutras. It was also clear to me that mathematics of the Sulbasutras sometimes called Vedic mathematics must have been used by Harappan architects and town planners. (It is worth keeping this Sulbasutra connection in mind for it holds one of the keys to the decipherment.) (Note 4)
It was for these reasons that I was certain that the language of the Indus seals had to be related to Vedic Sanskrit. For the same reason, I was equally certain that the language of the Indus script could not be Proto Dravidian or any other pre-Vedic language that may or may not have existed. (As a native speaker of a so-called 'Dravidian' language, and a writer in that language, I have never believed in the Aryan-Dravidian linguistic divide. Nor do I believe that 'Proto Dravidian' has ever existed outside the imagination of linguists.) While many history books still hold on to the idea of an 'Aryan invasion' and suggest that the Vedas were the creation of these invading Aryans who destroyed the Harappan Civilization of the Dravidians, I was aware that archaeologists no longer accepted the idea of any such invasion. In fact I had written three books and numerous articles refuting various aspects of the Aryan invasion theory.
As far as the script itself was concerned I was on less sure ground. There were many theories but few certainties. When I first saw it, I was enthusiastic about the readings presented by the well-known archaeologist Dr. S.R. Rao, especially after I myself seemed to be able to obtain some plausible readings following his approach. But soon I ran into difficulties concerning his linguistic assumptions. I also found that far too many symbols in the script had been given artificial readings, and seemed not to agree with what I knew of the civilization and the language. It was my firm belief that the decipherment would show the language to be an archaic form of Sanskrit, probably of the late Vedic period. Rao's readings had assumed a pre-Vedic (theoretical) language, which my own research told me was impossible.
During my stay in Europe and America, I managed to acquire a good deal of useful material on ancient scripts and decipherment, especially from the British Museum. Studying them I began to see that most attempts aimed at deciphering the Indus script had been ad-hoc, and even based on wishful thinking. The Finnish scholar Asko Parpola for instance had assumed a particular solution (based on the Dravidian theory) and had then gone on to find justification for it in the seals. This can hardly be called decipherment.
As I began to study the Indus script in comparison with others, it became reasonably clear that the script was unlike any previously known. Contrary to widely held belief, the individual symbols and their possible sound values did not present the greatest challenge. Rao had seen that some of the Indus signs were related to signs in North Semitic scripts like Aramaic and Phoenician. (He seems not to have compared the Indus with the Himyeretic another West Asiatic script.) But for some reason he seemed not to have used the Indian Brahmi script in assigning phonetic values to the Indus signs. The Indian-American cryptographer Subhash Kak had noted the similarities between the Indus signs and the Brahmi letters, but had apparently not pursued the idea much further. I myself could identify several individual signs based on their similarity to Brahmi; there was hardly any doubt in my mind that the script was the ancestor of the Brahmi.
So the situation in October 1996 when The Hindu published its review of Jha's Vedic Glossary on Indus Seals can be summarized as follows. Several of us were reasonably sure that the language was Sanskrit. I had in fact said so in several publications including my book The Politics of History. At least some of us had a reasonable idea that the script was partly alphabetical, and also that its signs and sounds bore relationship to Brahmi as well as to some West Asiatic scripts. And yet, as we now know, none of us had the key to its decipherment. We all seemed to be missing something crucial.
It is helpful also to keep in mind the following fact: so many claims of decipherment have been made over the past sixty years that one has good reasons to be suspicious of any new claim. So the natural tendency upon seeing another claimed decipherment is to dismiss it. But the review of Jha's book in The Hindu sounded somehow different. For one it noted that Jha, in his reading of the seals had found connections with the Sulbasutras. This made me feel that there might be something to it, for I knew on the basis of my own research that the mathematics of the Sulbasutras and Harappan archaeology were linked. Jha's reference to the Sulbasutras suggested that his book might be worth a more serious look than what one ordinarily gives a new claim over the decipherment. I ordered the book from the publisher, which I received in about three weeks.
Upon opening the book the first thing that struck me was that it was not easy to read, and also not particularly well organized. The author claimed that the language was Vedic Sanskrit, which of course I already knew. He also noted that the writing itself was related to Brahmi which also I knew. At the same time, in spite of the somewhat awkward presentation a common feature of most pioneering works it was clear that the author was a Vedic scholar who had also mastered paleography. The book was highly technical and very much down to earth. It had none of the metaphysical speculation about the religious thoughts and the philosophy of the Harappans that one finds in so many works devoted to 'decipherment'. This was encouraging.
After a relatively short time spent reading the book I came to recognize the key idea in Jha's approach: the Indus script is not an alphabet in the conventional sense, but a proto-alphabet in which a single symbol was used to represent all the vowels. Moreover, this common vowel symbol a U-shaped character is used only as the first letter of words that begin in vowels; otherwise, the Indus writing is done without the use of vowels. The importance of this finding cannot be overestimated.
Once I grasped this key point not made explicit in Jha's book, but a crucial step in his decipherment everything seemed to fall in place. I immediately saw that Jha had the decipherment. The rest was a matter of familiarizing oneself with the writing and filling in the details a by no means simple task, but one that could be approached with confidence knowing that it was on solid ground. It was now possible to read literally hundreds of seals where previous 'decipherers' had struggled to read one or two and even that incorrectly. The readings were so consistent that there was no longer any room for doubt. (Note 5)
Note added: This is confirmed by the fact that I was able to decipher a newly discovered example from Harappa dating to 3500 BC based on Jhas methodology and my own extension of it. (See news item.)
After studying Jha's decipherment, and with the benefit of hindsight, it is possible to see where previous attempts had gone astray. The first difficulty was that many previous scholars Father Heras and Asko Parpola come to mind had rested their faith on an unproven linguistic conjecture. They had taken the Aryan invasion theory and the Aryan-Dravidian linguistic divide not as theories but as proven historical facts that would be reflected in the writings, which the Harappans had left behind. In effect, instead of analyzing the script and the language on their merits, they had sought support in the seals (and the writing) for their own beliefs.
Technically also one can see that using any form of Proto Dravidian or any other hypothetical language was doomed to failure. It is an impossible task trying to read an unknown script using a non-existent language.
Another weakness of many of these attempted decipherment was insufficient attention to paleography. A detailed comparative study of ancient scripts is needed to grasp the basis of Indus writing. In particular, a very careful comparison had to be made with both Indian and West Asiatic scripts in assigning sounds to symbols. Upon such a study, what one finds is that the Indus writing was not a fully developed alphabet, but more developed than its contemporaries in Mesopotamia, Egypt and the Levant.
An additional difficulty is that in deciphering the Indus script one runs against the prevailing scholarly opinion on the evolution of alphabetical writing. For example, most books claim the Indian Brahmi script to be an off-shoot of Aramaic in the first millennium BC; what we actually find is that the Indus script evolved into the Brahmi almost a thousand years earlier. In fact there are examples of writing in western India dating to about 1500 BC, displaying a transition stage from the Indus to the Brahmi. It also appears that some of the features that we associate with Aramaic and Phoenician were present in the Indus script a thousand years earlier. Thus a radical reorientation was needed before the Indus script could be deciphered. (Note 6)
Other errors such as Rao's assumption of a pre-Vedic language, and the belief that the writing was predominantly from right-to-left when it is in fact the other way were minor and would probably have been corrected over time. But trying to read the script while remaining bound to the Aryan-Dravidian divide was a hopeless task. It was doomed from the start.
All in all, the decipherer had to display unusual scholarship as well as unusual independence of mind before he could achieve the breakthrough. We can next examine how this led to the decipherment of the Indus script.
Steps towards decipherment
In his approach to the decipherment, Jha's first major step was to recognize the Aryan invasion and the Aryan-Dravidan divide to be modern theories that lack empirical support. In fact, most archaeologists today totally reject the notion of any invasion of India in ancient times. He also saw that the so-called Dravidian languages like Tamil and Kannada had never existed independently of Sanskrit. Linguists have constructed hypothetical 'Proto-Dravidian' languages, but these are entirely theoretical; there is no evidence that they existed at all unless we take the theory itself as the evidence. This break from old theories allowed Jha to approach the problem with a fresh mind and focus on the data. Without this break it is unlikely that Jha would have been able to decipher the script.
Once we discard these various theories as unfounded, we are led naturally to ancient Sanskrit of the Vedic period. Jha, an eminent Vedic scholar himself, decided to search for Vedic words on the seals. In this he was helped by an ancient work known as the Nighantu. It is a concise glossary of key Vedic words compiled by the ancient sage Yaska. Jha also found that the Shanti Parva of the ancient epic Mahabharata preserves an account of Yaska's search for still older 'buried' glossaries in compiling his own glossary. From this he concluded that some of the seals must contain words found in Yaska's Nighantu.
This gave him the clue to the language that it must be an archaic form of Sanskrit same as the language of Yaska's Nighantu. In order to identify the letters on the seals, Jha next compared the Indus writing with all the ancient scripts of India and the neighboring regions. He found that letters of most of the ancient scripts like Brahmi (of India), Aramaic, Phoenician, Himyeretic and others from West Asia were related to Indus signs. Since he knew the language, and also had Yaska's Nighantu as reference for the vocabulary, this comparative study allowed him to identify the symbols and read the seals.
This can be understood by comparing Jha's Indus script decipherment to Jean-Francois Champollion's decipherment of Egyptian pictorial script known as 'hieroglyphics' in the 1820s. Champollion was helped by a bilingual inscription written in Greek and Egyptian known as the 'Rosetta Stone'. In a like manner Jha has used Yaska's Nighantu as the key in his decipherment. Where his study of ancient scripts allowed him to identify individual letters of the alphabet, Nighantu provided him with the key to its vocabulary. The two together yielded the decipherment.
(This comparison should not be taken too far, for there is no need for a bilingual inscription for the Indus script since we already know the language to be Vedic Sanskrit. Where Yaska's Nighantu helps is in giving us a vocabulary so that we don't need to search far and wide for words appearing on the seals. Also, there are many words on the seals that are not found in Yaska's glossary. These can now be read since we have the decipherment. The basic point is that two sources were needed to solve for the two unknowns the script and the vocabulary.)
In understanding the decipherment and its application it is helpful to have some idea of ancient writing systems and how they evolved.
Ancient writing: from pictures to letters
The Indus script is one of the more complex writing systems devised by the human mind. It represents the transition stage from the complex pictorial-cum-phonetic writing characteristic of early civilizations to the highly scientific Brahmi alphabet, which is the source of all Indian and Southeast Asian writing. Jha has made a plausible case that even West Asiatic scripts like Aramaic, and Phoenician derive ultimately from the Indus.
The human species has evolved basically four different modes for expressing language in written form. These are pictograms, word-signs (or logograms), syllabic signs and the alphabet. This classification is not watertight, and it is possible for a complex writing system like the Indus to incorporate features of several of these. In fact, every alphabet must accommodate syllabic signs to represents sounds.
The first of these, the pictogram, need not be tied to any particular language. Pictures are used to represent objects, and conventions are adopted to express relationships. A circle may be used to represent the sun. The best known example of pictograms is probably the Maya glyphic writing of Central America. To understand how such a system works it is helpful to think of them as engineering drawings. A trained engineer will have no difficulty interpreting such a drawing like a blueprint or a circuit diagram even though it may not contain any words. Traffic signs also work on the same principle. (It is not necessary for us to understand how this idea was later adopted to express abstract concepts based on the so-called 'rebus principle'.)
A major advance took place when humans decided to extend the concept to expressing syllables or sounds. Now each pictorial sign represented not an object but a syllable in the language. The result is what is called a word-sign or ideogram. We can see the process in some of the Indus symbols. A bird is used to represent the sound 'shaka', from the Sanskrit word shakuni (bird). Similarly a pipul leaf (Sanskrit ashvattha) became the symbol for the sound 'shva'.
Traces of pictorial origin can sometimes be found even in the later alphabetical stage. In the Indus script, parallel wavy lines are used to represent the nasal letter 'na'. It is not hard to see that this symbol parallel wavy lines must at one time have been the pictorial symbol for the word 'nadi' which means river in Sanskrit. This later became a single wavy line for the letter 'na'. This symbol seems to have been borrowed by West Asiatic scripts like Phoenician and Aramaic.
The difference between the word-sign and symbols in a syllabic script is not always clear-cut, especially in a complex script like the Indus. Basically a syllabic script is one in which each symbol represents a particular sound syllable. This is the main feature of most ancient scripts from India to Greece. What distinguishes many ancient methods of writing from the Indus to the Mesopotamian Cuneiform to Egyptian hieroglyphs to Linear B of Mycenian Greece is the use of symbols to represents syllables. Some of them like the Egyptian, and also the Indus to a lesser degree, retained pictorial features; but the basic unit of writing was no longer the word but the syllable. Even pictorial signs came to be used for syllables rather than words.
The next stage in the development of the alphabet was the introduction of vowels. Here the paths taken by the Greeks and the Indians as they moved from syllabic writing to full-fledged alphabets diverged. The Indians, as witnessed in the Brahmi, indicated vowel values with the help of strokes attached to consonants; traces of this can be found in the Indus writing itself where a rudimentary stroke system makes its appearance. The Greeks (and the Romans) on the other hand indicated vowels with separate letters. They essentially borrowed the Phoenician letters, which includes syllabic signs that Greek does not need. The Greeks began to use these surplus letters as vowels.
These two paths gave rise to the two major systems of alphabetical writing in the world today the Graeco-Roman used in Europe and America, and the Indic used in India and Southeast Asia. The development of the alphabet is one of the major achievements of mankind one that took several thousand years. And despite fundamental differences, both the Graeco-Roman and the Indic methods have the same basic goal: denoting different ways in which a particular sound (called 'consonant') is to be pronounced. This is accomplished by the Graeco-Roman and the Indic scripts in different ways. We may next take a very brief look at the principles of these two methods.
All of us know that modern alphabets basically have two classes of letters called consonants and vowels. (We can ignore semi-vowels an artifact that most languages can do without.) Consonants represent the sounds or the phonetics, while vowels give shape to the sounds. In English, a vowel always follows the consonant, which is modified by the vowel. For example, 'put', 'pet' and 'pit' are pronounced differently because the vowels, 'u', 'e' and 'i' give different values to the consonant 'p'.
Indian scripts follow a different principle. The vowel additions to the consonants are indicated not by adding letters as in English, but by adding strokes to the consonants being modified. Different kinds of strokes are added to the consonant in question for denoting different vowel modifications.
This is how writing is done in modern alphabetical systems. But the Indus script, like most ancient scripts was syllabic and did not have vowels. (Note 7)
Syllabic writing in the Indus script
Both the Indic and the Graeco-Roman are fully developed alphabets with a full complement of vowels needed by the language. But this was not true of ancient writing. Pre-alphabetical (syllabic) systems were written without the help of vowels. This was true of the Indus script also, though its scribes had evolved a rudimentary way of writing vowels as we shall see later. As a result, there is considerable latitude in reading passages written in such scripts.
In syllabic writing, each word is represented by a string of consonants or syllabic signs. (I have oversimplified somewhat, for a syllable, strictly speaking consists of a phoneme and a vowel, but we need concern ourselves with these technicalities.) It is up to the reader to supply the necessary vowels. In English for example, using consonants only, we would write 'bk' for 'book', 'lmp' for 'lamp', 'txtbk' for 'textbook' and so forth.
Since syllabic writing lacks vowels it is often possible to come up with more than one 'correct' reading for the same string of symbols by superimposing different vowels. This is a feature not only of the ancient Indus script, but also the Linear B script used for the Mycaenean Greek a thousand years later, and even modern languages like Hebrew and Arabic. As a result, knowledge of the context has an important bearing on the interpretation of the signs. For illustration, the examples given below have at least two different 'correct' readings depending on the vowels we choose. This is how the Indus script has to be read. The choice of one over the other is governed by the context. (I have given the English transcription for the Sanskrit letters in question).
Letter string Possible readings
s-m saama, soma or suma
bh-r-t bharata or bhaarati
p-r-v purava or paurava
d-sh-h-m dashahema or dashahoma
The Indus script: a primordial mix
We saw earlier that scripts can be classified into pictorial, logo-graphic, syllabic and alphabetical. But the Indus script is a primordial script that defies classification. It is a mix of at least three of these four methods. It consists of pictorial symbols, phonetic symbols making up a syllabary, as well as a rudimentary form of alphabetical writing. And we find that characteristically Indian feature the composite letter. They combine to make the Indus script one of the most complex writing systems ever created.
In an article such as this, it is not possible to go into the details of Indus writing beyond what has been given so far. It is suffices to know that the Indus script, like most ancient scripts was syllabic. But what distinguishes it from other syllabic systems is that it used a U-shaped generic vowel symbol for words that began in a vowel. It also included a rudimentary stroke system for indicating vowel endings. (This feature is not discussed in the present article.)
This single vowel symbol later came to be expanded to eleven vowel symbols used in modern Indian scripts. The rudimentary stroke system also was perfected making the Indic phonetically the most scientific system in the world. We can therfore say that the Indus script was not as developed as modern scripts, but was more developed than its contemporaries.
To understand the importance of this generic vowel symbol we need to recognize that many languages of West Asia like Aramaic and Phoenician in ancient times, as well as Hebrew and Arabic today, never had a pressing need for vowels. (These are examples of what are called 'Semitic languages'.) This is because words in Semitic languages never began in a vowel. It is still the case that Arabic and Hebrew newspapers are printed without vowels. This apparently presents no problems to its readers.
Sanskrit however is different. It is full of words that begin in vowels words like agni, ishvara, uma, om and many more. To meet this need the Indus scribes had evolved an ingenious device. They indicated vowel beginnings with a U-shaped symbol. This symbol, however, is used only at the beginning of a word, and for all vowels; in the remainder of the word the reader has to fill in the necessary vowels as we already saw. With this innovation, it becomes possible to write words with vowel beginnings also. The word 'agni' is written Ugn. Similarly, 'ishvara' becomes Ushvr, 'indra' Undr, 'arkagni' Urkgn and so on. Someone familiar with the language experiences little difficulty in reading it, but the reader must know both the language and the literary context. (See examples of Indus writing.)
What this means is that the Indus alphabet is not a true alphabet but a hybrid proto-alphabet: it is a syllabic system enhanced with a generic vowel symbol for words beginning in a vowel.
Recognizing this fact that it is a syllabic script with a generic vowel symbol holds the key to Jha's decipherment. In addition, composite letters a striking feature of all Indian scripts appear also in the Indus. All in all Indus writing may be seen as an intermediate stage in the transition from a primitive syllabic system to a scientific phonetic alphabet like Brahmi from which nearly all Indian scripts are derived.
In his book Jha gives both a complete alphabet and a methodology that anyone can use, provided the user is familiar with the language and vocabulary of the Vedic literature. It is obviously beyond the scope of an article like the present one to go into the details of this extraordinarily rich script. What has been presented in this section is only a glimpse into the Harappan writing, describing some of its key features. Still, it is hoped that enough details have been provided to enable the reader to get an idea of the Indus script. The examples also should help the reader grasp some of the essentials of Indus writing.
For instance, the messages on two famous seals with the image of the bull, and the image of the horned deity or Pashupati can now be read. (See photographs.) The first simply reads 'Indrah' refering to the Vedic God Indra associated with the bull as the symbol of strength. The Pashupati seal reads 'Ishadyattah marah' which means: 'Marah (or evil adversaries) controlled by Isha'.
The real test
The objective test of any decipherment is consistency: it must be internally consistent, and consistent also with empirical data in the form of known language and literature. This point needs to be emphasized because there have been claims on the part of some critics that the decipherment of the Indus script must also have 'interlocking evidence'. By this is generally meant that it must not violate the scholarly opinion (and theories) about ancient India that have been current for more than a century.
This is not a valid objection. To begin with, there is not one but several theories about ancient India often contradicting one another. No decipherment can possibly 'interlock' with all of them, by being "all things to all men". Second, most of these theories including the most famous of them the Aryan invasion no less had been discredited by science even before the decipherment. What critics often have in mind when they demand 'interlocking evidence' is that the decipherment must support their view of Harappan as pre-Vedic and its language as Proto Dravidian. The decipherment (and other evidence) fails to do this. So the theory must go not a consistent reading of the seals, and definitely not science.
(The readings do 'interlock' rather strongly with texts from the Vedic literature as we now know. This in fact holds one of the keys to the decipherment: the language is Vedic Sanskrit, and the vocabulary is from the Vedic literature. The two go together.)
As far as the claim that Jha (and the present author) have 'imposed' Sanskritic readings on the seals goes, we would like to know if it is really so easy to impose an arbitrary language and a highly complex alphabetical system and come up with hundreds upon hundreds of consistent readings that 'interlock' with numerous texts from Vedic India from the Rigveda to the Sulbasutras. If it were so easy, why did scholars following the Dravidian theory from Heras to Parpola fail to come up with even a handful of meaningful readings by 'imposing' their favourite language and theory? In fact, they do not have a single reading to show for their years of effort. All they have are metaphysical speculations.
The fact of the matter is: it is much easier to decipher and correctly read inscriptions than come up with such incredibly clever fabrications. As Hardy said of Ramanujam's great but baffling mathematical discoveries: "They must be true because no one would have the imagination to invent them."
The same is true of decipherments. Fabricating the kind of readings that Jha's decipherment yields (and my readings support) hundreds of readings that are uniformly consistent calls for diabolical skills. The reader is invited to study the decipherment and judge the readings on their own merit. There will of course be errors and need for improvements, but the kind of 'imposition' being suggested by some is beyond human capacity. (See Note 8.)
Summary: Indus script as the 'missing link'
In summary, Jha's decipherment tells us that the language of the seals is Vedic Sanskrit, while the writing itself is proto-alphabetical, representing an intermediate stage in the evolution from a primitive consonantal (syllabic) system to the scientific alphabetical writing which is the unique achievement of the Indian civilization.
The Indus script may therefore be seen as the 'missing link' between primitive alphabet and scientific alphabet in the evolution of writing.