Wednesday, October 7, 2015

The Indus Valley Civilization: Was it a Bilingual Culture?

The Indus Seals

Nearly a hundred years after the Indus Valley civilization was first discovered in the 1920s, the language of the Indus seals remains shrouded in a veil of mystery. 

The Indus seals date from the earliest period of the Harappan civilization from c.3500 BCE. Most of them are an inch square – roughly as big as a postage stamp – and generally made of steatite (soapstone). Typically, a brief script was incised along the top of the seal. The rest of the seal was occupied by an image in relief, which generally depicted one or more animals or a yogi seated in a meditating posture.

Till date, nearly 4000 inscribed texts have been found, containing around 420 unique signs, of which 31 signs are used frequently. Since the unique symbols are in excess of 400, the Indus script is believed to be logo-syllabic i.e. some of the signs express an idea or a word, while the others represent a sound. The inscribed texts are very brief with an average of five signs. 
Fig 1: Indus seals depicting animal images in relief. Source: Wikimedia Commons.
Fig 2: Indus seal depicting a meditating yogi. Source: Wikimedia Commons
 Even though the inscrutable inscriptions on the Indus seals remain undeciphered, we do know for what purpose the seals were used.

The Mesopotamians, with whom the Indus cities had active trade, used seals for stamping sealings (i.e. seal impressions) on clay, resin or other materials that were attached to sacks of grains, jars or some other article of merchandise. 

The Indus seals must have served a similar function. Most of the Indus seals have a projecting boss at the back, to provide a finger-hold while pressing the seal into soft clay. Clay sealings with Indus signs have been found in Mesopotamia. Some of these sealings have impression of fabric on the back, which indicates that they were stuck to sacks of traded goods coming from the Indus Valley.

Evidently, the Indus traders branded their products using customized trademarks in much the same manner as modern-day manufacturers!

Historians believe that the markings on the seals must have been related to trading transactions and contain information about the traded goods (name, quality etc.) or the trading merchant (name, title, location, clan etc.).
Fig 3: A Mesopotamian limestone cylinder-seal, and a seal impression made on clay by rolling the seal. Source: Wikipedia
The Indus symbols are also found on objects such as pottery, bangles and even copper tools. A large signboard (3 metres in length) discovered at Dholavira in 1991 had ten large Indus signs made of baked gypsum. The signboard fit the northern gate of the citadel. According R.S.Bisht, who made the discovery, the inscription could stand for the name of the city, the king or the ruling family.

Recently, in 2012, a palm leaf manuscript was discovered at a Harappan site in Afghanistan, containing seven lines of text, which is the longest script recovered from any Harappan site. The text remains undeciphered.

Clearly, a large volume of information about this ancient culture remains inaccessible to us. One can only wonder how much more we could have learnt about the trade, art, literature, religious beliefs and other aspects of the civilization of the Indus people, if only the script could have been deciphered! Fortunately, some recent findings suggest that we may be on the verge of a breakthrough, and it may finally become possible to decipher the Indus script.

But before we get there, let us briefly review what is already known about the origins and evolution of the Indus script and the underlying language. 

A Proto-Brahmi Script

When we talk of the Indus script (or any other script for that matter), we need to remember that there are two variables involved – the script itself, and the underlying language.

A script is a set of symbols used for writing a language. A single script can be used for multiple languages. For instance, the Latin script is used for writing English, French, German, Dutch and many other languages. Conversely, a single language can be written in multiple scripts. The Sanskrit language, for example, has no native script of its own, and has been written using nearly all the major scripts of South Asia.

In case of the Indus script, archaeologists had noted from a very early period that it is very similar to the Brahmi script, which forms the basis of all the subsequent scripts of India and South Asia. One survey found 198 scripts that derive from Brahmi, making it the most influential writing system in the world.[1] 

When Sir John Marshall (the Director-General of the Archaeological Survey of India from 1902 to 1928) carried out the excavations in the 1920s that led to the discovery of Harappa and Mohenjo-Daro, he had compiled a big list of Brahmi characters that were derived from the Indus script. In the official accounts of the archaeological expeditions at Mohenjo-Daro he wrote that:

“In the table which follows I have drawn up a list of those signs, from which some of the Brahmi characters appear to be derived......obviously a large number of signs were discarded and only the necessary ones chosen...In all this widely spread epigraphical material, ranging from the upper to the lower parts of the Indus Valley, and to ancient Sumer and Elam, there is not a single text which differs in archaic style from the others...It represents a standardized and advanced stage even at this early period, and the existence of the accents reveals the astonishing care and knowledge of phonetic principles which would hardly be conceded to the scribes of this remote period. But the archaeological evidence of the mounds in the Indus Valley is said to admit of no doubt in this matter.”[2]

In 1934, G.R. Hunter analyzed the Indus scripts in his doctoral dissertation and concluded that it was a precursor of the Brahmi script. According to Hunter, scripts like Sabaean and Phoenician were derived from the Indus.

The dominant scholarly opinion in India is along the same lines. Dr. Suniti Kumar Chatterji, the National Professor in Humanities, Government of India, states,

“Some of the Mohenjo-Daro signs resembles or are almost identical with Brahmi letters.  Some others are a bit complicated.  What is most important, in some of Mohenjo-Daro signs, it would appear that the Brahmi characteristic of tagging on vowel signs to the consonant letters is also found, besides combinations of two or more consonants.”[3]
Fig 4: Some common variants of the Brahmi letters. Source: Wikipedia

Fig 5: A fragment of Ashoka’s 6th pillar edict. Source: Wikipedia

 But when and how did the Indus script transform into the Brahmi? 

The archaeological records indicate that the Indus Valley civilization began to collapse at around 1900 BCE, due to a combination of extreme climatic factors - earthquakes, massive floods and prolonged droughts.  The Harappan archaeologist S.R.Rao tells us that,

“In circa 1900 B.C., most of the mature Harappan sites were wiped out forcing the inhabitants to seek new lands for settlement. They seem to have left in a great hurry and in small groups, seeking shelter initially on the eastern flank of the Ghaggar and gradually moving towards the Yamuna. The refugees from Mohenjo-Daro and southern sites in Sind fled to Saurashtra (Gujarat) and later occupied the interior of the peninsula.”

The migrating people took the Indus script to different parts of India. A terracotta seal depicting Indus signs has been found in Vaisali, Bihar, in Eastern India, dating to c. 1100 BCE. Recently, an Indus Text with four signs engraved on a Neolithic polished stone celt (c. 2000-1000 BCE) was discovered in Tamil Nadu, in Southern India.[4] 

Even more interesting are the punch-marked coins issued by the Indo-Gangetic kingdoms at around 500 BCE, which have a large number of symbols that are identical to those found on the Indus seals.[5] 

It is not clear, though, at exactly what point the Indus script transformed into the Brahmi script. D.C.Sircar, India’s most eminent epigraphist of the twentieth century, observed in 1953 that, “The ancient [Indus] writing…may have ultimately developed into the Brahmi alphabet several centuries before the rise of the Mauryas in the latter half of the fourth century BC.”[6] 

In Northern India, the first inscriptions in the Brahmi script appear during the reign of the emperor Ashoka of the Maurya dynasty in the 3rd century BCE. Step by step the Ashokan Brahmi script evolved into the multiple scripts in use in Northern India today such as Devanagari, Sharada, Bengali-Assamese, Newari, Oriya etc. In Southern India, there is evidence of writing using the Tamil-Brahmi script as early as the 5th century BCE.[7] In the middle of the 7th century CE, it developed into the Pallava script which is the ultimate mother of the four major systems of writing in the South - Telugu, Kannada, Tamil including Grantha and Malayalam.

While most Indian scholars are in agreement that the Indus script was proto-Brahmi, there is much debate regarding the language being recorded using the script. Opinions are generally divided between Proto-Dravidian and Sanskrit, with many proposed decipherments. 

Recent findings, however, tilt the balance heavily in favor of Proto-Dravidian as being the language of the Indus seal inscriptions. 

A Proto-Dravidian Language

In the modern day, the Dravidian language is spoken by the people of Southern India, comprising of the four southern states – Karnataka, Kerala, Tamil Nadu and Andhra Pradesh.

There are a couple of interesting anomalies in the geographical distribution of Dravidian speakers. Approximately 2.2 million Brahui people in the province of Baluchistan in Pakistan also speak the Dravidian language. This group is isolated from the nearest Dravidian-speaking population by a distance of more than 1,500 kilometers! How did this large community of Dravidian speakers assemble here? Many scholars are of the opinion that the Dravidian language was widespread in the Indus Valley civilization, and Brahui is a remnant of it.
Fig 6: The Dravidian speaking regions of India and the Brahui speakers of the Indus Valley.  Source: Wikipedia
Another anomaly is the presence of numerous Proto-Dravidian speaking tribal groups throughout Northern and Central India. For example, more than two million Oraon and Kisan peoples of Orissa and surrounding states speak a Proto-Dravidian language called Kurukh. Malto is another Proto-Dravidian language spoken by the tribal people in these areas.

Interestingly, both Kurukh and Malto are very closely connected with the language of the Brahui people of Baluchistan. All of them belong to the Northern Dravidian language group. This agrees with S.R. Rao’s observation that the tribes from the Northern regions of the Indus Valley (such as Harappa) migrated eastwards towards the Yamuna River and the Gangetic plains.

In contrast, there are nearly 10 million Gond people in Central India who speak a Proto-Dravidian language that belongs to the Southern Dravidian language group. This would imply that some of the tribes from the southern Indus sites in Sind and Saurashtra (Gujarat) migrated towards Southern India as well as Central India.

The scattering of Proto-Dravidian speaking tribal groups throughout India suggests that the Dravidian language must have been spoken in the Indus Valley. There are many Dravidian place-names along the northwest coast of India, in Maharashtra, Goa, Gujarat, and to a lesser extent in Sindh, which supports this contention, and also provides clues regarding the migration paths from the Indus sites. 

Ancient Tamil literature also offers important insights. According to the Tholkappiyam (c.3rd century BCE), the earliest work of Tamil literature, the Velir chieftains, who were a royal house of minor dynastic kings in the early historic period of South India, belonged to the Yadu dynasty of Krishna, and they had migrated south from the city of Dwarka in Gujarat. 

Dr.Sircar, apart from pointing out the phonetical connections between the cities Mathura (the birthplace of Krishna) and Madura (the capital of the Pandyan kingdom), “refers to the temples of two gods at Madura, Kaveripattinam and other places mentioned in the Tamil classic Silappadikaram (2nd – 3rd century CE) one of whom has been described by the poet Kari-Kannan as the dark complexioned one bearing the wheel (Krishna), and the other as the fair complexioned one bearing a flag of the palmyra (Balarama).”[8] 

As a result of these migrations, the Proto-Dravidian language was once widely spoken across India. The Encyclopedia Brittanica states:

“Between 2000 and 1500 BC, there was a fairly constant movement of Dravidian speakers from the northwest to the southeast of India, and about 1500 BC three distinct dialect groups probably existed: Proto-North Dravidian, Proto-Central Dravidian, and Proto-South Dravidian”.[9]
The usage of the Proto-Dravidan language by the Indus people is further supported by the fact that nearly all the scholarly attempts to identify the language of the Indus seal inscriptions agree (with one or two exceptions) that it is structurally ancestral to Dravidian. It is quite possible, though, that the seal language (once it is fully deciphered) may be found to include loan words from other languages, including Sanskrit.[10] 

The most notable scholarly attempts to decipher the Indus script using a proto-Dravidian assumption are those of the Spanish Jesuit archaeologist Henry Heras in 1953, the Russian scholar Yuri Knorozov along with Dravidian specialist Nikita Gurov in 1965, and the Finnish scholar Asko Parpola who led a Finnish team in the 1960s – 80s. 

The Tamil scholar Iravatham Mahadevan has also published several papers proposing Dravidian readings for the Indus signs. In November 2014, Mahadevan published a detailed interpretation of a frequent sequence of four signs in the Indus Texts, primarily on the basis of the Dravidian language (Old Tamil), to derive the phrase “Merchant of the City”.[11] 

However, the most interesting discovery – one which holds the promise to reveal the language of the Indus script – came about very recently. 

The Gondi Connection

A team of scholars used root morphemes of the proto-Dravidian Gondi language to decipher a set of 19 pictographs carved in the Late Harappan style on a boulder close to Hampi in Southern India. The complete deciphered sentence, as published in The Hindu on November, 2014, read:

“On the goddess Kotamma temple woollen market way there is a rocky roof shelter for shepherds and sheep to stay at night up to morning.”[12]

The foundation for this discovery was laid a few years back, when Gondi expert Dr.Motiravan Kangale of RTM Nagpur University wrote a book (published in Hindi and Kannada) where he showed that the Indus script is very similar to the symbols used by the Gond tribals of Central India – the largest tribal group of India with over 10 million people. Dr.Kangale provided a number of decipherments of the Indus seal inscriptions using the root morphemes of the Gondi language. The deciphered texts typically specify the traded good or the identity of the trader – which is exactly what would be expected of an Indus seal.

His pioneering work was noticed by K.M.Metry, a Kannada University Professor of Tribal Studies. Prof. Metry had found a number of ancient rock painting containing symbols in the region around Hampi. He realized that the writings were of a Late Harappan style i.e. simplified Indus writing, as found on the potsherds at Bet-Dwaraka (in Saurashtra, dated to around 15th century BCE). Some of the symbols looked uncannily similar to the signs of the Gondi script. 

He invited Gondi experts Dr.Motiravan Kangale and Prakash Kalame to take a look at a specific rock painting containing 19 characters, carved on a boulder on top of a hill at Talwargahtta, on the way to the famous Vittala Temple at Hampi. The team of scholars managed to successfully decipher the pictographs using root morphemes of the Gondi language, and other cultural clues. 

Dr.Kangale mentioned that such drawings are found in Chhattisgarh and in interior structures of Gotuls (learning centres for youths) in the Bastar region. Prof. Metry told the Bangalore Mirror:

“There is a direct link between the Late Harappan and the writings that have been discovered. This proves that the people of Indus Valley Civilisation moved to South India after the collapse of the Late Harappan stage.”[13] 

Fig 7: Seal from Mohenjo-Daro interpreted using the Gondi language, on the basis of Dr.Motiravan Kangale’s work. 
Fig 8: The Hampi pictograph discovered by Prof.Metry and deciphered by Dr.Kangale. Source: The Hindu
The correlations between the Gondi script and the Indus / Late Harrapan style of writing, as well as the decipherments of the Indus seal inscriptions and the pictographs at Hampi using the root morphemes of the Gondi language, makes it abundantly clear that the underlying language of the Indus script must be Proto-Dravidian. 

It is quite incredible to realize that in the hilly, forested regions of central India, there still exists a language, practically frozen in time, defying the normal processes of linguistic change, which connects us back to the ancient Indus civilization! One can only guess what other mysteries of the past have been preserved within the art, customs, and mythic lore of this large tribal group.

The identification of the language of the Indus seals with Proto-Dravidian, however, raises an important question: If the Indus people spoke Proto-Dravidian, then who composed the voluminous amount of Vedic literature in Sanskrit? 

A Bilingual Culture

Modern Indian scholars such as Dr.N.S.Rajaram, Natwar Jha, Shrikant Talageri and others have argued that the Rig Vedic period had ended by c.4000 BCE. The bulk of the hymns in the four Vedas - Rig, Atharva, Yajur and Sama - had been composed by that time.

The primary scientific evidence in support of this hypothesis is the fact that the Vedas talk of the Saraswati River multiple times, describing it as a great river, “pure in her course from the mountains to the sea”. Satellite imageries of palaeochannels, along with geological, hydrological and archaeological field work, have shown that the Saraswati River of the Vedas flowed along the dried-up river bed of the Ghaggar-Hakra river system in northwestern India. During the great meltdown of the Himalayan Ice Age glaciers, some 12,000 years ago, the Saraswati was a wide, perennial river, following a southwesterly course from the Himalayas, to the Arabian Sea. 

Sometime around 4000 BCE, parts of the river started drying up, and it turned into a non-perennial river, possibly due to tectonic upheavals that disconnected it from the Himalayan glaciers at its source. Isotopic study of groundwater along the palaeochannels of the Saraswati River in Jaisalmer and Ganganagar districts of Rajasthan by Nair and Rao has indicated that the shallow groundwater was 2000 to 6000 years old, and the deep-seated groundwater was 6000 to 12000 years old. This indicates that the Saraswati River had started drying up and turned into a shallow, non-perennial river sometime between 5000 - 4000 BCE.[14] 

The Harappan civilization, which flourished on the banks of the Indus and Saraswati Rivers from c. 3500 – 1000 BCE, was, therefore, a post-Vedic culture. According to Dr.N.S.Rajaram, “the mature Harappan civilization was the last glow of the Vedic age.”[15] The Vedic underpinning of the Harappan society is amply demonstrated by the large number of Vedic fire altars (with layers of fine ash) that have been discovered at Indus Valley sites such as Kalibangan, Rakhigari, Banawali, Lothal etc., which indicates that Vedic fire rituals formed an integral part of the religious life of the people. 

Thus, on one hand, the Harappans orally transmitted the ancient Vedic texts (and composed many post-Vedic texts in Sanskrit such as the Upanishads, the Epics etc.), while on the other hand, the Indus seals reveal that they conducted trade using the Proto-Dravidian language. How do we resolve this apparent contradiction? 

The only logical solution is that the Indus people must have been bilingual. Perhaps, the Harappans used Sanskrit as the liturgical language, to compose their ritualistic and philosophical texts, and Proto-Dravidian as the vernacular language?

One may, of course, ask that if Sanskrit was the liturgical language of the Indus people, then why do we not find inscriptions in Sanskrit amongst the ruins of the Indus sites? The answer to that question is alarmingly simple: Sanskrit was never meant to be written down!  

Sanskrit was the Language of the Gods (daivi vak) used primarily for the performance of religious rituals and ceremonies. The correct pronunciation of Sanskrit was considered as extremely crucial for its religious efficacy, due to the ritual vibrations created by these sounds. It was believed that incorrect pronunciation may produce undesirable outcomes. As a result, Sanskrit texts were always transmitted orally. The knowledge of Sanskrit was confined to a small section of educated men, and the texts were never put down in writing, in order to guard against the normal processes of linguistic change and colloquialization that takes place in any language that is widely spoken and written. This is why Sanskrit never had any native script of its own.

Even in the 4th century BCE, when Panini composed his voluminous work on Classical Sanskrit grammar called Ashtadhyayi (“Eight Chapters”), containing 3959 rules of Sanskrit morphology, syntax and semantics, he did not take recourse to writing. It is believed that he composed the text with the help of a group of students, whose memories served him as “notepads”. The entire text was committed to memory and transmitted orally, as is typical in Vedic learning.

In the 3rd century BCE, when Emperor Ashoka put up inscriptions across the length and breadth of India (19 Pillar edicts, and 14 major Rock edicts discovered till now), not a single inscription was in the Sanskrit language. Ashoka used the Brahmi script and the Prakrit language (a Sanskrit derivative), and a few edicts were in the Greek and Aramaic languages.

The first inscriptions in Sanskrit - the Mathura inscriptions of the time of Sodasa - belong to the first quarter of the 1st century CE, and contain verses in classical Sanskrit.[16] This was long after the Harappan civilization had ceased to exist. It was only from the 4th century onwards, with the rise of the Guptas, that Sanskrit became the predominant language of Indian epigraphs. Even then, the sacred texts containing the hymns, philosophies and mythologies, were preserved orally, and were set down in writing quite reluctantly at a comparatively late date.

In view of this deeply-entrenched belief that Sanskrit was not to be written down, we cannot realistically expect any Indus seal to be in Sanskrit. The inscriptions on the Indus seals were primary for the purpose of trade and commerce. It would have been sacrilegious to use Sanskrit for such a purpose. It is possible, though, that the inscriptions contain some loan words from Sanskrit. That would be expected from a bilingual culture.

The idea that the Indus Valley civilization may be bilingual has not been seriously explored by modern researchers. However, a statement to that effect was made by Harappan archaeologist Dr. D.P. Sharma at the International Conference on Harappan Archaeology held in Chandigarh in 2012. While explaining his research work on the Indus signs on a palm leaf manuscript discovered from a Harappan site in Afghanistan, he stated that,

Harappans were using proto Dravidian and Sanskrit as their language and their script was proto Brahmi only.[17]

This tradition of using Sanskrit as the liturgical language for worship and rituals, and Dravidian as the vernacular has been followed in Southern India from the early historical period till the present day, thereby demonstrating a linguistic continuity with the Indus Valley. 

Sanskrit mantras are still an integral part of the daily worship and temple rituals in all Tamil Nadu temples, accompanied by music and dance in Tamil. Dr.R.Nagaswamy, the erstwhile Vice Chancellor of the SCSVMK University, Kanchipuram, wrote an article in The Hindu where he said:

“The Saiva and Vaishnavite saints who lived mostly between the 6th and 9th centuries handed down to us the then existing beliefs and customs. Saint Appar belonging to the Vellala agricultural class, sings again and again that God is the essence of the Vedas and the embodiment of mantra and tantra. He also says that the God is none other than the Tamil and Sanskrit languages. Similarly, the Alvars proclaim that God is Tamil and Sanskrit. No one today could claim he is a better devotee than these noble saints. After all, both the Saiva and Vaishnava saints were deeply rooted in the Sanskrit agamas but have sung excellent poems in Tamil. To the ancient saints there was absolutely no conflict between the two languages.”[18]

Sanskrit was used for inscriptions since the 5th century CE in Southern India, and the copper-plate charters of the Pallavas, the Cholas and the Pandyas are written in both Sanskrit and Tamil languages.[19] Sangam Tamil classics frequently refer to the Vedas, and hold the four Vedas in very high esteem. The ancient Tamil kings performed Vedic sacrifices, and the judicial and administrative system was based on Vedic injunctions.[20] 

The co-existence of Tamil and Sanskrit since the beginning of recorded history, as well as the respect accorded to the Vedas by the ancient Tamil kings, highlights the Vedic nature and bilingual tradition of the Dravidian society.

This also explains the mutual influence that the Sanskrit and Dravidian languages have exercised on each other. It is well known that all the Dravidian languages, including Old Tamil, have borrowed a significant number of words from Sanskrit. What is perhaps not so well known is that the Rig Veda has more than a dozen words borrowed from Dravidian.[21] 

The eminent Dravidian expert Bhadriraju Krishnamurti has pointed out that Vedic Sanskrit contains retroflex consonants, which are absent in other Indo-European languages. This indicates close contact of the Sanskrit speakers with speakers of a language family rich in retroflex consonants such as Dravidian.[22] In addition, a number of grammatical features of Vedic Sanskrit, not found in its sister Avestan language, appear to have been borrowed from Dravidian languages. These include the gerund, which has the same function as in Dravidian, and the quotative marker iti.[23] All of these linguistic borrowings are indicative of a very early Dravidian influence on Vedic Sanskrit. 

In Northern India, the Dravidian language must have been widely prevalent after the migrations from the Indus Valley. However, a linguistic revolution swept across much of Northern India sometime after the 6th century BCE, and a new type of language arose from Sanskrit known as Prakrit. The modern vernacular languages of Northern India evolved from the Prakrits. The Ashokan inscriptions of the 3rd century BCE were written in Prakrit (in the Brahmi script), which shows that the people were already well versed with the new language and the script.

The first variant of Prakrit is known as Ardhmagadhi which originated in Magadha i.e. Bihar in Eastern India - the seat of power in India for a long time. Various forms of Prakrits soon emerged in different regions, patronized by different royal dynasties. Sanskrit, however, continued to be learnt by the members of the upper castes for scholarly communications, and the language co-existed alongside the vernacular Prakrits. The Sanskrit dramas of this period were composed in both Sanskrit and Prakrit, and they show that Sanskrit speakers were almost always multilingual.[24] 

The Prakrits eventually replaced Dravidian as the vernacular language in Northern India. The tribal belts, however, remained insulated from this linguistic change, and continued to speak Proto-Dravidian and practice their ancestral customs. It is very interesting to note in this context that, the Gond kings of the medieval period could write poems in Sanskrit. The inscription at Mandla district written in Sanskrit gives genealogical account of the Raj Gonds.[25] This means that even the Gonds were bilingual at one point. While one may argue that the Gond kings could have acquired knowledge of Sanskrit after coming in contact with the Hindus, it is equally plausible that both Sanskrit and Proto-Dravidian were patronized by the Gond chieftains right from the ancient times (much like the emperors of the Southern Dravidian kingdoms.) 

The emergence of the Prakrits, however, had an unintended effect: it created a linguistic divide between the North and the South, which was later exploited by a host of colonial-era historians to propound various race-based theories. 

It is not clear, though, as to what political, social, or religious reasons fueled the emergence of the Prakrits after the 6th century BCE. 

One of the possible factors could be the formation of the new religions - Jainism and Buddhism. The Jain prophet Mahavira (599–527 BCE), and his immediate disciples known as Ganadharas, had given sermons in Ardhmagadhi and most of the ancient Jain literature was composed in Prakrit.

The oldest Buddhist scriptures (the texts of Theravada Buddhism) were written using Pali, a form of Prakrit. The Gandhara language, employed by the North Western branches of Buddhism, was a “Buddhist Hybrid Sanskrit”. The Gandhara and some of the other languages used by the Buddhists were likely to have been purely literary languages, without any spoken counterparts, which indicates that the Buddhists were actively inventing new Sanskrit-based languages for written communications.

Perhaps, the leaders of the new religions felt that, since Sanskrit could not be written down, it was necessary to devise a Sanskrit-based language to convey the sacred teaching of the prophets in a written form to the people? A Sanskrit-based language would have an aura of “sacredness”, and thereby be suited to the purpose (as opposed to a vernacular like Dravidian).  

It is possible that even the Brahmi script was perfected at this time (based on the corpus of Late Harappan symbols) for writing the new language. The Jain Agamas mention the Brahmi script in a list of 18 scripts, and Jain legends recount that the script was taught by their founder Rishabha to his daughter Brahmi, whence the name comes.[26] 

According to some scholars, Jain philosophy reached Southern India as early as the 6th century BCE. An inscription in Tamil-Brahmi, found in a chieftain’s grave in Palani (Tamil Nadu) dated to 490 BCE, contained the term “vay-y-ra” (meaning “diamond”). The term “vayra” is a Prakrit, derived from the Sanskrit “vajra” – a term commonly used in Jainism and Buddhism to denote the indestructible nature of spiritual power. There are many Jain caves scattered throughout Tamil Nadu which contain epigraphic records in the Tamil-Brahmi script from the 2nd century BCE onwards describing the livelihood of Tamil Jains (called Camanar from the Prakrit samaṇa “wandering renunciate”). The Jain and Buddhist saints were great scholar-poets as well, and composed some of the great literary texts of the Sangam Period (300 BCE – 300 CE) in the Tamil language.

Given the fact that the Jains composed their texts in the earliest form of Prakrit (Ardhmagadhi), claim to have invented the Brahmi script, and made early inroads into Southern India, we must consider the possibility that it was the Jains who played a pioneering role in the invention of the Prakrits and the finalization of the Brahmi script.

However, it must have been with the rapid spread of Buddhism throughout India that the Prakrit languages were adopted across the country. The geographical spread of Buddhism in India during the reign of Emperor Ashoka in the 3rd century BCE (as indicated by Ashoka’s rock and pillar edicts) shows a strong correlation to the distribution of Prakrit languages in India. 
Fig 9: The geographical limits of Buddhism in India during the reign of Emperor Ashoka in the 3rd century BCE strongly correlates to the geographical distribution of Prakrit languages in India.
Overall, it cannot be doubted that there has been a long-standing bilingual (or multilingual) tradition in India since the earliest historical times. Sanskrit was the language of ritual and worship in both Northern and Southern India, while Dravidian was the vernacular, till it was replaced by the Prakrits in Northern India after the 6th century BCE. Since the Vedic tribes migrated into India following the collapse of the Indus Valley civilization in c.1900 BCE (and once again after the end of the Late Harappan phase in c.1000 BCE), it implies that the Indus people, too, must have been bilingual, speaking both Sanskrit and Proto-Dravidian.

After all, the practice of using a sacred language for worship, and a separate vernacular language for all other purposes, was widely prevalent in the ancient world. In that respect, the Indus Valley civilization is no exception. 

Sacred Language and Bilingualism

In the book Sacred Languages and Sacred Texts, Prof. John Sawyer of the Newcastle University mentions that during the Assyrian Period (c.2600 – 605 BCE),

“Scenes depicting the writing down of lists of booty or the like, regularly show two scribes: one writing with a stylus on clay, the other with a brush on papyrus or vellum. The clay tablet, written in Assyrian, was for the imperial authorities, while the other document was an official Aramaic translation for the benefit of the subject peoples, including the people of Israel and Judah, who could not read Assyrian cuneiform…Bilingual texts containing an Aramaic version, are extremely common, many of them designed to be used in schools.”[27]

Aramaic was the vernacular language throughout the Assyrian and Persian Empires, until it was eventually superceded by Greek, after Alexander. Incidentally, some of the rock edicts of the Indian Emperor Ashoka were in Greek and Aramaic. 

The Achaemenids of the 6th century BCE spoke Old Persian (which was written using the Old Persian cuneiform script), but they used Aramaic extensively for administrative functions. Amongst the Jews, Hebrew was the language of scripture and worship. However, the Jewish religious leaders, at least from the time of Ezra (4th century BCE), were bilingual, using Aramaic or Greek in their contacts with other people, but still speaking Hebrew among themselves.[28] 

In the Roman Catholic Church, the sacred language of liturgy was Latin (Ecclesiastical Latin or Liturgical Latin), throughout the medieval period. Latin could be read and understood only by the clergy, while the vernacular language was different in different nations. Vernacular languages were never used liturgically in the Roman Catholic Church until 1964, when the first permissions were given for certain parts of the Roman Liturgy to be celebrated in certain approved vernacular translations. In Islam, Arabic is the sacred language for praying and reading the Qur’an. Many Muslims use Arabic inside the mosque but speak in one or more vernacular languages outside.

So, the use of a sacred language for prayers and rituals was widespread throughout the ancient world. In most cases, the vernacular language was different from the sacred language, and the members of the clergy, who knew the sacred language, were generally bilingual or multilingual. The sacred language was believed to be imbued with divine attributes, because of which the sacred texts were not permitted to be translated into other languages. Translation of the Qur’an is actually forbidden by Sunni law, and the Avesta, the sacred scripture of the Zoroastrians, was not even written down, let alone translated into the language of the people, for over 1000 years till the 6th century CE.

There is no reason why a similar situation could not have prevailed in the Indus Valley. The Indus people could have been bilingual, using Sanskrit as the liturgical language and Dravidian as the vernacular. The evidence certainly points in that direction. On one hand, the Vedic texts are in Sanskrit, while on the other, the structural analysis of the Indus inscriptions and the more recent decipherments using the root morphemes of the Proto-Dravidian Gondi language makes it almost certain that the language of trade and administration was Proto-Dravidian. 

A bilingual Vedic culture would explain why the Dravidians of Southern India, who had migrated to their present locations after the collapse of the Indus Valley civilization (as indicated by the archaeological and literary evidence), have used Sanskrit as a liturgical language and Dravidian as the vernacular from the earliest periods. It accounts for the very early influence of the Dravidian language on Vedic Sanskrit, (viz. Dravidian words in the Rig Veda and Dravidian grammatical features in Vedic Sanskrit) as well as the presence of a large number of Sanskrit words in all the Dravidian languages.

The reluctance to put down Sanskrit in a written form, in order to guard against the normal processes of linguistic change and colloquialization, may have prompted the emergence of the Sanskrit-based Prakrits after the 6th century BCE. Buddhist and Jain monks extensively used Prakrit languages to record the sermons of their prophets, and convey them to the people. 

The Sanskrit-based Prakrits eventually replaced Dravidian as the vernacular across much of Northern India, excluding the inaccessible tribal belts. However, the bilingual tradition continued unabated. The Sanskrit dramas of the classical period show that the Sanskrit speaking educated people of Northern India were almost always bilingual or multilingual. Even the Proto-Dravidian speaking tribal Gond kings had knowledge of Sanskrit, and were known to be have been bilingual during the medieval period.

There is no doubt that bilingualism was deeply embedded in the Indian tradition throughout recorded history, and it seems very likely that this was the legacy of a bilingual Indus Valley civilization.


[1] Thomas R. Trautmann, Languages and Nations: The Dravidian Proof in Colonial Madras (University of California Press, 2006) 64, taken from Wikipedia 
[2] John Marshall, Mohenjo-Daro and the Indus Civilization: Being an Official Account of Archaeological Excavations at Mohenjo-Daro Carried Out by the Government of India Between the Years 1922 and 1927 (Asian Educational Services, 1997) 431. 
[3] Dr. Suniti Kumar Chatterji, “Brahmi - The Mother of Indian Scripts”, Central Hindi Directorate Ministry of Education & Social Welfare 1977 <> 
[4] Iravatham Mahadevan, “A Note on the Muruku Sign of the Indus Script in light of the Mayiladuthurai Stone Axe Discovery”, 06 May 2006 <> 
[5] C.L. Fabri, “The punch-marked coins : a survival of the Indus civilization”, Journal of the Royal Asiatic Society, 1935, 311, taken from Michel Danino, The Lost River. 
[6] D.C.Sircar, “Inscriptions in Sanskritic and Dravidian Languages”, Ancient India, no.3, 1953, 215, taken from Michel Danino, The Lost River. 
[7] T.S Subramanian, “Palani excavation triggers fresh debate”,  The Hindu 29 Aug. 2011 <> 
[8] Manohar Laxman Varadpande, Krishna Theatre in India (Abhinav Publications, 1982) 8. 
[9] Dravidian languages. Encyclopædia Britannica <> 
[10] Frank Raymond Allchin, George Erdosy, “The archaeology of early historic South Asia: the emergence of cities and states”(Cambridge University Press, 1995). 
[11] Iravatham Mahadevan, “Dravidian Proof of the Indus Script via the Rig Veda: A Case Study”, BULLETIN OF THE INDUS RESEARCH CENTRE Vol.4 Nov. 2014. 
[12] S. Harpal Singh, "Gonds may have migrated from Indus Valley", The Hindu 17 Dec. 2014 <> 
[13] Shyam Prasad S, "Prof sees Harappan script in Hampi", Bangalore Mirror 04 Nov. 2014 <> 
[14] A.R.Nair, S.V. Navada and S.M.Rao, “Isotope study to investigate the origin and age of groundwater along palaeochannels in Jaisalmer and Ganganagar districts of Rajasthan” (1999) taken from Vedic Saraswati: Evolutionary History of a Lost-River in Northwest India (Geological Society of India) 315-319. 
[15] Dr. N.S. Rajaram, "The Harappan Civilization and Myth of Aryan "Invasion", Archaeologyonline <> 
[16] “Epigraphical Studies in India - Sanskrit and Dravidian”, Archaeological Survey of India <> 
[17] Swati Chandra, “Harappan people used an older form of Brahmi script: Expert”, TNN 2012 <> 
[18] Dr.R.Nagaswamy, “Why worship in Sanskrit” <> 
[19] “Epigraphical Studies in India - Sanskrit and Dravidian”, Archaeological Survey of India <> 
[20] Dr.R.Nagaswamy, “Asoka and the Tamil Country: The Evidence Of Archaeology” <> 
[21] Bhadriraju Krishnamurti, The Dravidian Languages (Cambridge University Press, 2003) 6. Taken from Wikipedia (Dravidian Language) 
[22]Ibid 36 
[23] Ibid 36-37 
[24] Madhav Deshpande, “Efforts to Vernacularize Sanskrit: Degree of Success and Failure”, 2011 in Joshua Fishman, Ofelia Garcia, Handbook of Language and Ethnic Identity: The Success-Failure Continuum in Language and Ethnic Identity Efforts (Oxford University Press) 218. 
[25] Shiv Kumar Tiwari, Tribal Roots of Hinduism (Sarup & Sons, 2002) 4 
[26] Acharya Nagrajji, Āgama Aura Tripiṭaka, Eka Anuśilana: Language and literature (New Delhi: Concept Publishing, 2003) 224. Taken from Wikipedia (Brahmi script) 
[27] John Sawyer, Sacred Languages and Sacred Texts (Routledge, 2012) 80. 
[28]Ibid 77-78.