Evolution of languages in India

Click the image to zoom-in

It was a lazy Saturday-afternoon discussion over coffee with friends, discussing life, its struggles et al, when one of us mentioned about how government documents, especially forms to be filled for various compliances were written in a complicated language, difficult to interpret even for the well educated. Someone remarked that this tendency was probably a remnant of Mughal times when the royal court used Urdu, while the people spoke Hindi in the streets. That one sentence was wrong at so many levels that I felt compelled to intervene to tell him that Urdu and Hindi hadn't even been birthed when the Mughals ruled; nevertheless, the Mughal court operated in Persian, which too was not the language of the masses.

He quickly Googled my claim for accuracy, and quietly agreed with me; and then popped up a curious question: Why is it that the Mughal court transacted its business in Persian when the Mughals belonged to Mongolia, and came to India via Turkmenistan - thus having no direct lineage to Persia or the Persian kingdom? That question left me also dumbfounded - his logic was quite incisive, the Mughals indeed had no direct connection to Persia. That sent me into a long deep dive into the evolution of languages in India through the ages, going back to 2000 BCE, all the way till the present. My discovery was fascinating and revealed how language use is interrelated with not just geopolitical but more so social evolution of the Indian society.

The first thing I discovered was that the question itself contained a small historical wrinkle. The Mughals were not simply 'from Mongolia'. Babur, the founder of the dynasty, was a Central Asian Timurid prince: descended from Timur on his father’s side and from Genghis Khan’s Mongol Chagatai line on his mother’s. He belonged to a Turko-Persian cultural world where Persian was already a language of prestige, literature and administration. The Mughals weren't Persian by ancestry but chose Persian as the language, because that was the language which held most prestige in the world they knew.

And that, as it turns out, is the whole story. Languages do not travel only with rulers. They travel with soldiers, scribes, religious institutions, schools, courts, trade routes and the promise of a job. Sometimes a language becomes powerful because it is the language of prayer; sometimes because it is the language of poetry; sometimes because it is the language in which the tax collector writes down what you owe.

The evolution of language that I came across is unlike a relay race with one language handing the baton to another but rather more like a crowded Indian train station: several languages arriving, departing, exchanging travelers (speakers), and occasionally even transferring entire coaches from one to the other.

Before language: a subcontinent of scripts, sounds and dialects

When we talk about the evolution of languages in India, we often begin with Sanskrit. That is understandable: Sanskrit has an extraordinary literary and religious history, and its influence on many languages is unmistakable. But it is not the beginning of the whole story. The subcontinent has always been home to multiple language families, and Sanskrit is one important strand in a much larger linguistic fabric.

Major language families in the Indian subcontinent include [refer image above] Indo-Aryan, Dravidian, Austroasiatic and Tibeto-Burman, alongside smaller families and languages whose relationships remain uncertain.

Indo-Aryan languages include Hindi, Bengali, Marathi, Punjabi, Gujarati, Assamese and many others.
Dravidian family includes Tamil, Telugu, Kannada and Malayalam, as well as languages such as Gondi, Kurukh, Malto and Brahui.
Munda languages, part of the Austroasiatic family, are spoken in parts of central and eastern India;
Tibeto-Burman languages are prominent across the Himalayan and northeastern regions.

All these families existed in some form or other, even before the formalization of the two proto-languages: Sanskrit and Tamil. In fact, Sanskrit and Tamil themselves got formalized by litterateurs and scholars from a proto-Indo-Aryan and a proto-Dravidian dialects which existed far before any of the Indian languages emerged. A review by linguist Sonal Kulkarni-Joshi describes the diversity of languages and dialects and the long history of interaction among these families.

Notably, the dates and routes behind these families are not all equally certain. Linguists reconstruct earlier languages by comparing patterns in words and grammar, but a reconstructed language is not the same thing as a language recorded in an inscription. We do not have a written record of Proto-Dravidian, for example. A Bayesian study of the Dravidian family estimated that the family may be about 4,500 years old, while also emphasizing that its geographical homeland and dispersal remain uncertain. The fact that the Baluchi people in North Western part of India (now Pakistan), speak Brahui which is a Dravidian language, geographically isolated from its cousins, with the nearest being over 1,500 kilometers away in South India, speaks volumes about the lack of proper knowledge about the geographical homeland of proto-Dravidian languages.

The same caution applies to the early history of Indo-Aryan languages. The prevailing view among historical linguists is that Indo-Aryan languages entered the subcontinent in the 2nd millennium BCE, as part of the wider family of Indo-European languages (which would include Germanic family of languages precursors to present day European languages).

But the details of movement, contact and language change are debated, and the evidence does not justify turning linguistic families into racial or cultural boxes. A language family tells us about the historical relationship between languages; it does not tell us that everyone who speaks one of those languages belongs to one ancestry or one community. Kulkarni-Joshi’s review is particularly useful here because it discusses both the evidence and the limits of what historical linguistics can establish.

There is another important point: languages did not evolve in isolation. The subcontinent is often described as a linguistic area because languages from different families shared features after centuries of contact. Similarities in sound systems, sentence structure and other grammatical patterns can arise through bilingualism and long-term interaction, not only through common ancestry. The languages of India have not merely occupied the same map; they have changed one another.

Prakrit, Apabhramsa and the languages people actually spoke

So if languages in India were a potpourri even in ancient times (just like they're today), why is Sanskrit often described as an ancient language? The reason for Sanskrit's stature is both linguistic and political - Sanskrit was probably one the first among languages to get codified into formal rules of grammar and diction. And this formalization was a probably due to its particular social and cultural role. Sanskrit was a language of learning, ritual, philosophy and literary prestige. But Sanskrit was not the everyday speech of everyone across the subcontinent.

Alongside Sanskrit were the Prakrits: a broad group of Middle Indo-Aryan languages and speech forms. The word Prakrit is often explained to contrast with Sanskrit but the historical relationship between Sanskrit and Prakrit is more complicated than a simple “formal language versus its corrupted version.” Modern scholars commonly believe Prakrits as vernacular languages, regionally varied and the relationship between them and Sanskrit was understood differently by different grammatical traditions.

This distinction between a prestigious literary language and everyday speech is not unique to India. Western readers, can appreciate Prakrit vs Sanskrit by comparing Prakrits with Cockney or Highland English while Sanskrit to Standard British English. In India however, the distinction became especially visible in religious and political life. Buddhist traditions used Pali and other Middle Indo-Aryan forms; Jain traditions used languages such as Ardhamagadhi. These choices mattered. They made it possible to communicate religious ideas in languages beyond the most formal Sanskritic register, even as Sanskrit continued to hold immense prestige.

The inscriptions of Ashoka offer a striking example of how a ruler could address a multilingual population. Most of Ashoka’s inscriptions, from the 3rd century BCE, were written in dialects of Prakrit, with regional variation across the empire. The OpenEdition study of Ashokan inscriptions describes the use of different Prakrit dialects and early Brahmi script. This is not to mean that Ashoka pioneered the modern policy of linguistic secularism (although he may have!) Rather, it is testimony to the fact that even an emperor of Ashok's stature could not impose one language everywhere, and assume that his 'court language' would be equally intelligible to all his subjects; he too had to use a combination to reach masses over whom he governed.

Over time, the Middle Indo-Aryan languages changed further. Linguists use terms such as Apabhraṃśa for later stages and literary forms that connect Middle Indo-Aryan with the newer Indo-Aryan languages. By the period immediately before and around the Ghaznavid conquests of north India, Apabhraṃśa was extremely important in northern Indian literary culture. It wasn't one single language in the modern sense. "Apabhraṃśa" is a scholarly term referring to a group of late Middle Indo-Aryan literary forms that evolved from Prakrits and ultimately contributed to stratification in languages such as early Hindi/Rajasthani/Gujarati/Punjabi and related languages.

A scholarly history from Aligarh Muslim University describes 'Western Apabhraṃśa' as an important literary language of the Rajput courts of northern India immediately before the Turkic conquest. Thus, by around the 1st millennium CE, the ancestors of many present-day Indo-Aryan languages were taking recognizable shape. Bengali, Gujarati, Punjabi and other languages developed their own literary traditions over time. Marathi had already emerged a little earlier through a long history of regional speech and writing, not as a sudden offshoot of a single courtly decision - in fact, the development period of Marathi was contemporaneous with rapid regime changes and skirmishes with bordering kingdoms.

While Indo-Aryan languages were getting developed, inter-mixed and refined, Dravidian languages too were taking shape in the same era. Tamil, Kannada, Telugu and Malayalam followed their own evolution, with deep literary traditions of their own. Edicts and inscriptions going back to 1 to 10th century CE illustrate that Dravidian languages were not simply vernaculars but were written languages of political authority and administration at different times and in different regional settings. The earliest edicts are in Tamil in 1st CE, with Kannada and Telugu edicts dated around 450-500CE.

Notably however, Ashoka's rock edicts from 3rd century BCE mentioning the Cholas, Pandyas, Satyaputras and Keralaputras were written in Prakrit using Brahmi, not in Tamil or another Dravidian language; indicating that Indo-Aryan languages would have been understood in peninsular India even 2000 years ago.

Oncoming of Persian, and birth of Hindi/Urdu

Now to the question that started my deep dive: when did Persian enter the Indian lexicon? Persian most likely entered India via the Arab conquests of Sindh, but that would not have been a reason enough for its proliferation throughout northern India. Persian was already a language of administration and high culture across a vast region stretching through Iran, Central Asia and parts of South Asia. It had a long bureaucratic tradition, a rich literary canon and a network of trained scribes. The rulers who came from Central Asia did not have to be ethnically Persian to find Persian useful. They inherited a political world in which Persian was a ready-made language of governance. This was also the period when India's trade was at its peak, and Central Asia was not just a trade partner in itself, but also the route through which trade to trading partners further West, all the way to Rome, was carried out.

That explains why the Delhi Sultanate established Persian as a language of administration far before the Mughals arrived. The Mughals built on that inheritance. Under Akbar, Persian was established as the official language of administration across the empire in 1582, according to the Oxford Handbook of the Mughal World. Persian was not merely a decorative court language. It was used in revenue administration, official correspondence, chronicles and legal proceedings.

This choice also created a professional class. If you wanted a position in the bureaucracy, Persian literacy could be an advantage. Hindu trading communities, including Kayasthas, Khatris and Brahmins, learned Persian for trade initially but later transitioned to scribal roles having found their knowledge of Persian to be a gateway to being employed in the administrative system. The Oxford Handbook notes that Persian became an “admission ticket” to the Mughal revenue bureaucracy, and that many indigenous elites learned it for government employment. In other words, Persian was not simply imposed by foreign rulers on a passive population. It became embedded in local institutions and careers.

That is one reason the idea that Persian was “the language of the Mughals” can be misleading. It was the language of the state, diplomacy and trade in the 'region', but it was also a language learned and used by people across religious and regional lines.

The Mughal empire itself was multilingual. Persian dominated at the imperial level, but regional languages continued to be used in local administration, literature, devotional writing and everyday life. Village records, where they existed, could be kept in regional languages even while higher administration operated in Persian. In fact, the Mughals also patronized Sanskrit and vernacular literatures. Sanskrit scholars worked at court; Hindu epics were translated into Persian; Braj poetry received patronage; and regional languages such as Bengali, Marathi, Gujarati, Telugu, Kannada and Dakhni flourished in different settings. The Mughal court was not a language vacuum in which Persian erased everything else. It was one powerful centre in a multilingual world.

This is where the original coffee-table remark gets the history backwards. The Mughal court definitely did not use Urdu while the people spoke Hindi. The Mughal court used Persian for much of its official business. Urdu, in its present-day form, developed over time from Indo-Aryan speech in and around Delhi, shaped by contact with Persian, Arabic and other languages. It was not the language of the Mughal court but rather very much the reverse - the language of the masses.

Urdu and Hindi: related speech, different literary directions

Urdu did not arrive in India with the Mughals, nor was it invented by them. Urdu developed from local Indo-Aryan speech, especially the Delhi region’s Khari Boli, in a multilingual environment shaped by Persian, Arabic, Turkish and regional languages. The Encyclopaedia Iranica account of Urdu traces Urdu to the regional speech of Delhi and describes the influence of Persian after the Muslim conquests.

The language had several names over time: Hindavi, Hindui, Hindustani, Rekhta and others. The names were not always used consistently, and they did not map neatly onto today’s categories of Hindi and Urdu. The word “Urdu” itself is associated with zaban-e urdu-e mualla, often translated as the language of the exalted camp - this could mean the court or royalty, but given its origins and antecedents, it most likely was meant to denote high-culture: a language in which poetry and prose of high-quality was written. This explanation is more likely because Urdu’s development was not a single event in a royal court but rather a gradual process involving local speech, migration, literary patronage and contact across communities.

Urdu’s literary history grew in multiple regions. Dakhni, a southern form of Urdu, developed in the Deccan and was used in poetry and religious writing. Persian remained the language of administration in many courts, even as Dakhni and other vernaculars gained literary status. Later, poets and writers from the Deccan moved north, and the language associated with Delhi and Lucknow developed into a major literary medium.

Hindi and Urdu are often treated today as two entirely separate languages, but at the level of everyday speech they share a great deal. Their grammatical base is the same, and ordinary spoken forms are often mutually intelligible. The differences become more pronounced in formal vocabulary, literary traditions and scripts: Hindi is generally written in Devanagari and tends to draw more heavily on Sanskrit vocabulary in formal registers; Urdu is written in a Perso-Arabic script, usually Nastaliq, and draws more heavily on Persian and Arabic in formal registers. Northwestern University’s open textbook on Urdu explains that the similarities between Hindi and Urdu are often greater than their differences, especially in everyday speech.

The British period and gazettization of language

India is a living testament to diversity and coagulated co-existence. While the slogan "Unity in Diversity" may have been coined by the makers of modern India, it is simply a reflection of India through its centuries old unwritten tradition. This is as much or rather even more true of language than any other aspect of culture. Just as the Mughals chose Persian, not because it was 'their own language', every region had a distinct language being used in courts and in the bazaar - often the language of choice having no bearing on who was the ruling satrap locally.

Marathi, Telugu, Tulu flourished alongside Persian in the Bahmani kingdoms of Deccan, while Marathi was equally powerful in Maratha ruled feudatories of Central India, co-existing of course with Hindi/Urdu, and even Gujarati and Nimadi. Similarly, the Travancore kingdom transacted official business in Malyalam, with Tamil being used in several parts. And while kingdoms of Awadh and the Sikh empire used Persian as their official languages, the masses spoke Awadhi and Punjabi / Haryanvi in these regions.

With this level of diversity already prevalent by the time the British marched into India, they had no choice but to inherit the Mughal language system. Persian thus remained important in administration under the East India Company for a time. And even though the colonial state eventually replaced Persian in 1837, they did (and could not) replace it instantly with English. The East India Company ended Persian’s official role in judicial and revenue proceedings and allowed other languages and scripts to be used. In northern India, Urdu written in the Persian script became an important court language in many areas, while English rose in higher administration and education. The Allahabad High Court’s historical account describes the legal change and the resulting use of Urdu in Persian script in the North-Western Provinces.

This was not a simple switch from Persian to “the language of the people.” The colonial state needed languages that could be standardized, taught, documented and used in courts. It also needed a bureaucracy that could translate between local administration and the language of the ruling power. English became increasingly important at the top; vernacular languages were used in other administrative settings; and Urdu gained official standing in parts of northern India.

It is here that, Hindi and Urdu started getting more sharply distinguished in public life. The distinction was not only linguistic. It involved script, education, employment, literary identity and political representation. Advocates of Hindi promoted Devanagari and Sanskrit-derived vocabulary; Urdu supporters defended the Perso-Arabic script and a Persianate literary tradition. The two varieties had long shared a spoken base, but colonial institutions helped turn differences of script and vocabulary into markers of community and political claim.

This is not to say that the British separated Hindi and Urdu purposely or by diktat. The languages and literary traditions had independent, albeit parallel histories, even before colonial rule. But colonial administration, print culture, education, census categories and gazette notifications, helped harden distinctions that had previously been more fluid. The language question became entangled with the question of who would have access to government jobs, whose script would be recognized in courts, and whose literary tradition would be treated as the standard.

English, independence and the continuing argument over language

The British period ultimately brought English into the centre of Indian public life. English became the language of higher administration, law, education and elite employment. It did not replace Indian languages in everyday life, but it created a new hierarchy: English at the top of many institutions, regional languages in local and cultural life, and Hindi and Urdu in a complicated relationship across northern India.

After independence, India faced a difficult question: could a country with so many languages choose one language for the Union without alienating large parts of its population? The Constitution named Hindi in Devanagari as the official language of the Union, while providing for the continued use of English for official purposes during a transition period. Pakistan on the other hand mirrored what India did except that it replaced Urdu for Hindi. Two languages which even today share so much in common have eventually been separated out by nationalities. However, the continued patronage Urdu receives in India is testimony that political boundaries cannot replace the place languages have in people's hearts. In fact, recent reports indicate that even Sanskrit is finding patronage in Pakistan.

In the present-day, the Indian Constitution’s Eighth Schedule recognizes 22 languages, including Assamese, Bengali, Gujarati, Hindi, Kannada, Kashmiri, Konkani, Malayalam, Manipuri, Marathi, Nepali, Odia, Punjabi, Sanskrit, Sindhi, Tamil, Telugu, Urdu, Bodo, Santhali, Maithili and Dogri. The Ministry of Home Affairs’ list of scheduled languages makes clear that India’s constitutional language landscape is not a one-language arrangement.

The states, too, have their own official languages. This is not an administrative inconvenience that India has failed to solve; it is a reflection of the country’s linguistic reality. The reorganization of states along linguistic lines, especially in the decades after independence, recognized that language is not merely a tool for communication. It is tied to education, literature, public services and a sense of belonging.

And English too remains important partly because it can function as a bridge language between people who do not share a mother tongue, and partly because it is deeply embedded in higher education, law, business and government. The expectation that English would cease to be used after few years of independence had proved politically impossible. English has continued, and the Official Languages Act provided for its continued use alongside Hindi. But its continued dominance also raises questions about access and inequality. A person who is fluent in English may find it easier to navigate institutions, while someone equally educated in a regional language may face barriers. The problem is not that English exists; it is that access to English can become a proxy for access to opportunity.

And this brings me back to the government compliance 'forms' that started the conversation. The difficulty of government language is not a Mughal inheritance in any simple sense. It is a problem of bureaucracy, legal drafting, institutional habit and the distance between the language of administration and the language people use in daily life. Persian once served as a language of administration because it was the language of a trained scribal system but also because it was the language of international commerce back then. English later occupied a similar position in colonial and postcolonial institutions, and has expanded to occupy that position now globally, not just regionally. And yet, official documents may be written in English, Hindi or a state language, but they can still be needlessly difficult to understand.

The lesson is not that one language is inherently more accessible than another. It is that institutions often use language to signal authority, expertise and status. A form written in dense legal English can be just as alienating as a Persian document would have been to someone who had never learned Persian. The language changes; the distance between the state and the citizen can remain.

Coda: What the language story tells us about India

The evolution of languages in the subcontinent is not a story of one civilization replacing another, and hence a language replacing another. Neither is it a story of 'official languages' trampling over local dialects. Rather, its testimony to India's culture that no ruler of India - foreign or local - has ever been able to impose any linguistic tradition on his/her subjects. And as a result, the story of languages in India is a story of how multiple layers have coexisted.

Sanskrit became a language of religious and intellectual prestige. Prakrits and other vernaculars carried everyday speech and religious teaching. Regional languages developed literary traditions and political importance. Persian became a language of administration and high culture under the Delhi Sultanate and the Mughals, and was learned by local elites who wanted to participate in government. Urdu and Hindi developed from related speech forms, then acquired increasingly distinct scripts and literary registers. English entered through colonial rule and remains influential because institutions continue to use it.

At every stage, language was connected to power - but not only to power. It was also connected to devotion, poetry, trade, education, migration and the ordinary need to speak to one another. A language could be the language of a court in one setting, a language of poetry in another, and a language learned by a local clerk hoping to secure a livelihood in a third.

The Mughals’ use of Persian is therefore not mysterious once we stop assuming that rulers must govern in the language of their ancestry. Empires often adopt languages that already have administrative reach and cultural prestige. The Mughals were Central Asian rulers who inherited a Persianate political world and made Persian their language of state. That choice was practical, but it also shaped the careers of Indian scribes, the culture of the court and the vocabulary of languages spoken across the subcontinent.

And Urdu? It was not the language of the Mughal court ever! It emerged over time from local speech and multilingual contact, and later became a major literary language in its own right. Hindi and Urdu are not simply two unrelated languages that happened to meet; they are closely related varieties whose modern distinctions reflect history, script, literary convention and politics as much as grammar.

The next time you hear someone say that a language is “foreign” because it was used by a ruler, please do ask a few questions. Was it foreign to the people who learned it and used it? Was it foreign to the scribes who built careers in it? Was it foreign to the poets who made it their own? And if a language has been spoken, written, adapted and loved for centuries, at what point does it stop being an outsider’s language?

The answer, I suspect, will not end up with a linear correspondence to the family tree of rulers. It will upend the very core of 'ruler decides the language' hypothesis, and speak about lives of the people who used the language, loved it, nurtured it and made it a part of their existence and of their progeny.

Search This Blog