In 2013, the last known speaker of Bo, one of the Andaman Island’s ancient languages, died. Her name was Boa Sr, and she was approximately 85 years old. With her death ended a language that had survived for an estimated 65,000 years – the longest continuous linguistic lineage on earth, a relic of the earliest human migrations out of Africa. No recording was made in time to preserve more than fragments. The People’s Linguistic Survey of India, completed in 2013, documented 780 languages spoken in India. Of these, UNESCO classifies 42 as endangered. But the number of genuinely endangered languages – those with fewer than 10,000 speakers and no institutional support – is considerably higher than that, and the pace of language shift toward Hindi, English, and dominant regional languages is accelerating.
What India’s Language Map Actually Looks Like
India is the world’s most linguistically diverse large country. The 2011 Census recorded 19,500 distinct “mother tongue” names, though the official consolidation process reduces this to around 1,369 “rationalised” mother tongues. The People’s Linguistic Survey of India, led by G.N. Devy, documented 780 distinct languages – defined as distinct linguistic systems, not dialects of a larger language. The difference between 19,500 and 780 reflects the definitional choices involved in distinguishing dialects from languages, but both numbers testify to a reality that India’s two-tier official classification (22 Scheduled Languages + the rest) fundamentally understates.
India loses approximately one language every two weeks, according to estimates by linguists working on documentation projects. This is not a precise figure – the data is genuinely uncertain, and the definition of “loss” depends on what threshold of abandonment you use. But the direction of the trend is not in dispute. Communities shift to dominant languages for economic opportunity, education access, and social mobility. Children grow up speaking the parents’ heritage language at home but conducting their economic and civic lives in Hindi, Telugu, Tamil, or English. Within two generations, the heritage language often has no speakers under 40.
Four Languages in Focus: Toda, Great Andamanese, Kodava, and Korku
Toda is spoken by the Toda people of the Nilgiri Hills in Tamil Nadu – one of the most studied tribal communities in anthropological literature due to colonial-era fascination with their social practices. The Toda language is a Dravidian language with features found in no other language family member; it has approximately 1,600 speakers. Toda has been documented extensively by linguists, has a small body of published material, and benefits from the high profile that Toda culture has had in academic literature. It also suffers from severe pressure: the Toda community is small, intermarries with other Nilgiri communities, and the language is not taught in any formal school system. Young Todas typically speak Tamil or Kannada for economic purposes and use Toda only within the community’s internal ceremonies.
Great Andamanese was once a family of distinct languages spoken across the Andaman Islands. Most are extinct. The few remaining speakers of what is now called “Great Andamanese” (a heavily mixed contact variety) live on Strait Island in the Andamans, and as of 2010, fewer than 50 people retained any knowledge of it. The language family is of extraordinary linguistic importance because the Andaman Islands were one of the earliest destinations of human migration out of Africa, and the linguistic ancestors of Great Andamanese may be the oldest surviving language lineage on earth. Its loss is irreversible and represents the erasure of cognitive and cultural knowledge that has no analogue anywhere else.
Kodava (or Coorgi) is spoken by the Kodava community of Kodagu district in Karnataka. With approximately 100,000 to 200,000 speakers, it is not endangered in the same acute sense as Toda or Great Andamanese, but it faces significant pressure from Kannada and English. The Kodava community is relatively affluent, well-educated, and widely dispersed nationally (due to high military service rates and urban migration). This success paradoxically threatens the language: successful Kodavas raise children in Bengaluru or Delhi who speak Kannada and English, not Kodava. The community has active literary and cultural organisations and a formal script was recently standardised, which represents a positive institutional development. Korku and Ho are tribal languages of central India (Korku in Madhya Pradesh and Maharashtra, Ho in Jharkhand) that have active digitisation efforts – smartphone apps, Unicode encoding, and community radio broadcasts – that represent promising models for language maintenance with limited institutional resources. See our analysis of community radio’s role in empowering rural India for how audio technology serves marginalised communities.
What We Lose When a Language Dies
The loss of a language is not merely a cultural loss – it is a cognitive and epistemic loss. Every human language encodes a distinct way of understanding the world: different classification systems for plants, animals, relationships, and time; different grammatical structures that shape how speakers conceptualise causation, agency, and identity; different bodies of knowledge – ecological, medical, navigational, social – that exist in oral form within the language and nowhere else. When a language dies without documentation, all of that knowledge dies with it.
Linguists studying biodiversity languages – languages spoken in biodiversity hotspots, often by indigenous communities with intimate knowledge of local ecosystems – have found that the geographic overlap between language diversity and biological diversity is not coincidental. Communities that have inhabited ecosystems for centuries develop fine-grained taxonomic and ecological knowledge encoded in language. The Toda people of the Nilgiris have names for hundreds of plants and animals in their ecosystem, many of which have ethnobotanical significance (medicinal, dietary, ritual) that has not been systematically documented in any other knowledge system. When Toda dies, some of that knowledge dies too – unless linguists and biologists work together to document it before it does.
“Every language is a unique window on the world. The extinction of a language is the extinction of a world view.”
G.N. Devy, Chair, People’s Linguistic Survey of India
The Three-Language Formula and Its Politics
India’s official educational language policy, the three-language formula, has been in various forms of contested implementation since the 1960s. In theory, students in most states learn three languages: their regional language, Hindi, and English. The National Education Policy 2020 reaffirmed and strengthened the mother-tongue medium of instruction in early education and the three-language formula. In practice, the formula’s implementation has been deeply politicised. Tamil Nadu has historically rejected the formula as a covert mechanism to impose Hindi, insisting on a two-language formula (Tamil and English). This disagreement has been a recurring flashpoint in Centre-state relations and reflects genuine concerns about linguistic assimilation that are not simply nationalist sentiment.
The mother-tongue medium debate is closely related. Research in educational psychology and learning outcomes consistently shows that children learn more effectively when early education is in their mother tongue. ASER data for India shows better foundational learning outcomes in states where early primary schooling is in the child’s home language. Yet parents in minority-language communities often choose Hindi or English-medium schooling for their children because they understand that economic opportunity depends on fluency in dominant languages. This rational individual choice, aggregated across millions of families, produces language shift that no amount of official policy preference for mother-tongue education can fully counteract without structural changes in the economic returns to minority language fluency.
Documentation Efforts and Digital Hope
The Endangered Archives Programme at the British Library has funded dozens of documentation projects in India, including recordings of tribal music and oral literature in languages with few remaining speakers. The Central Institute of Indian Languages in Mysuru has documented hundreds of Indian languages, though the quality and depth of documentation varies considerably. The Endangered Language Fund, Firebird Foundation, and ELDP have supported field documentation by Indian linguists. The Central University of Kerala has run a Language Endangerment Research programme. These are significant efforts, but they are nowhere near proportional to the scale and urgency of the problem.
Digital tools have created new possibilities. Community-developed apps for Korku and Ho have been created with Unicode support for scripts. Smartphone penetration in tribal areas has created platforms for language maintenance through WhatsApp groups, audio recordings, and social media in minority languages. The success of Odia Wikipedia and the growth of Bengali digital content demonstrate that digital presence can slow language shift even for medium-sized languages. For genuinely endangered languages, however, the challenge is that the communities most at risk are often the least connected digitally, the least resourced for documentation projects, and the least likely to have advocates in policy-making positions. Language endangerment in India is ultimately an equity problem: the languages at risk belong to communities at the margins, and the urgency of their linguistic loss is proportional to the communities’ marginality.
| Language | Location | Approximate Speakers | Status |
|---|---|---|---|
| Bo (Andamanese) | Andaman Islands | 0 (extinct since 2010) | Extinct |
| Great Andamanese | Andaman Islands | <50 | Critically endangered |
| Toda | Nilgiris, Tamil Nadu | ~1,600 | Severely endangered |
| Andamanese (mixed) | Andaman Islands | <100 | Critically endangered |
| Kodava | Kodagu, Karnataka | ~100,000 | Vulnerable |
| Ho | Jharkhand | ~1.4 million | Digital efforts underway |
How to Support Language Preservation
Support organisations working on language documentation: the People’s Linguistic Survey of India, Bhasha Research and Publication Centre, and the Central Institute of Indian Languages all do critical work. If you speak a heritage language at home – any language other than Hindi or English – use it with your children. Language shift happens one family at a time; so does language survival. Demand that your state government implement the Forest Rights Act provisions that protect tribal communities’ right to use their lands – language survival and land security are often linked. Source: People’s Linguistic Survey of India (2013); UNESCO Atlas of the World’s Languages in Danger; PLSI volumes on individual states.
The Policy Framework: What Exists and What Doesn’t
India’s Constitutional framework provides some foundation for language rights. The Eighth Schedule lists 22 official languages; Article 350A provides for instruction in the mother tongue in primary education where minority language communities are sufficient in number; States Reorganisation was conducted partly on linguistic lines. These provisions were designed for the major regional languages, not for the hundreds of smaller languages spoken by tribal and other communities. There is no statutory framework for the rights of speakers of non-Scheduled languages, no mandatory documentation obligation before development projects displace language-speaking communities, and no funded programme at the scale required to document the hundreds of languages at risk within the next generation.
The Tribal Language Task Force within the Ministry of Tribal Affairs has initiated some documentation projects, and the Sahitya Akademi has a programme for publishing literature in tribal languages. The State Institute of Languages in some states – most notably Karnataka – have done more systematic work. But the gap between the resources being applied and the scale of the linguistic emergency is vast. Language endangerment in India is unlikely to be addressed seriously until it becomes politically salient in a way that it currently is not. The communities whose languages are disappearing are among the most politically marginalised in the country; their linguistic loss is, in this sense, a direct expression of their broader marginalisation. Addressing one without the other is not possible. See also our reporting on the intersections of education access and community identity in why 50 million Indian students drop out before class 10, a crisis with its own language dimensions.
Model Cases: What Language Survival Looks Like
Welsh in the United Kingdom, Maori in New Zealand, and Catalan in Spain are instructive examples of language survival under demographic pressure. Welsh dropped to fewer than 500,000 speakers in the mid-20th century before policy intervention – mandatory Welsh-medium education in Wales, bilingual government services, and BBC Wales broadcasting in Welsh – reversed the decline. Welsh now has approximately 900,000 speakers, with a younger speaker population than 50 years ago. Maori was nearly extinct before the kohanga reo (language nest) movement in the 1980s established immersive Maori-language preschool education. Today, Maori is an official language of New Zealand with significant government investment in language revitalization. The common thread in both cases is institutional support at scale, investment in early childhood language transmission, and political decision to treat the minority language as a public good worth funding.
India could adopt elements of both models. The model of mother-tongue medium instruction through at least Class 5, endorsed by the NEP 2020, is the most important structural intervention for intergenerational language transmission. Community-led language documentation, supported by government funding, can preserve what is at risk before it is lost. Tribal autonomy councils in scheduled areas could be empowered to make educational decisions in their languages. The Forest Rights Act, if fully implemented, would protect the land-based communities whose territorial connection sustains language use. None of this is beyond India’s institutional capacity. What is missing is political will of the kind that comes from viewing linguistic diversity not as a complication to be managed but as a national asset worth investing in. See also how education access failures compound this problem in our analysis of why 50 million students drop out before class 10.