NEP's three-language formula: Rhetoric vs reality

CHENNAI: We humans tend to make snap judgements without enough evidence and cling to them even when proven wrong. This flaw distorts not only personal judgement but also public policy. Far too many decisions, affecting millions of people, are based on ideology, intuition, or untested assumptions, often with costly consequences. The antidote? Evidence-based policymaking. This means grounding decisions in real-world data, rigorous research, and measurable outcomes — to identify what works, improve what doesn’t, and abandon what harms. It is not enough if policies are well-meaning; they should be well-proven.

Launched in 1968, India’s three-language formula was meant to promote multilingualism, national unity, cross-regional mobility, and employment opportunities. It stands as a textbook example of an ideology-driven policy that ignores real-world data and resists course correction despite decades of patchy implementation and underwhelming outcomes.

Instead of rethinking its relevance, the National Education Policy (NEP) 2020 has reinforced it, making a third language compulsory in schools nationwide. This analysis highlights the gap between the rhetoric and reality of India’s three-language formula, and demonstrates why NEP 2020’s continued push for it is a serious policy misstep in need of urgent, evidence-based review.

Rhetoric: The three-language formula has been successfully implemented across India (except Tamil Nadu).

Reality: If that were true, India — excluding Tamil Nadu which has steadfastly stuck to its two-language formula since 1968 — should have been a multilingual success story. But according to the 2011 Census, only 7% of Indians reported knowing three languages.

In Hindi-speaking States like Rajasthan, Uttar Pradesh, Bihar, Madhya Pradesh, and Chhattisgarh, the number drops below 2%. Even bilingualism remains limited—just 26% nationwide, and under 15% in these five States. This is a massive policy failure.

Few government initiatives have been implemented so widely and for so long, yet yielded such meagre results.

Educational surveys also paint a dismal picture. The GoI’s National Achievement Survey (NAS) 2017 found that just less than half of Class 8 students could read a simple paragraph, write a basic essay, or understand grammar in their regional language or Hindi. NAS 2021 showed marginal gains, but the overall picture remained bleak.

English proficiency, tested only in NAS 2018 at the Class 10 level, was similarly poor. Tellingly, NAS never assesses third-language proficiency, conveniently masking policy failure.

The Annual Status of Education Report (ASER) conducted by the NGO Pratham in rural schools echoes these concerns. In 2018, 27% of Class 8 students couldn’t read a Class 2-level text in their regional language or Hindi — a figure that worsened to 30.4% in 2022. In 2016, 73.8% of Class 8 students couldn’t read basic English sentences; in 2022, this still stood at 53.3%. Like NAS, ASER also avoids evaluating third-language proficiency.

When so many students struggle with their first and second languages, what’s the justification for mandating a third? Shouldn’t the focus be on teaching two languages well—rather than three poorly?

Rhetoric: Learning a third language improves cognitive abilities.

Reality: NEP 2020 endorses this claim in a single sentence — without citing any global research. But cognitive science on multilingualism is far more nuanced.

While bilingualism has well-documented cognitive benefits — enhanced executive function, problem-solving, and delayed cognitive decline — these advantages don’t scale linearly. Research shows diminishing returns beyond two languages.

According to The Cambridge Handbook of Third Language Acquisition, cognitive gains arise only when learners are challenged but not overwhelmed. If students are still struggling with their first (L1) and second (L2) languages, learning a third (L3) may exceed their cognitive capacity, causing mental fatigue, and diminished learning efficiency. It also reduces practice time for L1 and L2, causing their attrition, with L2 being more vulnerable.

Cross-linguistic interference can cause pronunciation, grammar, and vocabulary mix-ups. Achieving equal fluency in three languages is rare; typically, one dominates while the others weaken.

Research also shows that language similarity matters. Speakers of Indo-Aryan languages like Marathi, Punjabi, or Odia find it easier to learn Hindi due to shared grammar, vocabulary, and phonetics, benefiting from what linguists call ‘facilitative transfer’. But speakers of Dravidian languages such as Tamil, Austro-Asiatic language like Santali, and Tibeto-Burman languages such as Mizo, face ‘non-facilitative transfer’, making L3 learning harder and the burden deeply unequal.

This is compounded by a troubling gap between stated policy and ground realities. The 2011 Census recorded 19,569 mother tongues, but 18,200 of these were lumped under the label ‘Others’. The fortunate 1,369 mother tongues were grouped under 121 languages, of which only 22 are included in the Eighth Schedule of the Constitution.

Article 350A mandates primary education in one’s mother tongue, and both UNESCO and NEP 2020 endorse it. Yet, most schools teach in the State’s dominant regional language — not the child’s actual mother tongue.

As a result, millions of children speaking tribal and other minority languages such as Tulu face a four-language burden: (1) their mother tongue (no classroom support), (2) the State’s regional language (medium of instruction), (3) Hindi/Sanskrit (imposed), and (4) English (poorly taught). The result: cognitive overload, rote learning, language loss, and soaring dropout rates.

Among tribal students, the dropout rate is 35% — three times the national average. This isn’t new. Educationist Dhir Jhingran highlighted this crisis in a landmark study in 2009 — but little has changed.

NEP 2020’s rigid trilingual mandate ignores these complexities, and disproportionately burdens the most vulnerable.

Ten non-scheduled languages, having over one million speakers each, should be added to the Eighth Schedule and made mediums of instruction in primary schools to abide by the Constitutional mandate and ensure greater linguistic inclusion.

Rhetoric: The three-language formula offers real choice and there is no imposition of Hindi or Sanskrit.

Reality: NEP 2020 promises that no language will be imposed on States, and students are free to choose any three languages, as long as at least two are Indian. But this freedom exists only on paper.

Take a school in Tamil Nadu. Let’s say 30% of students want Telugu, 30% Malayalam, 20% Kannada, 10% Hindi, and 10% Sanskrit as their third language. In theory, the school should offer all five as third-language options. In practice, it can’t. Hiring qualified teachers, developing and printing textbooks, and scheduling classes for five different languages are logistically and financially unviable—especially when student preferences change year to year.

On the other hand, Hindi and Sanskrit come pre-packaged — centralised resources, readily available textbooks, and a steady supply of teachers. As a result, economic and logistical pressures will quietly nudge schools in non-Hindi-speaking states toward offering Hindi or Sanskrit — or both — as the default third language. What NEP 2020 presents as ‘flexibility’ is, in reality, a subtle standardisation that undermines linguistic diversity and constrains true student choice.

This implicit imposition is compounded by NEP 2020’s explicit ideological bias. It devotes more discussion to Sanskrit — a language spoken by fewer than 25,000 people and with little practical utility today — than to English, which is essential for global careers, higher education, business, science and technology. In stark contrast, countries like China, South Korea, Japan, Russia and Brazil are ramping up English instruction to stay competitive globally.

NEP 2020 neither acknowledges this trend nor lays out any roadmap to strengthen English proficiency among Indian students. Beyond English, foreign languages like French, German, Spanish, Mandarin, and Arabic offer substantial career advantages worldwide. Hindi or Sanskrit as a third language do not.

According to Naukri.com, over 50,000 Indian jobs require French or German; fewer than 100 mention Sanskrit, mostly in religious or academic contexts. By limiting foreign language options to just one (invariably English), NEP 2020 undermines the only tangible benefit of learning a third language — better career prospects.

By treating language learning as a matter of cultural pursuit instead of economic opportunity, NEP 2020 fails the very students it aims to empower. India must stop romanticising token multilingualism and start prioritising practical language skills that open doors – not close them.

Rhetoric: India’s public school system can successfully implement the three-language formula.

Reality: While students are free to learn multiple languages outside school, it’s neither practical nor affordable to implement the three-language formula in public education.

In Tamil Nadu alone, hiring just one third-language teacher for each of the State’s 8,074 government and aided high schools and higher secondary schools would cost approximately Rs 560 crore annually — just in salaries. Add textbooks, teacher training, and infrastructure, and the financial burden becomes unsustainable.

Most rural schools can’t even fill existing posts in core subjects, let alone hire new third-language teachers.

Wouldn’t that money be better spent on strengthening foundational literacy, math and science — or preparing students for the future through subjects like Artificial Intelligence? Countries like China, Estonia, and South Korea are already teaching AI in schools. Shouldn’t India follow suit?

The truth is, relevance — not regulation — should guide language learning. A Keralite nurse in Mumbai picks up Marathi for her patients, speaks Malayalam at home, and uses English for documentation. No policy forced it; life did. This is contextual multilingualism: organic, need-based, and effective.

The world has changed — our education policy hasn’t. Last year, in Berlin, I followed a German-only tour of the Reichstag effortlessly through Google Translate’s live voice-to-text tool. With AI-powered translation tools like DeepL now rivalling human translators, learning a third language no longer needs a classroom.

Let’s focus school learning on what matters most: strong literacy in mother tongues and English. As for a third language, let students choose it on their own terms — guided by interest, life context, and the tools of the digital age. It’s cheaper, smarter, and more in tune with how languages are actually learned in the real world.

Rhetoric: National unity requires a common language (Hindi).

Reality: The idea that Hindi is essential to national unity may sound intuitive — but it’s neither supported by data nor rooted in India’s history. In a country as linguistically rich as India, privileging one language is not only unjustified but potentially destabilising.

Let’s look at the numbers. The 2011 Census claims that 43.63% of Indians speak Hindi. However, as noted scholar GN Devy points out in ‘India: A Linguistic Civilization’, this figure is inflated by categorising 53 distinct languages as ‘dialects’ of Hindi.

Many, like Awadhi, Bhojpuri, Brajbhasha, Magadhi, Chattisgarhi, and Rajasthani, are completely independent languages with deep literary traditions and predate Hindi by centuries. Some, like Banjari, are not even mutually intelligible with Hindi. Strip away this misclassification, and true Hindi speakers are only 25% of the population.

The 2011 Census also reveals that 63% of Indians have never left their birthplace, 85% remain within their native district, and 95% never migrated out of their home state. Daily life — in education, work, administration, commerce, and culture — is carried out overwhelmingly in regional languages. And where inter-state migration does occur, it is mostly away from the Hindi-speaking belt — toward the South, West, and Delhi where economic opportunities are concentrated. In such a scenario, projecting Hindi as a ‘natural’ lingua franca simply doesn’t stand up to scrutiny.

The notion that one language is necessary for national unity is not an Indian idea — it’s a European one. In the 19th and 20th centuries, countries like Germany, Italy, Poland, Hungary and Romania pursued monolingualism as a strategy for forging national identity. But applying that logic to India — one of the world’s most linguistically diverse civilisations — is like flattening a rainforest into a manicured lawn. It may look tidy, but it destroys the ecosystem.

A better analogy is the European Union, which recognises 24 official languages and treats linguistic diversity as foundational to democratic inclusion. Imagine imposing German or French as the EU’s only official language — how long would it last?

If India needs a model, it’s not in the West but in Southeast Asia. In ‘From Third World to First, Singapore’s first Prime Minister Lee Kuan Yew recalls how he resisted intense pressure from the Chinese majority (74.3%) to declare Mandarin the sole national language. Recognising that this would alienate Malays (13.5%), Tamils (9%) and other minorities, and to ensure fairness, Lee chose English — a colonial legacy but a neutral language — as Singapore’s lingua franca.

Singapore adopted a bilingual education policy: English for economic mobility, and mother tongues — Mandarin, Malay, or Tamil — for cultural identity. The result: social cohesion, inclusive governance, and meteoric economic growth. Today, Singapore ranks among the top nations globally in education, innovation, and prosperity.

India’s own post-Independence path shows that accommodation, not imposition, keeps the Union strong. We reorganised States along linguistic lines, recognised 22 official languages in the Eighth Schedule, and retained English as an associate official language. Historian John Keay, in Midnight’s Descendants, argues that this linguistic flexibility helped preserve India’s unity — unlike Pakistan, where attempts to impose Urdu as the sole national language alienated Bengalis and led to the creation of Bangladesh.

When less is more

NEP’s three-language formula fails the test of evidence-based policymaking. The data is clear: trilingualism hasn’t worked, and even bilingual proficiency remains elusive for many Indian students.

The policy overlooks the cognitive overload, budget constraints, implementation challenges, limited real-world utility, and the growing availability of AI-powered translation tools. It also disregards proven success stories.

Tamil Nadu’s two-language formula has delivered better educational and economic outcomes than most other States for over 50 years. Internationally, Singapore’s bilingual model has built one of the world’s most inclusive and globally competitive societies.

India should choose educational pragmatism over linguistic nationalism. It’s time to discard the failed three-language formula and focus on bilingual proficiency — English for employment opportunities, regional languages for cultural preservation — and use AI-powered tools for learning any additional language, when and if students need it.

India doesn’t need a national language to stay united. It needs a language policy that respects its linguistic plurality and puts students first. As Singapore’s success shows, the most effective approach isn’t picking favourites — it’s picking what works.

-The writer is a former IAS officer of Tamil Nadu cadre, and the former Vice-Chancellor of Indian Maritime University, Chennai