Somewhere in northern Australia, the last fluent speaker of a language is an elderly person who can no longer find anyone to have a proper conversation with. The language existed for thousands of years, encoding an intimate knowledge of a specific landscape, a set of social relationships, a particular way of construing time and space and causality. When that person dies, the language dies with them. This is not a hypothetical scenario. It is happening now, repeatedly, quietly, across the world. Linguists estimate that a language dies approximately every two weeks — that the current rate of language extinction is without historical precedent and, on current trajectories, will transform the linguistic landscape of humanity within a century.
Of the approximately 7,000 languages spoken today, UNESCO's Atlas of the World's Languages in Danger categorises more than 2,500 as endangered. The most widely cited projection, from linguists Mark Pagel and colleagues as well as work by David Crystal, suggests that between 50 and 90 percent of the world's languages may disappear by 2100, leaving a world dominated by a small number of major languages — primarily Mandarin, English, Spanish, Arabic, Hindi, and a handful of others. This would represent the most severe contraction of linguistic diversity in human history, far exceeding what occurred during any previous period of population change or conquest.
But this is also the story of resistance, revival, and the extraordinary human capacity to decide that something worth losing is worth saving. Welsh, which seemed to be dying steadily through the mid-twentieth century, has stabilised and grown through decades of deliberate policy and cultural mobilisation. Hebrew, which had not been anyone's everyday spoken language for nearly 1,700 years, was revived as a living vernacular through one of the most remarkable acts of linguistic will in history. Maori, Hawaiian, and Catalan have all been pulled back from the edge of functional extinction through combinations of policy, education, and community commitment. The story of how languages die is also, inseparably, the story of what it takes to save them.
"Each language is a cathedral, a unique architecture of thought built over thousands of years. When it falls, it takes with it a way of being human that cannot be reconstructed." -- David Crystal, 'Language Death' (2000)
Key Definitions
Language death: The process by which a language ceases to be spoken, either because all its speakers have died or because its speakers have shifted to using a different language (language shift). Language death through shift is far more common than extinction of speaker populations. A language is typically considered 'dead' when it has no living native speakers, though it may persist in scholarly use (Latin, Sanskrit, Classical Greek).
Moribund language: A language that is no longer being acquired by children as a first language within a community. Moribund status is particularly significant because it means the language has approximately one generation of native speakers remaining. Languages can survive in moribund status for decades but cannot recover without deliberate intervention to restart intergenerational transmission.
Language shift: The process by which a community gradually abandons one language in favour of another, typically over two to three generations. Language shift usually begins with bilingualism in the community's younger generations, proceeds through a stage where the original language is used only in restricted domains (home, religious ceremony), and culminates in the younger generation having no productive competence in the original language.
Language revitalisation: Deliberate efforts to halt or reverse language shift and restore a language to everyday use within a community. Revitalisation ranges from documentation and archival work (recording the language before it disappears) to full-scale revival efforts aimed at creating new speakers. The Maori term for revitalisation, 'te kohanga reo' (language nest), describes the immersive early-childhood approach that has become influential globally.
Linguistic relativity: The hypothesis, associated with Benjamin Lee Whorf and Edward Sapir and refined by contemporary researchers including Lera Boroditsky, that language shapes cognitive processes — that speakers of different languages attend to the world differently, organise experience differently, and think about time, space, number, and causality in language-influenced ways.
The Scale of the Crisis
Why 7,000 Languages Become 2,700
The projection that the world will have approximately 2,700 languages by 2100 rests on current rates of language death and the proportion of existing languages that are already moribund or severely endangered. David Harrison at Swarthmore College, whose 2007 book 'When Languages Die' provides one of the most comprehensive accounts of the crisis, notes that the 7,000-language figure itself represents a dramatic contraction from pre-colonial diversity: researchers estimate that North America alone may have had 500-1,000 languages before European contact, compared to the 150-200 still spoken today.
The linguistic crisis is geographically concentrated. Papua New Guinea, a country of approximately 9 million people, hosts approximately 850 languages — more than in any other country on earth — many of them spoken by communities of a few hundred people and under severe pressure from Tok Pisin and English. Australia had approximately 250 indigenous languages at the time of European colonisation in 1788; today, fewer than 20 are spoken by all age groups. The Americas, Africa, and Southeast Asia similarly host large numbers of severely endangered languages.
The Geography of Endangerment
A striking feature of the linguistic crisis is that it is inversely related to political power. Languages with state support, educational use, and economic value are not endangered. It is the languages of politically marginalised communities — indigenous peoples, rural minorities, diaspora communities without institutional support — that are dying. This makes language death inseparable from histories of colonialism, dispossession, and cultural suppression.
Researcher Nicholas Evans at the Australian National University, in 'Dying Words: Endangered Languages and What They Have to Tell Us' (2010), documents this correlation in detail: virtually every severely endangered language is spoken by a community that has experienced cultural dispossession in the modern period. Language death is not a natural process but a social one, driven by power asymmetries that policy can address.
Why Languages Die
The Economics of Language Choice
The immediate driver of language shift is almost always economic: speakers choose the dominant language because it offers access to employment, education, government services, and social mobility that the minority language does not. When parents make the calculation that their children's opportunities will be better in English, Mandarin, or Spanish than in the heritage language, they often choose to raise them in the dominant language — a rational individual choice with collectively catastrophic consequences for linguistic diversity.
Sociologist Joshua Fishman, whose 1991 book 'Reversing Language Shift' remains the foundational text in language revitalisation theory, argued that understanding this economic logic was essential to designing effective interventions. Revitalisation efforts that do not address the economic disadvantage of speaking a minority language — that do not create domains where the minority language has tangible value — tend to fail. Welsh revitalisation succeeded partly because it created jobs for Welsh speakers (in Welsh-medium education, broadcasting, and public administration) that gave the language economic as well as cultural value.
Colonial Suppression
Colonial language policies were often explicitly designed to extinguish indigenous languages. Residential school systems in Canada, the United States, and Australia removed indigenous children from their communities and prohibited the use of indigenous languages, in many cases through physical punishment. The Canadian Royal Commission on Aboriginal Peoples (1996) documented systematic suppression of indigenous languages as part of a broader programme of cultural assimilation. The intergenerational trauma produced by these policies continues to affect language transmission in indigenous communities: many communities lost the generation that should have transmitted the language, creating gaps that are structurally difficult to close even with strong revitalisation motivation.
Urbanisation and Migration
Urbanisation accelerates language shift by removing speakers from the community contexts where minority languages are functional. Urban migrants find that minority languages serve little purpose in city environments dominated by national languages; their children grow up primarily in national language contexts. The global rural-to-urban migration of the past century has therefore been a major driver of language shift, affecting hundreds of minority languages worldwide as their rural speaker communities decline.
What Is Lost When a Language Dies
Ecological Knowledge
Languages encode knowledge of specific environments accumulated over thousands of years of intimate observation. Many indigenous languages have vocabularies for ecological phenomena — plant varieties, animal behaviours, environmental indicators — that have no equivalents in national languages and represent knowledge systems developed through long-term, detailed ecological relationship. When the Solega language of southern India loses speakers, it loses vocabulary encoding decades of observation of the Biligiri Rangan Hills ecosystem. Ethnobotanist Wade Davis, in 'The Wayfinders: Why Ancient Wisdom Matters in the Modern World' (2009), has argued that indigenous knowledge systems represent irreplaceable intellectual contributions to human understanding of the natural world.
Alternative Cognitive Architectures
The strongest version of linguistic relativity — that language determines thought — is no longer widely accepted, but substantial evidence has accumulated that language influences habitual patterns of attention and cognitive organisation. Lera Boroditsky's research at Stanford and UC San Diego has documented differences in how speakers of different languages conceptualise time (left-to-right, east-to-west, uphill-to-downhill), number (languages with approximate versus exact number words produce different numerical cognition), and causality (languages that mark accidental causation differently than intentional causation affect how speakers remember and attribute responsibility for events).
The Guugu Yimithirr speakers of Queensland, studied by linguist Stephen Levinson, provide a striking example: their language uses absolute cardinal directions (north, south, east, west) for all spatial reference, rather than the egocentric coordinates (left, right, in front, behind) used in most European languages. Guugu Yimithirr speakers develop extraordinary geographical orientation as a result, maintaining accurate compass-point awareness even in unfamiliar environments. This is not merely a linguistic difference but a cognitive one — a different way of relating to space that has practical and conceptual consequences. When a language that encodes such differences disappears, the cognitive strategy it supported typically disappears with it.
Oral Literature and Cultural Heritage
Every living language has a literature, even if it is entirely oral. Epic narratives, songs, proverbs, riddles, ceremonial texts, and oral histories constitute cultural heritage as significant as written traditions, but far more vulnerable. When a language becomes moribund, its oral literature typically dies with its last fluent speakers unless intensive documentation efforts capture it. The loss is not merely aesthetic: oral literature often encodes historical knowledge, legal traditions, cosmological frameworks, and social norms that have no written equivalent.
The Hebrew Revival: History's Unique Case
Eliezer Ben-Yehuda
Hebrew had not been spoken as an everyday vernacular since approximately 200 CE, when Aramaic displaced it among Jewish populations in the Middle East. For nearly 1,700 years, Hebrew survived as a language of Torah study, prayer, poetry, and scholarly correspondence — used by educated Jews across the diaspora but not as anyone's mother tongue. The dream of reviving it as a spoken language seemed to most contemporaries to be utopian fantasy.
Eliezer Ben-Yehuda, born Eliezer Yitzhak Perlman in Russian Lithuania in 1858, made the revival of Hebrew his life's work after becoming convinced that Jewish national revival required a Jewish national language. He immigrated to Palestine in 1881 with the explicit intention of speaking only Hebrew in his daily life. He insisted that Hebrew be the language of his household and raised his son Ben-Zion — born 1882 — as the first child in modern times to grow up with Hebrew as a first language. This was an extraordinary act: Ben-Zion initially had no one to play with who spoke his language, because the Hebrew-speaking community that Ben-Yehuda envisioned did not yet exist.
The Method and the Dictionary
Ben-Yehuda understood that revival required solving the vocabulary problem. Biblical Hebrew had no words for newspaper, train, dictionary, towel, or thousands of other modern concepts. Ben-Yehuda coined new words, often by extending biblical roots according to the morphological patterns of biblical Hebrew, and established a Hebrew Language Committee (Va'ad HaLashon, later the Academy of the Hebrew Language) to systematically expand the lexicon. His complete 'Dictionary of Ancient and Modern Hebrew,' completed posthumously in 1959, ran to 17 volumes.
The Jewish educational institutions established in Palestine — the Alliance Israelite schools, and later the institutions of the Zionist movement — adopted Hebrew as their language of instruction, creating the first generation of children for whom Hebrew was not primarily a religious language but a school language and, eventually, a vernacular. By the time of the British Mandate, Hebrew was sufficiently established as a living language that it was recognised as one of three official languages alongside English and Arabic.
The State and the Language
The establishment of the State of Israel in 1948 gave Hebrew what Ben-Yehuda could not have achieved alone: a state apparatus, an educational system, and the need for a common language among Jewish immigrants from dozens of countries. Modern Hebrew (Ivrit) is now the first language of the majority of Israel's Jewish population and has approximately nine million speakers worldwide. It is the only case in linguistic history where a language that ceased to be spoken as a vernacular was successfully revived to full everyday use across a society.
Welsh: Revitalisation Through Policy
From Decline to Stability
Welsh is one of the oldest living languages in Europe, descended directly from the Brittonic Celtic languages spoken across Britain before the Roman and Anglo-Saxon invasions. At the 1911 census, 43 percent of Wales's population spoke Welsh — nearly a million people. By 1961, the proportion had fallen to 26 percent and was still declining. The Welsh Language Society (Cymdeithas yr Iaith Gymraeg), founded in 1962, began a campaign of civil disobedience demanding official status for Welsh that kept the language politically visible. Sociologist Joshua Fishman later cited Welsh as an important case study in the relationship between official recognition and revitalisation success.
The Educational Pillar
The Welsh-medium school system was the most important structural intervention in Welsh revitalisation. The first Welsh-medium primary school (Ysgol Gymraeg) opened in Llanelli in 1947, teaching all subjects through Welsh to children in a predominantly English-speaking community. The system expanded steadily through the 1960s and 1970s. Welsh-medium and bilingual schools now educate approximately 22 percent of children in Wales. Research by Virginia Yip and colleagues at Cambridge documented that children in Welsh-medium education achieve comparable outcomes to English-medium peers in core subjects while gaining functional bilingualism.
Broadcasting and Official Status
S4C, the Welsh-language television channel, launched in 1982 after a sustained campaign that included Plaid Cymru leader Gwynfor Evans threatening to fast to death. The channel gave Welsh a presence in the broadcast media that provided both cultural legitimacy and employment for Welsh speakers. The Welsh Language Acts of 1993 and 2011 established Welsh as an official language in Wales and required public bodies to provide Welsh-language services. By the 2021 census, 538,000 people in Wales could speak Welsh — approximately 17 percent of the population. The proportion has stabilised and shows modest growth in younger age groups, suggesting that revitalisation is working.
Other Revival Stories
Maori and the Language Nest
The Maori language of New Zealand (Te Reo Maori) was in severe decline by the 1970s, with fewer than 70,000 fluent speakers, most of them elderly. The kohanga reo (language nest) movement, established in 1982, created immersive Maori-language early childhood environments where all communication was in Maori. The model spread to Kura Kaupapa Maori (Maori-medium schools) and eventually to a Maori-language television channel (Maori Television, 2004). The 2023 census found approximately 185,000 people able to hold a conversation in Maori — a significant increase — though full revitalisation remains a long-term project.
Hawaiian
Hawaiian came extremely close to extinction. By the 1980s, Hawaiian had fewer than 50 children who were native speakers, virtually all on the private island of Ni'ihau. The establishment of Punana Leo (language nest) immersion schools in 1984 and subsequent Hawaiian-medium public schools created a new generation of Hawaiian speakers. By 2015, there were approximately 18,000 speakers, including thousands of children. Hawaiian's revival is directly modelled on the Maori kohanga reo approach and is studied as a case of successful early intervention.
Practical Takeaways
The crisis of language extinction is not inevitable. Welsh, Hebrew, Maori, and Hawaiian demonstrate that determined communities, with appropriate policy support, can halt and reverse language decline even from advanced stages. The key variables are consistent across cases: intergenerational transmission, meaning languages must be acquired by children; institutional support through education and media; and economic domains where the language has tangible value.
For cultural observers, the significance of the crisis goes beyond linguistics: it is a crisis of diversity of human thought and experience, of ecological knowledge, of cultural heritage. For policymakers in multilingual societies, the evidence supports investment in minority language education, official recognition, and the creation of employment domains for minority language speakers as genuinely effective tools.
The projection of 2,700 languages by 2100 is not fate. It is a forecast based on current policy choices. Different choices — to fund Welsh-medium schools, to establish language nests, to recognise indigenous languages officially, to create economic value for speaking minority languages — produce different outcomes. The extraordinary cases of Welsh and Hebrew suggest that languages that seem doomed can survive, and that the effort to save them is worth making.
References
- Crystal, D. (2000). Language Death. Cambridge University Press.
- Harrison, K. D. (2007). When Languages Die: The Extinction of the World's Languages and the Erosion of Human Knowledge. Oxford University Press.
- Fishman, J. A. (1991). Reversing Language Shift: Theoretical and Empirical Foundations of Assistance to Threatened Languages. Multilingual Matters.
- Nettle, D., & Romaine, S. (2000). Vanishing Voices: The Extinction of the World's Languages. Oxford University Press.
- Evans, N. (2010). Dying Words: Endangered Languages and What They Have to Tell Us. Wiley-Blackwell.
- Boroditsky, L. (2011). How language shapes thought. Scientific American, 304(2), 62–65.
- Davis, W. (2009). The Wayfinders: Why Ancient Wisdom Matters in the Modern World. House of Anansi Press.
- Spolsky, B. (2014). The Languages of the Jews: A Sociolinguistic History. Cambridge University Press.
- UNESCO. (2010). Atlas of the World's Languages in Danger (3rd ed.). UNESCO Publishing.
- Hinton, L., & Hale, K. (Eds.). (2001). The Green Book of Language Revitalization in Practice. Academic Press.
- Welsh Government. (2017). Cymraeg 2050: A Million Welsh Speakers. Welsh Government.
- Skutnabb-Kangas, T. (2000). Linguistic Genocide in Education — or Worldwide Diversity and Human Rights? Lawrence Erlbaum Associates.
Frequently Asked Questions
How many languages are endangered?
Linguistic diversity is in rapid decline. Of approximately 7,000 languages currently spoken worldwide, UNESCO's Atlas of the World's Languages in Danger categorises more than 2,500 as endangered, meaning they are at risk of ceasing to be spoken within the coming generations. Linguists estimate that between 50 and 90 percent of currently spoken languages may disappear by the end of the twenty-first century, depending on the success of revitalisation efforts. A language is considered 'moribund' when it is no longer acquired by children as a first language — at that point, the language has approximately one generation remaining before functional extinction. The crisis is geographically concentrated: Australia, the Americas, and parts of Asia and Africa host disproportionate shares of the world's most endangered languages.
Why do languages die?
Languages die when their speaker communities shift to using another language for most or all communicative purposes, a process called language shift. The drivers are typically asymmetries of power and opportunity: dominant languages offer better access to education, employment, government services, and social mobility. When parents calculate that their children will have better life prospects speaking a dominant language, they often choose to raise them in that language rather than passing on a minority or indigenous tongue. Colonial history has been a major driver of language death globally: policies of forced assimilation, prohibition of indigenous languages in schools, and the economic and social subordination of indigenous communities systematically eroded the conditions in which minority languages could be transmitted across generations.
What is lost when a language dies?
When a language dies, its speakers lose access to a unique way of articulating human experience. Each language encodes distinctions that other languages do not make and reflects specific ecological, social, and historical experience of its speaker community. The Guugu Yimithirr language of Australia uses cardinal directions (north, south, east, west) rather than relative directions (left, right) for all spatial reference — its speakers have extraordinarily precise geographical orientation that differs fundamentally from speakers of languages using egocentric spatial reference. Research by Lera Boroditsky at Stanford and UC San Diego has documented multiple ways in which different languages lead their speakers to attend differently to time, number, colour, and causality. When a language dies, these ways of construing experience are lost, along with oral literature, ecological knowledge, and the cultural heritage encoded in vocabulary that has no equivalent in surviving languages.
How was Welsh saved?
Welsh is one of the most remarkable language revitalisation stories in the world. In 1911, approximately 43 percent of the population of Wales spoke Welsh. By 1961, this had fallen to 26 percent, and the trajectory suggested eventual extinction. The revitalisation effort was built on several pillars: the establishment of Welsh-medium schools (Ysgolion Cymraeg) from 1939 onwards, which created generations of new Welsh speakers; the establishment of S4C, a Welsh-language television channel, in 1982; the Welsh Language Act of 1993 and the Welsh Language (Wales) Measure 2011, which gave Welsh official status alongside English; and devolution in 1999, which gave the National Assembly for Wales authority to promote the language. By 2021, the proportion of Welsh speakers had stabilised and increased in some areas. The Welsh government's target is one million Welsh speakers by 2050.
Is the revival of Hebrew unique in history?
Hebrew's revival as a spoken vernacular language is essentially unique in linguistic history. Hebrew had ceased to be a spoken everyday language around 200 CE, surviving for nearly 1,700 years as a language of religious study, prayer, and scholarly correspondence — but not as anyone's mother tongue. Eliezer Ben-Yehuda, who immigrated to Palestine in 1881, dedicated his life to the project of reviving Hebrew as a spoken language. He raised his son Ben-Zion as the first native Hebrew speaker in modern times, coined thousands of new words for modern concepts, compiled a comprehensive dictionary, and organised societies to promote everyday Hebrew use. The establishment of the State of Israel in 1948 gave the revived language a state, an educational system, and millions of new speakers. Modern Hebrew (Ivrit) now has approximately nine million speakers and is the first language of most Israeli-born citizens.