World Languages

The number of languages spoken in the world today is not a settled figure, but a reasoned estimate that shifts each year as fieldwork, revitalization efforts, and language deaths are recorded. The most widely cited institutional source, Ethnologue, published by SIL International, reports approximately 7,168 living languages in its 2024 edition. Other catalogs, including Glottolog maintained at the Max Planck Institute for Evolutionary Anthropology, arrive at somewhat different totals because they classify dialects and closely related varieties differently. Scholars therefore treat all such counts as approximations.

Linguistic diversity is distributed very unevenly. A small number of countries host an outsized share of the world's languages. Papua New Guinea is home to approximately 840 languages, Indonesia to approximately 710, Nigeria to approximately 525, India to roughly 450 depending on classification decisions, and China to approximately 300. A dozen countries together account for more than two-thirds of the world's linguistic diversity, while many other nations are dominated by only one or two widely spoken languages.

Speaker populations are even more uneven than language counts. The ten largest languages by number of native speakers together account for approximately 46 percent of the world's population. The remaining 54 percent of people speak thousands of smaller languages, many with fewer than ten thousand speakers. UNESCO estimates that roughly 40 percent of the world's languages are endangered to some degree, and scholars at the Endangered Languages Project and the Linguistic Society of America have warned that a significant share may no longer be actively spoken by the end of the current century if present trends continue.

This page offers an encyclopedic overview of the world's languages: how they are counted, how they are grouped into families, how they are written, which are most widely spoken, how some are endangered and how others are revitalized, and how new technologies are reshaping what it means to speak and translate. It links throughout to country profiles on CountryReports and to the institutions whose research underpins the field.

A language family is a group of languages descended from a common ancestor, identified through systematic correspondences in vocabulary, sound, and grammar. Linguists currently recognize more than one hundred and forty distinct families, along with a smaller number of language isolates that have no proven relatives. Glottolog, the reference catalog maintained at the Max Planck Institute for Evolutionary Anthropology, lists families and their internal structure and is widely used by researchers at Harvard, MIT, Stanford, and the Linguistic Society of America.

A handful of families account for the overwhelming majority of speakers. Indo-European is the largest by speaker population, encompassing Spanish, English, Hindi, Bengali, Portuguese, Russian, Persian, French, German, and many more, with approximately 3.2 billion total speakers. Sino-Tibetan includes Mandarin, Cantonese, Tibetan, and Burmese, and is second by speakers, with approximately 1.4 billion. Niger-Congo, the largest family by number of member languages, includes Yoruba, Igbo, Swahili, Zulu, and Shona, and accounts for approximately 700 million speakers across sub-Saharan Africa. Afroasiatic, with Arabic, Hebrew, Hausa, and Amharic, counts about 500 million.

Further major families include Austronesian, spoken from Madagascar to Easter Island and including Indonesian, Malay, Tagalog, Javanese, and Maori, with approximately 380 million speakers; Dravidian, concentrated in southern India and including Tamil, Telugu, Kannada, and Malayalam, with about 250 million; Turkic, including Turkish, Uzbek, Kazakh, and Azerbaijani, with about 200 million; Japonic, comprising Japanese and the Ryukyuan languages, with about 130 million; and Uralic, including Finnish, Hungarian, and Estonian, with about 25 million. Language isolates such as Basque in Europe, Ainu in northern Japan, and, by some classifications, Korean are notable cases where no common ancestor with neighboring languages has been demonstrated.

Writing is a comparatively recent human technology. The earliest known writing system is Sumerian cuneiform, which emerged in Mesopotamia around 3200 BCE. Since then, hundreds of scripts have been devised, adopted, adapted, and abandoned. The Unicode Standard, the computing industry reference that encodes characters for digital text, currently includes more than one hundred and fifty scripts, covering most living writing systems and a great many historical ones. Scholars at the Harvard Department of Linguistics, the MIT Department of Linguistics and Philosophy, and the Stanford Department of Linguistics have contributed significantly to the decipherment and analysis of ancient scripts.

Writing systems are typically grouped into several structural types. Alphabets, such as Latin, Cyrillic, Greek, Georgian, and Armenian, use separate symbols for both consonants and vowels. Abjads, including Arabic and Hebrew, represent primarily consonants and leave vowels partly or entirely unmarked. Abugidas, the dominant writing type of South and Southeast Asia and parts of Africa, include Devanagari, Bengali, Tamil, Thai, and Ethiopic script. Each consonant in an abugida carries an inherent vowel that can be modified by diacritics. Syllabaries, such as Japanese hiragana and katakana and the Cherokee syllabary devised by Sequoyah, assign one symbol to each syllable. Logographies, most prominently Chinese characters used across Chinese, Japanese kanji, and historically Korean and Vietnamese, encode morphemes or words.

Some scripts remain undeciphered. Linear A of Minoan Crete, Rongorongo of Rapa Nui, and the Indus Valley script have resisted interpretation despite more than a century of effort. Others have been decoded in modern times: Egyptian hieroglyphs were unlocked through the Rosetta Stone, and Mayan glyphs were progressively deciphered through collaborative work in the late twentieth century. Contemporary preservation and digitization projects at institutions such as Harvard and the Smithsonian Center for Folklife and Cultural Heritage work to ensure that manuscripts in vulnerable scripts are recorded for future study.

A small set of languages carries much of the world's daily communication, commerce, government, and science. Each of them is also the primary language of at least one large state and has an extensive written literature. The profiles below draw on Ethnologue figures for native-speaker populations and on institutional descriptions from bodies such as Instituto Cervantes for Spanish, the Confucius Institute Headquarters (Hanban) for Mandarin, the Organisation internationale de la Francophonie for French, and the British Academy for English and the history of its study.

English functions today as the principal global lingua franca of business, science, aviation, diplomacy, and the internet. Approximately 390 million people speak English as a first language, primarily in the United States, the United Kingdom, Canada, Australia, New Zealand, Ireland, and other Commonwealth countries, while more than a billion additional people use it as a second or foreign language. Mandarin Chinese is the single largest language by native speakers, with approximately 940 million first-language speakers concentrated in China and in Chinese-speaking communities abroad. Spanish, with approximately 485 million native speakers, is the official language of Spain, Mexico, and most of Central and South America, and a co-official language in the United States of the federal government's public communication.

Hindi, with approximately 345 million native speakers in India, is one of twenty-two scheduled languages of the Indian constitution. Arabic is notable for its diglossia: Modern Standard Arabic is used in formal writing, broadcast media, and across the Arab world, while a cluster of regional varieties, mutually intelligible in varying degrees, serves daily speech in Saudi Arabia, Egypt, Morocco, Iraq, and elsewhere. French, Russian, Portuguese, Bengali, and Japanese round out the most widely spoken languages, each anchored to a major nation or group of nations and each with a substantial literary tradition. Smaller but regionally dominant languages, including Indonesian, Turkish, Korean, and Swahili, serve as lingua francas across wide areas of Asia and Africa.

A language is considered endangered when children are no longer learning it as a mother tongue or when its speaker community is shrinking, aging, or shifting to a dominant language. The UNESCO Atlas of the World's Languages in Danger estimates that roughly 40 percent of living languages are endangered to some degree. SIL International, which compiles Ethnologue, and scholars affiliated with the Language Documentation and Conservation journal at the University of Hawai'i use the Expanded Graded Intergenerational Disruption Scale, known as EGIDS, to categorize vitality from institutional use through dormant and extinct. UNESCO uses a five-level scale of vulnerability ranging from vulnerable through severely endangered and critically endangered to extinct.

The forces behind language endangerment are well documented. Globalization and economic integration concentrate opportunity in dominant languages. Formal schooling in national languages often excludes or stigmatizes local ones. Urbanization draws rural speakers into cities where their first language is rarely used, and mass media reinforces the prestige of a small set of widely spoken languages. Historical colonization in the Americas, Oceania, Africa, and parts of Asia suppressed indigenous languages, and the after-effects are visible today. The Linguistic Society of America has published guidance for researchers working in partnership with speaker communities, and the Endangered Languages Project, a University of Hawai'i and Google partnership, coordinates documentation, recording, and community resources.

Revitalization is possible, and the historical record contains some striking examples. Maori in New Zealand has been supported since the 1980s by immersion preschools known as kōhanga reo, by broadcast media, and by formal recognition as an official language. Hawaiian has been revived through Hawaiian-medium schools and university programs. Welsh, Catalan, and Basque all benefited from bilingual education and official status in the late twentieth century. Irish is taught across the Irish school system and is an official language of the European Union. Hebrew is the most dramatic case: it was revived from a language of religious and scholarly use into a fully spoken daily language for millions of people in Israel during the twentieth century. Digital archives such as the Rosetta Project and the Endangered Languages Project preserve recordings for languages where daily use has ceased.

When speakers of different languages need to communicate over time, they sometimes develop simplified contact languages called pidgins. A pidgin has a reduced vocabulary, a simplified grammar, and typically no native speakers: everyone who uses it also uses another language. When children grow up speaking a pidgin as their first language, it expands and regularizes into a creole, a full natural language with the same expressive capacity as any other. Haitian Creole in Haiti, Jamaican Patois in Jamaica, Papiamento in Aruba and Curacao, Tok Pisin in Papua New Guinea, Bislama in Vanuatu, and Afrikaans in South Africa are all widely spoken languages that developed out of contact situations, though Afrikaans is sometimes described as a daughter of Dutch rather than a creole in the strict sense.

Lingua francas are languages used for communication across groups that do not share a first language. They have existed for as long as there have been trade routes and multilingual polities. Latin served as the lingua franca of scholarship, liturgy, and diplomacy across medieval Europe. Sanskrit played a similar role across large parts of South and Southeast Asia. Arabic followed the trade and scholarly networks of the early Islamic world. Swahili emerged along the East African coast as a trade language and is now spoken across much of eastern and central Africa. Malay served maritime Southeast Asia and is the basis of modern Indonesian and Malaysian. English is today's most widespread lingua franca, used routinely in international business, science, air traffic control, and on the internet. Its role is tracked closely by the British Academy and by European Commission research on multilingual policy.

Sign languages are fully natural human languages that use visual and gestural signs rather than sound. They are not manual encodings of spoken languages. They have their own grammar, vocabulary, and community norms, and they develop, change, and diversify in the same ways as spoken languages. Research at Gallaudet University, the Linguistic Society of America, and university linguistics departments at MIT, Stanford, and other institutions has produced an extensive literature on sign-language phonology, morphology, and syntax. Many countries recognize their national sign language as an official or protected language.

The World Federation of the Deaf recognizes more than three hundred sign languages in active use around the world. Among the best studied are American Sign Language, used in the United States and much of English-speaking Canada; British Sign Language in the United Kingdom; French Sign Language in France; Chinese Sign Language in China; Japanese Sign Language in Japan; and Indian Sign Language in India. Historical relationships between these languages do not track those between the surrounding spoken languages: American Sign Language, for example, is derived largely from French Sign Language rather than from British Sign Language. International Sign is a contact signing form sometimes used at multinational events such as meetings of the World Federation of the Deaf, and functions somewhat like a pidgin across sign-language communities.

The monolingual adult is a minority of the world's population. Scholars estimate that the majority of people speak at least two languages at a functional level, and in many regions three or four are routine. Papua New Guinea, with roughly 840 languages, is perhaps the most linguistically diverse country on the planet, and individuals there often speak a local vernacular, the lingua franca Tok Pisin, and English. India and Indonesia are similarly multilingual, with hundreds of regional languages alongside national languages. Nigeria has approximately 525 languages alongside English as the official language of administration. South Africa recognizes eleven official languages, a reflection of the country's broad linguistic diversity, to which South African Sign Language was added in 2023 to bring the total to twelve.

Research at academic centers including the Max Planck Institute for Evolutionary Anthropology has documented cognitive, social, and economic advantages associated with multilingualism, including flexibility in problem-solving, stronger executive function in some tasks, and expanded educational and economic opportunity. Government approaches range widely. France has historically promoted the French language as a unifying national institution, a policy reinforced by the Organisation internationale de la Francophonie and by the Académie française. Switzerland manages four national languages through a cantonal model. Canada is officially bilingual in English and French at the federal level. The European Commission operates in twenty-four official languages and funds extensive language education programs, and the European Charter for Regional or Minority Languages protects additional historic languages such as Catalan, Welsh, and Sami.

All languages change continuously. Sounds shift, grammatical patterns are reorganized, words are borrowed from neighbors, and new vocabulary arises to describe new technologies and ideas. The Great Vowel Shift, a sweeping change in the pronunciation of English long vowels that took place roughly between 1400 and 1700, is a well-studied example of how substantially a language can change in a few generations. Grammaticalization, the process by which full lexical words gradually become grammatical markers, can be observed in many families. Language death, when a language is no longer actively spoken, is also part of the larger history of change, as is language birth, where a new language emerges from long contact, from creolization, or from the divergence of dialects.

Historical linguists reconstruct earlier stages of languages through the comparative method, systematically aligning vocabulary and sound correspondences across related languages to infer ancestral forms. Proto-Indo-European, the unattested ancestor of most European and many South Asian languages, has been reconstructed in considerable detail from comparative evidence by scholars at Harvard, the University of Cambridge, and other institutions. Proto-Austronesian, Proto-Bantu, and Proto-Sino-Tibetan are among the other reconstructed proto-languages. In the modern era, flows of loanwords follow the movement of technology, science, and popular culture: English has absorbed thousands of words from other languages over its history, and in turn is the source of many loanwords into Japanese, Korean, French, and almost every other widely spoken language, in domains ranging from computing to cuisine.

The past fifteen years have transformed what machines can do with human language. Neural machine translation, which replaced earlier statistical and rule-based systems during the mid-2010s, now powers everyday translation services including Google Translate, DeepL, Microsoft Translator, and the proprietary translation systems embedded in many consumer products. Speech recognition and speech synthesis have become usable for dictation, accessibility, voice assistants, and broadcast captioning in the largest languages. Large language models have further improved fluent cross-language output and have brought usable translation to many more language pairs than were previously supported.

The benefits of these advances are unevenly distributed. Models perform best on languages that have large digital text corpora available to train them; the United Nations archive, major news organizations, and parallel corpora of books and websites have supplied the bulk of that training data. So-called low-resource languages, which includes the majority of living languages, are comparatively underserved. Recent initiatives aim to address this gap. Meta's No Language Left Behind project has released open-source translation models for more than two hundred languages. Masakhane, a grassroots research network focused on African languages, has built open corpora, benchmarks, and models for dozens of languages from the continent. The Endangered Languages Project hosts community-contributed materials that can support small-scale corpus building. Research at academic centers including Stanford, MIT, and the Max Planck Institute explores AI-assisted documentation, speech transcription, and revitalization tools that can be deployed in partnership with speaker communities.

CountryReports publishes its country profiles, global content, articles, and reference features in five languages: English, Spanish, Chinese, Hindi, and French. Together these five languages are the first language of well over two billion people and are studied as second or foreign languages by hundreds of millions more. The five-language strategy is intended to give the widest possible global readership direct access to the site's educational content, while also fitting the school, university, and institutional markets where CountryReports is most widely used.

Users can switch the interface language at any time from the language menu in the site header. Freely accessible content is available in all five languages without a membership, and content that is part of the paid tier follows the entitlements described in the site terms of service and the related language availability policy. Additional language coverage is considered periodically based on audience data, and community feedback on translation quality is welcomed through the site's contact page.

Detailed citations, data references, and institutional sources for all CountryReports content are listed on the Sources page. The following international organizations, reference catalogs, and academic institutions are the primary authorities we rely on for content about the world's languages. Each link points to the institution's homepage.

International Organizations and Reference Catalogs

• UNESCO Atlas of the World's Languages in Danger — Vitality classifications, endangerment indicators, and country-level data on threatened languages worldwide.
• UNESCO — Programs on linguistic diversity, mother-tongue education, and intangible cultural heritage.
• European Commission — Multilingualism Policy — Policy framework for official and regional languages across the European Union.

Academic Research Centers and Journals

• Max Planck Institute for Evolutionary Anthropology — Research on linguistic typology, historical linguistics, and the Glottolog language catalog.
• Linguistic Society of America — Professional society for linguistics in North America, with resources on endangered-language documentation, ethics, and public education.
• Language Documentation and Conservation (University of Hawai'i) — Open-access academic journal on the documentation and revitalization of endangered languages.
• Harvard Department of Linguistics — Research and graduate training in historical linguistics, phonology, syntax, semantics, and fieldwork.
• MIT Department of Linguistics and Philosophy — Theoretical and experimental linguistics, including sign-language research and computational linguistics.
• Stanford Department of Linguistics — Sociolinguistics, language variation, computational linguistics, and fieldwork on under-documented languages.
• British Academy — United Kingdom national academy for the humanities and social sciences, with programs on language learning and linguistic research.

Cultural and Language Institutes

• Smithsonian Center for Folklife and Cultural Heritage — Documentation of living cultural traditions and languages, including the Mother Tongue Film Festival.
• Organisation internationale de la Francophonie — Intergovernmental body for French-speaking countries, promoting the French language and linguistic diversity.
• Instituto Cervantes — Spanish Government institute promoting the Spanish language and Hispanic cultures worldwide, including the annual El español en el mundo report.
• Confucius Institute Headquarters (Hanban) — Chinese Government institute supporting the study of the Chinese language and culture internationally.

If you notice an error or would like to suggest a correction, please use our contact page to get in touch.