“why don’t we just let all the languages die out?”, or some variant thereof, is the second question anyone asks me when I tell them I’m a linguist. (The first question is “How many languages do you speak?”) While that's very jarring, I understand where the notion is coming from. The purpose of language is communication. What would be more efficient for communication, having to navigate the seven thousand languages that currently exist or just a few?

This idea makes it particularly difficult for people outside of linguistics to see language endangerment and extinction as an emergency. Those more inclined towards social justice might realize that language is one of the strongest bonds that holds a culture together, and the fact that a language is only spoken by a small community does not make it less worthy of respect. However, the need to keep languages alive runs even deeper than altruism.

When there's no one left to talk to: A guide to endangered languages. Image 1.

Ben Macaulay



Ben Macaulay is a PhD student at the CUNY Graduate Center specializing in tonal phonology and morphosyntax

Linguistics, or the investigation of patterns within and across languages, has yielded findings that not only shed light on the details, but on other areas such as auditory/visual perception and cognition in general. These findings increase exponentially with the number of languages that are documented and allowed to thrive and develop into new language states.

Even if the languages we’ve lost had no surprising qualities, having continued access to speakers who can participate in studies and experiments is an invaluable resource. For example, despite being the most studied living language, English is still the subject of myriad experiments that continue to add to our knowledge about the language.

which languages are most critically endangered? Why don't we just figure it out and do something about it? Unfortunately it’s not that simple, for a few reasons. Nobody knows which languages are truly the most endangered; many of the most imperiled tongues are precisely the ones we know very little about. To label a language as endangered, we must first figure out what counts as a discrete language. Many of you reading this piece speak in a considerably different way than your grandparents; would you be justified to say that their version of English is an endangered language variety? Probably not. There is also not one single parameter that encompasses “endangered-ness.” Even Ethnologue (the Wikipedia for linguists) defines endangered-ness in two dimensions: the number of remaining native speakers, and the EGIDS level (a qualitative measure of how transmission of a language between generations has been disrupted by social or political factors). 

Keeping that in mind, let’s look at some examples of endangered languages and the different circumstances that put them at risk. To start is an example of a language with no tie to the linguistic community: not only does it lack the resources offered to documented endangered languages, but we actually don’t know whether there are still speakers or not.

endangered Languages

Consonants vs. vowels

Much of what we know about how people conceptualize consonants vs. vowels (“sonority”) is from working on Imdlawn Tashlhiyt Berber, a variety of Berber spoken in a single village in Morocco.



Much of how linguists approach prosody (how we use stress and tone in phrases) is due to the work done in the 1970’s on Yidiny, a language in Australia whose last speaker died shortly thereafter.

As this was the work of a single field researcher, and the findings were controversial, phonologists have been trying to reanalyze the Yidiny data ever since. Only recently has technology been able to perform meaningful reanalysis of Yidiny recordings. Had Yidiny, like many native languages of Australia, not been hunted to extinction, the scope of human knowledge about phonology and language, in general, would be significantly different.




Sri Lanka

When there's no one left to talk to: A guide to endangered languages. Image 2.

FAMILY: unknown (Wiki: Sinhala-based creole)





Sri Lanka has been colonized by numerous peoples in its history. In addition to the British, Dutch and Portuguese, the Sinhalese and Tamils that inhabit most of Sri Lanka today are also relatively recent inhabitants of the island.  Speakers of Prakrit (an Indo-Aryan language closely related to Sanskrit) came to Sri Lanka in the 5th century and made contact with the Veddah people.

While the Veddah’s language, as spoken at that time, is long gone, a creole based on this language was spoken by the Veddah at least through 1997, however reports from as far back as 1935 have noted the status of this language as endangered. Sinhalese and Tamil have both borrowed words from the Veddah language (such as the Tamil Komanam, ‘loincloth’).

Contact with Veddah is the reason that Sinhala, the Indo-Aryan language of Sri Lanka (related to Hindi, and more distantly, English), sounds radically different from the other members of its family. However, accessing the Veddah language is difficult. The Veddah people face stigma in Sri Lanka, and assimilation to dominant cultures there may have obscured any remaining speakers of the language. Wikipedia lists Veddah as a language that is extinct; Ethnologue lists its number of speakers as unknown, and the language as dying.

While many languages are in an undocumented state of limbo like Veddah, some languages with few remaining speakers have remained in close contact with the linguistic community. Wikipedia lists eight languages with a single (confirmed) remaining speaker, two of which are language isolates (languages that have no known relatives). These languages are particularly interesting as they often have unique features that cannot be investigated through related languages. The two such languages are Yagán and Kusunda.




Tierra Del Fuego, Chile

When there's no one left to talk to: A guide to endangered languages. Image 3.

FAMILY: language isolate



The Yaghan people live in Tierra del Fuego, the archipelago at the very southern tip of South America. Their language, Yagán, is down to one remaining native speaker, Cristina Calderón. With only one speaker, this language can no longer be used as communication. There are no conversations in Yagán, the last was in 2005 before the death of the second-to-last speaker, Cristina’s sister-in-law.

Yagán is well-documented. There have even been investigations into features like sound symbolism: grammars of Yagán mention things like the propensity of words describing curves to end in -m, (the way English has -mp in hump, bump, lump). There’s even an online dictionary.





When there's no one left to talk to: A guide to endangered languages. Image 4.

FAMILY: language isolate




Kusunda is a language once spoken in central Nepal. Now, most of the hundred or so Kusunda people speak the dominant languages of the region. Reports on the number of remaining Kusunda speakers vary: Wikipedia and a short documentary by Asia Calling report that Gyani Maiya is the last speaker of the language, however Ethnologue lists three. 

An interesting feature of Kusunda is in its phonology. Most languages in the region, such as Nepali and Hindi, get their distinctive sound from “retroflex” consonants, sounds like [t] and [d] made with the tongue curled backwards towards the palate (contrasted with “dentals” which touch the teeth). Kusunda only has one category for these sounds, and where it is made in the mouth depends on which vowel follows.

Some effort is being made to teach the language to Kusunda children, and linguists have been documenting the language as well. 



Ayapa Zoque

Tabasco, Mexico

When there's no one left to talk to: A guide to endangered languages. Image 5.

FAMILY: Mixe-Zoquean



You may know Ayapa Zoque by its Spanish name, Ayapaneco. This language was the subject of a Vodafone marketing gimmick and started a trend of endangered language thinkpieces starting with a 2011 article in The Guardian. While all agree that the cause of language revitalization is a noble one, some have criticized Vodafone for casting the revitalization effort through the lens of a “white savior complex.” Missing from their narrative about flying in a linguist to get the last two native speakers to stop being so fussy, was that the state of Ayapa Zoque is the result of an effort to suppress indigenous languages of Mexico that goes back hundreds of years. (And that Mexico has its own linguists.) In the linguistic community, this is not the attitude we take: speakers are the top priority, and revitalization efforts are on their terms.

Despite the controversy, this language revitalization effort is really the only one in the Americas that has garnered any mainstream attention recently. Similar, albeit less flashy, efforts are common in communities across North America. Even with the attention this language has received, it has been difficult to find any real linguistic information about it (aside from a word list). 




Brittany, France

When there's no one left to talk to: A guide to endangered languages. Image 6.

FAMILY: Celtic (Indo-European)



The number of remaining speakers is not the only measure of how endangered a language is. Breton is the worst-off of the four living Celtic languages, of which only Welsh is not classified as endangered. While having a speech community of 206,000 may look healthy, 75% of Breton speakers are over 65. The number of speakers is in rapid decline as the younger generations have not acquired Breton from their elders. Irish, another Celtic language, is in dire straits despite the efforts of the Irish government to revive it. Without any real help or recognition from France, Breton is dying. It is probably the most endangered language that you can learn at home. Breton is a perfect example of language endangerment without a low number of speakers: if the language is not being passed down to the next generation, it will die whether it starts with 206,000 speakers or 20. 

Celtic languages have a number of unique properties that distinguish them from other branches of the Indo-European language family (which includes English). For example, they all have a phenomenon called “initial mutations” where the first sound in a word can change in order to show some grammatical information. For example, mamm means “mother” in Breton, but e vamm means “his mother.” What’s interesting about Breton is that many of the quirks of Celtic as a whole have been reverted under heavy contact with French. All other Celtic languages put the verb at the beginning of the sentence (even under contact with English), but Breton puts it in the second position as French does.

Due to its location in Western Europe and a resurgence in Celtic culture, there are efforts to revitalize Breton, through instruction and culture. There was a short-lived video series to learn basic phrases in the language and there is also a lot of music with lyrics in Breton from artists like Dan Ar Braz, who also sings in Scottish Gaelic, the runner-up in Celtic’s race to extinction.




Hokkaido, Japan

When there's no one left to talk to: A guide to endangered languages. Image 7.

FAMILY: Language isolate

SPEAKERS LEFT: Approximately 10


While completely unrelated, Japanese and Ainu share many features: the verbs come at the end of the sentence, they have pitch accent systems (where stress is denoted with a tonal melody instead of loudness), and they have very similar sound inventories. These similarities are most likely due to contact. Some ways that Ainu is different are that it allows consonants at the end of the syllable (Japanese only has [n]), and uses a construction called “noun incorporation” (like “bartend” instead of “tend bar”).




Schleswig, Denmark/Germany

When there's no one left to talk to: A guide to endangered languages. Image 8.



When we think of languages like Danish, we don’t think of them as endangered (and rightly so). However, is what we think of as Danish one language? Linguistics does not have an answer to to this question. The shortcut we use to determine whether two forms of speech are varieties of the same language (dialects) or separate “languages” is whether speakers of one can understand speakers of the other and vice-versa (mutual intelligibility). This creates a definitional issue: how can we define languages by the perception of speakers when speakers are all different? For example, English and American speakers of English can generally understand each other, however there are varieties of English in Scotland and Ireland that are easier to understand for other speakers in the UK. This is what is called a “dialect continuum”, and it is an approach many linguists take to classification.

Some linguists think of Danish, Swedish and Norwegian as a single language with a dialect continuum, while others break the Scandinavian group into even smaller languages. One such language is Jysk, spoken on the Jutland peninsula of Denmark, and northern Germany. It is highly divergent from Standard Danish and is often not understood by speakers of other varieties of Danish. A sub-variety of this language, Sønderjysk, spoken at the Denmark-Germany border is something I often hear when working with Danish speakers in New York. In Denmark, Jysk is often the subject of comedy skits (the tea set is a symbol of hospitality, an attribute associated with this community much like we do with the American South). But some speakers of Jysk do not find the state of their language to be a laughing matter. Many of them consider themselves bilingual, and Jysk to be a completely separate language from Standard Danish. Poets like Vagn Rasmussen publish poetry in Jysk.  

Danish is infamous for having a large inventory of vowels and sonorants (vowel-like consonants). Jysk is an even more extreme case, reducing many words to single vowels. A well-known Jysk tongue-twister, a æ u å æ ø i æ å, means “I am on the island in the stream” and contains no consonants. 

Because the transition from dialect to separate language is gradual and often goes unnoticed (if it is indeed a legitimate transition in classification), the need to protect endangered language varieties is not always apparent. Some Jysk speakers are outspoken and ask for recognition, but so far the language has gotten little attention aside from a dictionary. There has been no count of Jysk speakers.





When there's no one left to talk to: A guide to endangered languages. Image 9.

FAMILY: BANZSL family of sign languages


Sign languages are an interesting case from the perspective of language endangerment: while there are cases of predominantly hearing communities using signed languages, the survival of a signed language is sensitive to the presence of hearing people in its community. Sign languages often arise spontaneously in communities with high percentages of deaf people (“village sign languages”) and often thrive and develop into more complex languages, as all languages do. A rise in the percentage of the community that is hearing can endanger the sign language, as hearing people may not see the need to learn the signed language. 

Auslan (an acronym for Australian Sign Language), is the sign language local to Australia. It is not classified as endangered yet, however members of the Deaf community in Australia have raised concerns about Auslan in the face of Australia’s policies on cochlear implants. Cochlear implants attach directly to auditory nerves in order to allow for the perception of speech in ways that other hearing aids do not, however there is a push to have children implanted early as there are worries that implants later in life will not allow implantees to properly hear speech. Deaf communities like the one in Australia see the implantation of cochlear implants in children as invasive and unnecessary, and do not appreciate the implication that lives without hearing are pathological and must be fixed. As of 2004, 80% of profoundly deaf children in Australia are given cochlear implants. With the increased ability for deaf Australians to assimilate into hearing culture, there are worries that many of them will turn away from Auslan.

Many users of Auslan are active on the internet, where the ease of creating and uploading video has allowed users of signed languages to communicate and share media in ways previously impossible. An example of this is on YouTube, where many users of signed languages upload versions of (hearing) songs that incorporate sign, either in place of vocals or as a full music video. This example is in Auslan.

Pink — Dear Mr President in Auslan

A sign language music video
© Show and Tell Productions


What can be done to save these languages?

The reasons for endangerment are not the same for every language. Some languages are threatened by a low speaker count, or changes to the makeup of the community. Others are in contact situations where it is of greater benefit for younger people in the community to use the dominant language of the region. Some aren’t even recognized as their own language and are thus not considered as targets for documentation or revitalization efforts. Many languages have the disadvantage of being outlawed in the country they are native to. 

The problem of language endangerment is probably never going to be completely solved. If anything, we are still feeling the loss of countless languages (and entire language families) that we failed to protect. What can be done now is on two fronts. The first is field work: linguists reach out to communities that speak endangered languages and help them document and revitalize the language in a way that is sensitive to their specific needs. The second is the speakers themselves. For example, Irish continues to dwindle even when the government awards stipends to Irish speakers of a certain proficiency. If Irish continues to be “a thing of the past” in people’s minds, if Irish doesn’t permeate home life and elevator chat and Instagram captions, then it will die. Not all people who speak endangered languages realize that there’s a problem either. Some know that their language is endangered but don’t know that it’s possible or worthwhile to do anything about it.

Many languages suffer endangerment because of modernization: people move out of small communities to find opportunity in the city, and their first language is no longer an effective tool for communication. The upside of this is that now a lot of work on these languages can be done from the city. New York is home to speakers of over 800 languages, many of which are at risk. Non-profits like the Endangered Language Alliance document these languages and help them thrive in the small enclaves of speakers that form within the city. For example, holding literacy classes or creating children’s books to help pass on languages like Mixtec (Oto-Manguean, Mexico) and Garifuna (Arawakan, Honduras/Belize) in the new communities of speakers that have arisen in the outer boroughs.


Linguists, field work

Linguists reach out to communities that speak endangered languages and help them document and revitalize the language in a way that is sensitive to their specific needs.


Speakers, communication

Not all people who speak endangered languages realize that there’s a problem either. Some know that their language is endangered but don’t know that it’s possible or worthwhile to do anything about it. It is.


cover image via shutterstock.com