Language isolates, like Chimané from Bolivia, are unrelated to any other known tongue. Studying them is revealing how languages evolve and influence our perception of the world around us.
LAST February, amid the fjords of southern Chile, an elderly woman died – and a language fell silent. Cristina CalderÓn, a much-loved 93-year-old, was the last known native speaker of Yaghan, which could at one time be heard across the Tierra del Fuego – the Land of Fire – that forms the jagged tip of South America. The loss of any tongue is a tragedy, but Yaghan’s extinction will be felt particularly keenly because this was no ordinary language. It was an “isolate”: a language utterly distinct from those used anywhere else in the world.
 |
| The Chimané people have provided insights into the power of number words REUTERS/David Mercado |
Language isolates comprise about 200 of the estimated 7400 languages in use today and many are dangerously close to following Yaghan into oblivion. Estimates suggest that 30 per cent of all languages will have vanished by the end of the century. Isolates – some used by just a few hundred people – are particularly vulnerable.
But as their vulnerability has risen, so has an awareness that isolates can tell us a lot about human communication and cognition. In the past few years alone, they have offered us fresh insight into the interplay between cultural and linguistic evolution and provided support for a controversial hypothesis that links our understanding of reality with the language we use. “Each of these isolates is a… whole different window on the mind,” says Lyle Campbell at the University of Hawai’i at Mānoa.
What’s more, there is new hope that the research might also identify better strategies to help us save them from extinction.
Broadly told, the story of language is a family affair. The similarities between Danish and Dutch, or Swahili and Zulu, speak to a world where groups of languages are far from isolated. Instead, they fall into roughly 140 common lineages.
Isolates are different. Instead of boasting gaggles of siblings, these languages stand totally apart. Probably the most evocative example is Basque, spoken by a community straddling the border of France and Spain. While vin is a favourite French tipple and Spaniards may enjoy a glass of vino with their dinner, the Basque equivalent is ardoa. And while farmers in Spain and France engage in agricultura or agriculture, their Basque counterparts prefer nekazaritza.
Having such unusual vocabulary compared with their surrounding languages can give isolates an air of the exotic, and foster a sense of pride and resilience among speakers. This may help explain why Basque has survived for millennia – a bronze tablet in the shape of a hand found in 2021 shows that an ancestor of modern Basque was being written down 2000 years ago, while Basque itself may have a history stretching back 5000 years or more.
Rare languages
Truth be told, however, isolates are more conventional than they might appear. Many were once members of larger families, but have been left isolated because those other languages vanished – usually through a blend of cultural influences, conquest and simple luck. Ket, for instance, a language isolate of Siberia, was once part of the small Yeniseian language family. It is now alone, potentially in part because of the spread of the Russian language following Russia’s conquest of Siberia between the 16th and 18th centuries.
But, paradoxically, it is precisely because isolates are relatively conventional that they are so linguistically valuable. Each is a unique experiment that can cast new light on the solutions people have found for effective communication.
“There’s so much diversity in how humans construct a cognitive world with their languages,” says Anna Belew at the University of Hawai’i at Mānoa, who is also programme manager at the Endangered Languages Project. “And isolates, just by virtue of not having sister languages, tend to hold fairly unique ways of encoding the world.”
Decode them and we gain a richer understanding of the rules of language. Kutenai, for instance, is an isolate spoken by the Kutenai people, an Indigenous community straddling the US-Canada border. It is one of the few languages in the world with a grammatical “fourth person”, making it possible to fully clarify the meaning in phrases like “he held his hat”. Kusunda, spoken in Nepal, lies at the opposite extreme. It is missing some of the grammatical features found in most languages, such as words like “no”.
Isolates are also challenging assumptions about the fundamentals of languages and language evolution – as revealed by Al-Sayyid Bedouin Sign Language (ABSL).
Used by a Bedouin community living in Israel’s Negev desert, ABSL arose in the first half of the 20th century, improving communication in a population where, through quirks of genetics and social organisation, the rate of deafness is 50 times the global average. But what really makes ABSL special is how its words are formulated. Practically all languages rely on what is known as duality of patterning, whereby users combine a relatively small number of meaningless sounds or gestures to make a rich and extensive vocabulary. For example, essentially all of the words in the English language are constructed from a few dozen fundamental sounds, each represented by one or more letters of the alphabet.
The alternative – a language in which every word is unique and can’t be broken into smaller, repeatable units – would be bewilderingly complex. In such a hypothetical spoken language, it might require people to remember that a certain timbre of hiss means breakfast, while mild squawking means bathroom. Yet that is precisely what ABSL does. Unlike all spoken languages, and indeed other sign languages, users must memorise a new and distinct gesture for every word they want to express. ABSL shows that duality of patterning isn’t the universal feature of language we assumed it to be, and it also offers us clues as to why.
 |
| Cristina Calderón was the last speaker of Yaghan Víctor Alejandro Correa Rueda CC BY 2.0 |
In a 2021 study, Simon Kirby at the University of Edinburgh and Monica Tamariz at Heriot-Watt University, both in the UK, compared ABSL to an unrelated but similarly young sign language – Israeli Sign Language – which does exhibit duality of patterning. They noted that Israeli Sign Language, like many formal sign languages, is usually learned by groups of students in schools. The two researchers reasoned that students interacting informally with one another – in the playground, for instance – would find and use linguistic features such as duality of patterning, because all students of a new language tend towards simplification. As other students adopt these simplifications, they take root and ultimately spread through the community.
Child talk
New ABSL learners, on the other hand, typically pick it up through exposure to the language of elders. The learners may still come up with innovative ways to simplify the language. But older signers are unlikely to adopt what they might perceive as child talk. The result is that simplifications don’t spread, and the language remains complex.
Kirby and Tamariz tested this idea with a simple computer simulation, containing a four-word “language” being used by a small, artificial population. If hypothetical students were taught by the oldest members of the population, who themselves were expert language users, it took more than 50 generations for duality of patterning to emerge. If, however, learners were taught by a younger member who was still picking up the language, duality of patterning emerged in just four generations.
The work suggests that language evolution can be shaped to a striking degree by the structure of a particular society. More specifically, Tamariz says the research identifies new learners as a powerful force in shaping languages.
One of the other big questions linguists are grappling with is the interplay between language and cognition. Some researchers argue that the way we communicate influences our perception of the world and our understanding of abstract concepts such as numbers, an idea known as the Sapir-Whorf hypothesis. This controversial idea has its critics, but language isolates might provide crucial evidence in its favour.
The speakers of one isolate in particular often feature whenever linguists debate the Sapir-Whorf hypothesis. The Pirahã people, an Indigenous group living in northern Brazil, are famous among researchers for speaking a language – also called Pirahã – that some research suggests has no words for numbers. What’s more, studies suggest that Pirahã speakers struggle to count, implying that words for numbers are indeed necessary to do maths.
But there is a problem with that conclusion, according to Benjamin Pitt at the University of California, Berkeley. Pitt says that researchers have “implicitly or explicitly” compared the Pirahã speakers’ understanding of numbers with the number knowledge of people in higher-income countries. He says this is hardly a fair comparison, given that the way of life in a country like the US is wildly different from that in the Amazon.
To really understand whether number words are necessary for counting, Pitt says we need to compare people who either do or don’t know such words, but who still share a common culture. Elsewhere in the Amazon, Pitt and his colleagues found an Indigenous community of such people.
The Chimané people live in lowland Bolivia. Their language, Chimané, is another isolate and it does contain number words. Crucially, however, knowledge of those words varies across the society: some Chimané can count indefinitely, while others stumble at 10 or 12.
Pitt and his colleagues worked with 30 Chimané people, only 15 of whom were confident counters, and asked them to complete some counting tasks. For instance, volunteers were presented with a row of pebbles and asked to produce their own row containing the same number of pebbles. Most volunteers could do so reasonably accurately when told to make their row parallel to the row already on the table – perhaps, says Pitt, because they could simply match up the pebbles without having to count them.
In the past, some researchers have argued that people can use a range of different cognitive tools to count, downplaying the importance of language. The research by Pitt and his colleagues suggests this idea is wrong and language is crucial. Especially in the realm of numbers much larger than 4, Pitt says Chimané shows that language sometimes offers people “new conceptual abilities” – which may otherwise be unavailable to them.
But whatever gems they can offer linguists, isolates are, by definition, vulnerable (see “Tongues in trouble”, below). Unlike other languages, whose insights may be preserved in related tongues, isolates have no family to fall back on. “We really need to get information on the isolates,” says Campbell, “because if we lose them, that information is gone forever.”
Learning Basque
Yet it is just possible that the very act of studying isolates might improve our ability to save them. Ane Ortega-Etcheverry at the Begoñako Andra Mari Teacher Training University College, Spain, knows from experience that isolates can survive against the odds. A child of Francisco Franco’s dictatorship – which, for decades, waged a war against Spain’s minority tongues – she had scant opportunity to learn Basque while growing up. But after embracing her ancestral tongue as an adult, Ortega-Etcheverry devoted years to exploring why it is so resilient to extinction.
In a 2014 study, she and her colleagues interviewed a range of non-native Basque learners, uncovering what she calls a “continuum” of attitudes among participants. At one end were so-called euskaldunberri, or new Basque speakers, including those being taught the language at school. At the other end were euskaldun, or proficient Basque speakers.
Ortega-Etcheverry and her colleagues were surprised to find that euskaldunberri becoming fully-fledged euskaldun had less to do with technical proficiency and more to do with culture. To put it another way, being taught to speak Basque fluently is less important than having family and friends familiar enough with the language to use it in conversation. Some felt that the formal Basque they studied in school made them less euskaldun than learning a rugged local dialect.
Ortega-Etcheverry is already spreading the word from her research. She is now working with speakers of Nasa Yuwe, an isolate from Colombia that, with about 60,000 speakers according to some estimates, is classified as vulnerable. “School is not enough,” says Ortega-Etcheverry, arguing that if endangered languages are to thrive, activists need to create a welcoming cultural atmosphere.
Get the strategy right and Ortega-Etcheverry thinks we can turn the tide and help vulnerable language isolates survive. It might just ensure that when older community members like Cristina CalderÓn die, the languages they speak live on.
Tongues in trouble
Pello Salaburu at the University of the Basque Country describes language isolates as “black swans” – outliers that force scholars to challenge their preconceptions around cognition (see main story). But many isolates may soon vanish.
Al-Sayyid Bedouin Sign Language was recently classified as “definitely endangered” by researchers at the University of Central Lancashire, UK, working with UNESCO and the Foundation for Endangered Languages. It is thought to be used by no more than 4000 people in Israel. Chimané, with about 4800 speakers in Bolivia, is considered “vulnerable” by the Endangered Languages Project. Some isolates are at even greater risk: Kusunda, from Nepal, has a single surviving fluent speaker.
Other isolates have more speakers, but are still threatened – and their disappearance would result in the loss of fascinating linguistic features. Purépecha, for instance, has about 117,000 speakers in Mexico and boasts a distinctive numerical system. While it is common for languages to make some numbers above 10 by combining two earlier numbers, such as “twenty-one”, Purépecha is unusual in using this approach for the numbers 7, 8 and 9 too.