We are lucky to know anything at all about the Ubykh language. In the 1800s, tens of thousands of people spoke it on the Black Sea coast. When Russia conquered the region, the Ubykhs resisted until they were forced into exile in the Ottoman empire. Transported thousands of miles by a traumatised community now scattered across Turkey, Ubykh survived until 1992 when its last fluent speaker died. It was one of at least 244 languages that has become extinct since 1950, and soon – unless anything changes – my grandmother’s language will have joined them.

Over the next 40 years, language loss has been predicted to triple without intervention. Yet we hear about language endangerment far less often than we hear about other wounding losses to our planet’s diversity or history. Deforestation in Costa Rica is being reversed following the realisation of the enormous natural and scientific resource that may disappear with its trees. International archaeologists rallied to preserve and restore ancient remains in Syria following the destruction wreaked by Islamic State. But the efforts of those labouring to document or preserve minority languages are rarely celebrated.

The databases that do exist, such as Ethnologue, chart unfathomable cultural riches contained within more than 7,000 known living languages. But a staggering 44% of these are now classed endangered, many of them with fewer than 1,000 speakers left. One-nation-one-language narratives lull us into assuming France speaks French, China speaks Mandarin; this ignores the tens and even hundreds of regional languages, many of whose speakers have experienced everything from active persecution to bans in school to simply feeling stigmatised for speaking their mother tongue.

Some communities are lucky enough to have the political or cultural autonomy to protect their languages – think of Welsh or Māori – but many aren’t so fortunate. Some rue and rally; others resign themselves to decline, not because they’ve actively chosen to abandon a language, but because maintaining it in the face of a more dominant one takes enormous huge resolve and resources.

Often it is linguists that are on the frontlines – people such as Georges Dumézil, who doggedly sought out Ubykh, a rumoured Caucasian language with an incredible number of distinct sounds. Decades of search finally led him to Tevfik Esenç, who had been raised by Ubykh-speaking grandparents. Their partnership is how we know Ubykh has more than 80 consonants and just three vowels, a ratio placing it at the very edge of language evolution – and an important addition our understanding of the sheer variety of human communication.

The study of endangered languages often reveals that Indigenous peoples identified and classified flora and fauna, from tubers to species of dolphin, long before western science encountered them. Many have extensive vocabularies connected to traditional practices that are equally at risk; in some cases, linguists have arrived just in time to record these, interviewing elders before they pass away.

Documenting languages is important not least because it means communities can better revive them if they eventually choose to do so. In my wider work investigating linguicide – the deliberate erasure of a language – it is clear that linguistic and human rights often go hand in hand. The displacement and disempowerment of Indigenous people across the United States unfolded alongside the loss of a dizzying variety of languages; attempts by communities to reclaim and celebrate their heritage often focus on language revival. Why does this matter? In Canada, research showed that among groups where more than half could maintain a conversation in their native language, youth suicide rates were low to absent, whereas they were six times higher in groups where that wasn’t the case. A language alone does not save a community from poor mental health, of course, but it may be an indicator of the cultural resilience that does. In 2012, a government inquiry in Australia found that Indigenous languages played such an important role in communities’ health and life expectancy that it argued they should be recognised in the constitution. Some 14 years later, the constitution still only recognises English. In Europe, instruments such as the Charter for Regional Or Minority Languages promise better protection, although many countries have not ratified it, including France and Italy.

All this is taking place against a backdrop of homogenisation – with major languages such as English, Mandarin and Spanish dominating (according to Ethnologue, 88% of the world’s population is a native speaker of one of only 20 languages). Linguists have observed that migrants tend to become monolingual in their adopted country’s language by the third generation.

I have seen this effect first-hand. I grew up only understanding, not speaking, the glorious soundscape of standard Italian and “dialët” from the mountains of Piacenza that my Nonna and Mum spoke. It had been so devalued in Italian public life that’s the only name she ever had for it: a dialect of Italian. It’s actually a variety of Emilian called Piaśintein, a descendent of vulgar Latin. In the north, transmission to children has basically stopped and so it can feel like an artefact from the past. Yet following my Nonna’s death, weaving it into conversation with my Mum is a way of keeping a part of her alive.

But not just her – the unique time, place and culture it represents; the fronted vowel sound ø, which can sound to outsiders more Scandinavian than Italian; the nature words, especially those for i funz, the valley’s famous mushrooms. And much else besides.

From Ubykh to Piaśintein, language documentation holds out hope, at least, for revival. For others – Australia’s Walangama, Argentina’s Abipón – the little that survives may never be enough. Who can say what we have lost in their now disappeared inventories of words for plants or animals, or in their wise sayings? As we speak, there are activists demanding legal and cultural recognition for thousands of endangered languages. We should listen to them before it is too late.

Sophia Smith Galer is a journalist and author of How to Kill a Language (William Collins). To support the Guardian order your copy at guardianbookshop.com. Delivery charges may apply.

Further reading

Rare Tongues by Lorna Gibb (Atlantic, £12.99)

Proto by Laura Spinney (William Collins, £10.99)

Through the Language Glass by Guy Deutscher (Arrow, £10.99)