Of the roughly 7,000 languages spoken today, UNESCO warns that a large proportion are endangered, with many likely to disappear within this century as their last fluent speakers age. This prospect provokes a powerful instinct to intervene, but the question of how far we should go is harder than the instinct suggests. By "save" I mean not one action but a spectrum — from cheaply recording a language to the costly, sustained work of restoring it to everyday use. This essay argues that we should document every endangered language as a matter of course, but attempt full revival only where the speaker community itself wants and drives it, because effort imposed from outside is both unlikely to succeed and ethically questionable.
The case for trying hard to save dying languages is strong. A language is not merely a set of labels but a unique repository of knowledge, history, humour and ways of perceiving the world; when it dies, that perspective is lost as completely as an extinct species, and often irreversibly. For the communities concerned, language is bound up with identity and dignity, and its loss can feel like a continuation of the historical injustice that frequently caused it, since many indigenous languages were actively suppressed under colonial rule. On this view, helping to preserve a language is a matter of cultural justice as much as of academic interest, and the international community has a real interest in maintaining the diversity of human thought.
Crucially, this effort is not futile, which answers the most common objection. The revival of te reo Māori in New Zealand — driven by community "language nests" for young children, Māori-language broadcasting and official recognition — has increased the number of speakers and learners and restored the language's public status. The revival of Welsh, through bilingual education and dedicated media, tells a similar story in a Western context. These cases prove that decline is not inevitable and that determined, well-resourced effort can reverse it.
Yet there are serious reasons for caution that a mature answer must concede. Full revival is extremely expensive and slow, and resources spent on it are resources not spent on health, education or poverty reduction; for a poorer country, the trade-off is real. A single shared national or global language can also bring genuine economic opportunity and help unite a diverse society, and speakers themselves frequently choose the dominant language for precisely these reasons. Where that is so, an externally driven revival can feel less like liberation than like an imposition by outsiders romanticising a language its own community has chosen to leave behind.
The way to reconcile these positions is to distinguish between documentation and revival. Recording a language — its grammar, vocabulary and oral literature — is comparatively cheap, preserves the knowledge permanently, and should be done universally, since the cost of losing the record entirely is so high. Active revival, by contrast, is justified only where the community itself wants it and is prepared to lead, because, as both the Māori and Welsh cases show, revival succeeds through the will of speakers rather than the enthusiasm of outsiders. This distinction allows us to honour the value of every language without committing limited resources to projects that cannot or should not succeed.
In conclusion, we should try to save dying languages, but the extent of that effort should be calibrated rather than uniform. Every endangered language deserves to be documented, because the knowledge it holds is irreplaceable and the recording is cheap; full revival deserves wholehearted support precisely where the community drives it, as the te reo Māori revival proves is possible. We should not, however, impose revival on communities that have chosen otherwise. The right answer is therefore neither "save them all at any cost" nor "let them go", but document universally and revive by consent.