The Last Native Speaker
On the death of languages, the cognitive monoculture, and what civilization loses when the last grammar of a world goes silent
Friend,
I want to tell you about a woman named Marie Smith Jones.
Marie Smith Jones lived in Anchorage, Alaska. She was born on May 14, 1918, in the Eyak village of Cordova on Prince William Sound. She died on January 21, 2008, at the age of eighty-nine. Her obituary in the New York Times described her as a tribal chief, a great-grandmother, an environmental activist, and the last living native speaker of the Eyak language. The first three descriptions, while accurate, were not what made her death worth a national obituary. The fourth description was. When Marie Smith Jones died, the Eyak language died with her.
I want to be precise about what that means, because it is the kind of statement that can be received as either smaller or larger than it actually is.
Eyak was the language of a specific people, the Eyak, who had lived on the southern coast of what is now Alaska for at least the previous several thousand years. The language was related to the broader Athabaskan language family that includes Navajo and Apache, but was distinct from any of its relatives at a depth that linguists estimate at three to four thousand years of separate evolution. It contained words for relationships and natural phenomena that no other language in the world had. It encoded a particular way of slicing up reality — a particular conceptual geography, a particular map of kinship, a particular way of speaking about time and motion and space — that had developed in continuous use over millennia and that had no equivalent anywhere else.
When Marie Smith Jones died, that conceptual geography ceased to be the daily mental architecture of any living human being. Some of the words were, by then, recorded. Linguists from the University of Alaska Fairbanks had been working with her for decades to document what they could. Audio recordings exist. Grammars exist. A dictionary exists. But the language as a living thing — as the medium through which a human mind organizes experience and through which a human community constructs shared meaning — that was gone.
You could, in principle, learn Eyak today from the documentation. A few people have made the attempt. They can produce sentences that approximate what Marie Smith Jones would have said. But the language, as it lived in her — with the rhythm, the connotations, the ten thousand undocumented decisions about how this idea connects to that idea, the entire substrate of meaning that exists only in fluent speakers — is not in the documentation. Cannot be in the documentation. Documentation captures the surface of a language. The interior dies with the last speaker.
This letter is about what civilization loses, on a scale I do not think is widely appreciated, when languages die. It is about why I have come to think the global wave of language extinction now underway is one of the most underrecognized civilizational events of the present moment. And it is about what anima mundi, the soul of the world, looks like when read through the lens of the world’s vanishing tongues.
There are roughly seven thousand languages spoken on Earth as of 2026. By the most credible projections — those of Lyle Campbell at the University of Hawaii, those of David Crystal at Bangor University, those of the Endangered Languages Project consortium — between half and ninety percent of these languages will be gone, in the sense Eyak is gone, by the end of this century. Roughly one language is dying every two weeks.
We are, in other words, in the midst of an extinction event whose scale and pace, in linguistic terms, has no precedent in human history.
