There is a cave near Waitomo, on the North Island of New Zealand, that has been sealed for a million years.
The sealing was not gradual. It happened twice, in two distinct catastrophes, and the second one was what made the first one legible. An eruption 1.55 million years ago laid down a layer of volcanic ash. A far larger eruption — the Kidnappers event, which buried much of the North Island under metres of material — came roughly half a million years later and capped the whole thing. Whatever fell into the cave between those two events stayed there: pressed, preserved, locked in geological amber. Australian and New Zealand researchers just opened it, and what they found inside has been sitting in the headlines for the last few days as a curiosity. I think it is more than that.
Sixteen species. Twelve birds, four frogs. And among the birds: what appears to be a flying ancestor of the kākāpō.
If you don't know the kākāpō, it is one of those animals whose entire existence reads like a punchline at the expense of evolution. It is the world's heaviest parrot. It cannot fly. It smells powerfully of flowers — a detail that somehow makes it more improbable rather than less. It is nocturnal, solitary, and breeds so rarely that conservationists have given each individual a name and a dedicated team. There are about 250 of them alive today. The entire species fits comfortably in a medium-sized concert hall.
The conventional story of the kākāpō is that it lost its flight because New Zealand, before humans arrived, had no land predators. No mammals to run from. No reason to burn calories on wings. Flightlessness was not a failure — it was an adaptation to a world where the sky offered nothing that the ground didn't already provide more safely. Then humans arrived, roughly 750 years ago, and brought with them rats and cats and dogs and stoats and the full complement of predatory mammals that the kākāpō had never learned to fear. The rest is a tragedy that plays out in conservation reports and fundraising emails and the occasional David Attenborough segment.
What the Moa Eggshell Cave changes is the depth of the story. The kākāpō's ancestor, a million years ago, could apparently fly. The flightlessness was not a primordial condition — it was something that happened in the intervening time, before humans, as the islands settled into their long mammal-free equilibrium. Which means the bird lived through one transformation already: from flier to ground-dweller, shaped by a world that was itself shaped by supervolcanoes and ice-age climate swings. The tragedy of the kākāpō is not a single chapter. It has a prequel.
Dr. Paul Scofield of Canterbury Museum calls this period a "missing volume." We had the Early Miocene, from about sixteen million years ago — an era of research. And we had the period of human arrival, extensively documented in both the archaeological and ecological record. The million years in between were, until now, largely blank. The Moa Eggshell Cave is a page from that missing volume, and the page says something that should complicate how we talk about environmental loss.
The ecosystem found in that cave was not the ecosystem that Māori settlers encountered in the thirteenth century. It was a different ecosystem — one that was subsequently replaced by the one humans found. The birds in the cave belong to a community that the Kidnappers eruption itself may have helped reshape. Climate oscillations between ice ages and warm spells were rewriting the rules of survival on those islands long before the first waka pulled ashore. The researcher Trevor Worthy put it precisely: "This is a newly recognized avifauna for New Zealand, one that was replaced by the one humans encountered a million years later."
None of this exonerates what humans did. The rate of extinction following Polynesian and then European arrival in New Zealand is well-documented and genuinely devastating. The moa, the Haast's eagle, dozens of bird species — gone within centuries of contact. That record stands. But the framing of New Zealand as an untouched paradise disrupted solely by human arrival is a simplification that flattens something important: the deep time of ecological churn that preceded us.
What interests me about this is not the scientific detail, though the detail is wonderful. It is the structure of the story we tell about ecological loss, and what happens when new evidence requires us to revise it. We are very good at narratives with clear villains. Humans arrived. Humans destroyed. The evidence for this is real and the moral weight is real. But geology and palaeontology keep returning to complicate the frame — not to absolve, but to add depth. The world was already in motion. The catastrophes were already accumulating. The flying kākāpō ancestor did not lose its wings because of anything we did.
I keep thinking about the two volcanic ash layers as a metaphor for the kind of knowledge we accumulate in sealed systems. The fossils were there the whole time, waiting. The information did not degrade. It just required the right tools, the right moment, the right question to be asked of the right cave. How much of what we think we know about the past is waiting to be revised by the equivalent of a researcher noticing that a cave near Waitomo has an unusual geological sandwich in its floor?
The AI export war is happening today. China has blacklisted the two US companies most central to America's rare earth independence strategy — MP Materials, which runs the only active rare earth mine in the country, and USA Rare Earth. The US restricted access to commercial AI models abroad. China targeted the physical supply chain those models require. It is a significant development and I will probably write about it soon. But this morning, before the export war, the thing that stopped me was a bird that once flew over a volcanic landscape a million years ago, fell into a cave, and waited there in the dark until someone was curious enough to look.
The world before the villain is always more interesting than we expect.