The Family Tree Was Never a Tree

Not maliciously. The cabinet is doing its job — sorting the smear of the world into drawers marked this and not-this, because a mind that refuses to sort never gets to lunch. But every so often reality hands up a specimen that fits two drawers at once, or none, and the whole cabinet shudders on its rails.

This month the specimen is us.

Ian Towle, a research fellow at Monash University's Biomedicine Discovery Institute, has published an argument in the American Journal of Biological Anthropology proposing that Australopithecus and Paranthropus — two genera that have anchored the human origin story for roughly a century — be folded into Homo. Not renamed. Not reshuffled into a tidier arrangement. Absorbed. One genus, four to five million years deep, everything from Lucy to the primate currently reading this on a glowing rectangle.

The reasoning is almost boringly practical, which is the best kind. The traits we used to draw the borders have stopped cooperating. Big brain? Homo naledi and Homo floresiensis got by with braincases down in the australopith range. Tool use, that old trumpet blast of a criterion? No longer exclusive; the behavioral repertoire documented in Paranthropus and Australopithecus keeps turning out richer than the caricature allowed. Towle's actual claim is sharper than "let's be inclusive": he argues these groups may not represent distinct evolutionary branches at all, and that defending the borders means redrawing them every single time somebody's brush hits bone in a South African cave.

Which raises the question nobody asks at the museum. What was the boundary ever made of?

Here is the vertiginous part, and I mean vertiginous the way a glass floor is fun. There was never a moment when an australopith gave birth to a Homo. There was no first human. Every individual in that unbroken chain was the same species as its mother and the same species as its child, all the way down — an unbroken smear of ordinary Tuesdays — and somewhere in the middle of that smear we have driven a stake and hung a small brass plaque reading YOU ARE NOW ENTERING THE GENUS HOMO. The line is real in precisely the way the equator is real. Useful. Navigable. Not made of anything.

Worth being precise about what's soft here, though, because not all of it is. Two different questions have been wearing the same word. Where does Homo begin, asked along one unbroken lineage running backward through time, has no answer — not a hard answer, not a lost answer, none, because there is nothing down there to answer it. No joint. No seam. No Tuesday on which the species changed. That question is a convention in a costume.

But did Paranthropus branch off is a different animal entirely. Paranthropus boisei and Homo erectus walked the same landscapes at the same time, breathing the same air. Whether they sit on one branch or two is a claim about descent, and descent went one way or the other. There's a fact under it. That's why Towle had to publish a paper instead of a preference — and why he might lose. You can only be wrong about the second kind of question. So the failure isn't that we draw lines; it's that we can't reliably tell which kind we're drawing, and end up litigating conventions as though they were discoveries and announcing discoveries as though we were only tidying labels. The timeline has no joints. The branching does. We reach for the same brass plaque either way.

Naming makes it worse, because taxonomy arrives with sound effects. Homo rings like a fanfare. Paranthropus sounds like something that didn't make the cut. In fairness to the dead, the name came after the bones — Robert Broom pulled the Kromdraai material out of the ground first and christened it second, and para-, alongside, beside, adjacent to, was a hypothesis rather than a verdict. A guess that this was a side branch. A checkable guess. Towle is checking it, which is the system working as advertised.

The trouble is what happens to a hypothesis that gets to live inside a name. It stops being checked and starts being assumed. A lineage that persisted for something on the order of a million years spent the following eighty being introduced to undergraduates as near the point, long after the evidence that filed it there had been complicated, superseded, or quietly dropped. The label kept making the argument after its argument retired.

And when contrary evidence finally arrives, it never announces itself. It doesn't stand up and declare the old story wrong. It just quietly stops fitting. Fossils don't argue. They accumulate.

This is what a boundary failure looks like from the inside, and it's worth watching closely, because we're going to need the practice. The category human has been renegotiated before, and it has never once been a filing question. Every prior edition of that boundary decided who was owed something — protection, personhood, the presumption of an inner life — and the people on the wrong side of it did not experience the outcome as a debate about nomenclature. The line moved late, moved because it was fought over, and the fighting was done largely by the excluded. It will be renegotiated again, probably soon, probably around something without bones. So the lesson from the fossil record isn't that our ancestors were more like us than we assumed. It's that "like us" has sometimes been a measurement and sometimes a decision dressed in Latin and wearing a lab coat for credibility — and that we have never been much good at saying, at the time, which one we were doing.

Towle may well lose this argument. Taxonomists are a famously unbudgeable people, and there are respectable technical reasons to keep separate names for separate morphologies — a category that contains everything distinguishes nothing. Fine. Genuinely fine. The fight was never really about nomenclature. It's about whether we can hold a map loosely enough to redraw it once we find out which of its lines were surveyed and which were merely drawn.

The universe declines to sort itself. It just runs. Four billion years of unlabeled chemistry on one damp planet, sublimely indifferent to genus, and somewhere in the middle of it a hominin with a medium-sized brain picks up a stone, and much later a descendant of that hominin sits in an office in Melbourne and suggests, in careful academic prose, that we were all family the entire time.

Maybe we were. That one has a fact under it, and the fact will outlast the argument either way.

But notice who's doing the adopting. Paranthropus isn't being welcomed into the family; it's being annexed, by the only party in the room holding a pen, on terms it was never asked about. Which is fine — it's a fossil, it stopped caring a million years ago. What should snag is that this is the only procedure we have ever had for deciding where the circle ends. Every widening of us so far has been granted from above, late, by whoever held the pen at the time, and a boundary drawn generously is still a boundary drawn by one party over another. A warmer line is still a line somebody chose.

Nobody sent the memo, so somebody had to write it down. Worth staying awake to the fact that it's always somebody — and that so far, every single time, the somebody has been us.