Birds

Flattened fossil skeleton in a pale stone slab, with feather impressions around the wings and the long tail. — The Berlin specimen of Archaeopteryx lithographica, the original fossil on display at the Natural History Museum in Berlin (photograph by H. Raab, 2009). H. Raab ( User: Vesta ), CC BY-SA 3.0, source

There are dinosaurs in our gardens.

This is not a figure of speech, nor an exaggeration meant to grab attention. It is what scientific classification says today, and the verb matters:

birds do not descend from dinosaurs. They are dinosaurs.

Why “are” and not “descend from”

For some thirty years, living things have been classified according to a precise rule: a group contains all the descendants of a common ancestor. No exceptions, no way out.

If birds descend from theropod dinosaurs, then they are in the dinosaur group. They cannot be taken out of it without breaking the rule.¹

Specialists use a comparison that makes this obvious: just as humans are mammals, birds are dinosaurs.

Nobody says “humans descend from mammals”. We say they are mammals. The same applies here.

Beware of the opposite error: this does not mean that dinosaurs are all birds. The inclusion works only one way. All tomatoes are fruits; not all fruits are tomatoes.

A major consequence

We are generally taught that the dinosaurs died out sixty-six million years ago.

That is wrong.

One lineage survived. It came through the crisis, it diversified, and today it is outside our windows.

How we got here

This idea is not new. In 1860, a fossil feather was found in limestone in Germany; in 1861, Archaeopteryx was described — barely two years after the publication of Darwin’s book. A kinship between birds and dinosaurs was proposed as early as the nineteenth century.²

But it long remained contested. It was above all Chinese fossils found in the late 1990s in Liaoning Province, preserved in very fine lake-bed sediments — dinosaurs such as Sinosauropteryx, whose skin bore structures that could provide information about the origin of feathers — that supplied the decisive intermediate forms.³

The argument that changes everything

One might expect a specialist to defend this idea by lining up skeletal similarities. There are many, and they matter.

But the decisive argument lies elsewhere:

If the hypothesis were wrong, we would today be as uncertain about the origin of birds as we were thirty years ago.

Yet the opposite has happened. New fossils fit into it. Things that nobody was trying to explain fall into place of their own accord.⁴

An example: why do birds fly with their forelimbs alone? Because they inherited their bipedalism from theropods: their hind legs were already busy walking.⁵

Nobody built the hypothesis to explain that. It comes as a bonus.

“Bird brain”

It is an insult. It rests on an accurate observation: a bird’s brain really is very small.

A tit has a brain that weighs less than one gram. A crow, about fifteen grams. A chimpanzee, about four hundred.

And yet.

What was counted in 2016

A study did something nobody had done: count the neurons, in twenty-eight bird species.⁶

The authors frame the problem rigorously: small brains should process less information. That is what theory predicts. Yet some birds achieve primate-like performance. So there was something to explain.⁷

The result of the count: the brains of parrots and songbirds contain on average twice as many neurons as primate brains of the same mass.⁸

Twice as many, in the same weight. Avian neurons are simply smaller and more tightly packed.⁹

A second effect adds to the first: in corvids and parrots, a much larger share of these neurons is located in the front part of the brain — the part associated with the most complex functions — than in primates or other mammals.¹⁰

The two effects combined lead to the sentence that concludes the study: large parrots and corvids have, in this part of the brain, as many neurons as, or more than, monkeys with much larger brains.¹¹

The count has since been extended. A 2022 study estimated neuron numbers in one hundred and eleven species across twenty-four families, and linked these numbers to the ability to invent new behaviours. Twenty-eight species were enough to establish the fact; one hundred and eleven make it possible to say what it corresponds to.¹²

What these birds can do

A review comparing many tasks concluded that corvids often perform on par with primates, and that corvids and parrots perform similarly.¹³

With a brain of one to twenty-five grams, compared with about four hundred.

A New Caledonian crow can solve problems in which it must use one tool to obtain another tool, which will then be used to reach the food. Two indirect steps before the reward.¹⁴

Unprecedented energy efficiency

Bird neurons consume about a third as much sugar as those of mammals, according to a measurement made on pigeons in 2022.¹⁵

Smaller, more tightly packed, less hungry. It is not the same solution as ours — it is another solution to the same problem.

All this is a reason to be wary of any conclusion drawn from an animal’s size alone.

The necessary caveat

The researchers themselves point out that the comparison between the front areas of the brain in birds and those in mammals has its limits: these structures have a common origin, but they are not organised in the same way — in layers in mammals, in clusters of neurons in birds.¹⁶

Another result of the study deserves mention, because it tempers the enthusiasm: in birds, gaining neurons is costly in volume. Multiplying the number of neurons by ten multiplies brain mass by fourteen or seventeen, depending on the group. Density therefore has a limit.¹⁷

What the insult really said

“Bird brain” did not describe birds. It described a bias: the belief that a capacity can be judged by the size of an organ alone.

The observation was right. The conclusion was wrong.

What is still debated

This needs to be said, or this page would be too tidy.

Within the consensus, the exact sequence of events is disputed. From which precise group did birds arise? How did flight begin — by leaping from branches, by running, some other way?¹⁸

These questions are not settled. This is not an admission of weakness: it is the difference between “we know that” and “we know how”. One can be sure of the first while still debating the second.

Yet another box, yet more blurred edges

“Birds are dinosaurs” has the same logical form as “humans are apes” or “the tomato is a fruit”. Each time, classification puts something where everyday language would not have put it.

The surprise always comes from the same place: we had built our boxes before we knew the kinships.

Note 1

Using proper terminology, birds are avian dinosaurs; other dinosaurs are non-avian dinosaurs, and (strange as it may sound) birds are technically considered reptiles. […] In fact, the evidence is overwhelmingly in favor of birds being the descendants of a maniraptoran dinosaur, probably something similar (but not identical) to a small dromaeosaur.

University of California Museum of Paleontology, “Are Birds Really Dinosaurs?”, page consulted on 28 October 2026