The Layer We
Call Essential

Ask how smart an animal is and you have already agreed to something: that smart is a quantity, that it runs along one axis, and that every mind can be given a place on it.

The axis needs a yardstick, so we have built several. Absolute brain mass. Brain mass against body mass. The encephalisation quotient. Cortical surface. Neuron counts. These are real measurements, taken carefully by careful people, and none of them is the problem.

The problem is the ordering. A measurement tells you how much of something there is. It does not tell you that more of it is better, or that the creature with the most of it is the one the others are failing to be.

And the ordering came first. We had the ladder long before we had the instruments, and then we went looking for the instrument that would show us on the top rung — which is a different activity from measuring.

Neuron number is the yardstick that feels least arguable. Cells are countable. You can do the stereology and get a figure, and the figure does not care what you were hoping for.

In 2014, Mortensen and colleagues did the counting on the neocortex of the long-finned pilot whale — a dolphin, despite the name. They found roughly 37.2 billion neocortical neurons, alongside some 127 billion glial cells.

That is, in the paper's own words, almost twice as many neocortical neurons as humans. The first animal ever counted with more neocortical neurons than us, and it is a mid-sized dolphin that most people have never heard of.

Neuron density is lower in the whale; the total is higher because the brain is bigger. Both facts are in the paper, and both are worth keeping.

Here is the sentence in that same paper. The absolute number of neurons in the human neocortex, the authors conclude, is not correlated with the superior cognitive abilities of humans — at least not compared with cetaceans — as had previously been hypothesised.

Read it slowly. The yardstick was neuron count. The count came back against us. And the conclusion drawn is not perhaps the ranking is wrong. It is the count must not be the right yardstick.

To be fair to the authors: as a narrow claim about neuroscience that is entirely defensible, and probably correct. Neuron number alone plainly does not explain human cognition. They are reporting an honest negative result.

But look at what stayed fixed while everything else moved. Human superiority is not the thing being tested here. It is the thing the test is being fitted to.

Which tells you something useful: the ranking was never resting on the count. So no count could ever disturb it. Another yardstick will be found, and if that one comes back wrong too, another will be found after it.

In 2006 Paul Manger published a long paper arguing that cetacean brains got big to make heat — thermogenesis against a cooling ocean — rather than to think. He argued further that cetaceans are considerably less intelligent than generally supposed.

One of his three claims was this: there is no neural basis for complex cognition in cetaceans. Among the evidence offered were the indistinct lamination, the pyramidalised layer II, and the absence of layer IV.

In 2008, sixteen authors replied. Those features, they answered, are one aspect of an unusual combination of conservative and derived traits — and cetacean brains, having diverged from our common ancestor more than 95 million years ago and arrived at high encephalisation entirely independently, in all likelihood represent a non-primate route to neuroanatomical (and cognitive) complexity.

You know this argument. It is the one made about us, in clinics and classrooms, in the same century. Here it is being had by neuroscientists, about whales, in a review journal — and the shape of the inference is identical.

Sperm whales talk in codas — short patterned bursts of clicks. For decades the Eastern Caribbean repertoire was described as around 21 discrete coda types. A modest vocabulary. Roughly what you would expect from a lesser mind.

In 2024, a team analysed 8,719 codas from at least 60 whales recorded between 2005 and 2018, and found the repertoire was not a list at all. It has four features that combine freely: rhythm and tempo, which hold steady regardless of context, plus two they named rubato and ornamentation, which shift with the conversation going on around them.

Combine those four and you get roughly 143 combinations actually in use — nearly an order of magnitude past the old count.

Notice the naming. Rubato and ornamentation are borrowed from music, because the categories on the shelf did not have a slot for what the whales were doing. The vocabulary had to be built before the structure could be seen.

Which is the quiet lesson of the whole zine. When your instrument has no category for a thing, the thing reads as absence. Twenty-one types was never what the whales had. It was what we could hear.

In Shark Bay, Western Australia, some bottlenose dolphins break a marine sponge off the seafloor, wear it over the closed beak, and use it to probe the sediment for fish. It is called sponging.

Not the dolphins. Some. Krützen and colleagues found it in 15 of 141 known mothers, and mitochondrial DNA put its transmission almost entirely down a single matriline — mother to daughter, learned rather than inherited.

And spongers live differently. Compared with non-spongers in the same water, they are more solitary, work the deep channels, dive for longer, and spend more of the day foraging.

So inside one population, sharing one gene pool and one bay, there is a minority with a different sensory route to food, a different social pattern, and a different shape of day — passed along a family line, by learning.

We have a word for a profile with peaks and troughs rather than a single level: the spiky profile. And a word for surfacing and diving by capacity — dolphining, which we named after them before we knew this.

In 2011, off the island of Pico in the Azores, two researchers watching sperm whales found something else in the group: an adult male bottlenose dolphin whose spine curved so severely that his rear half held an S shape.

Over eight days he travelled with the whales, foraged with them, and played with the adults and their calves. When he rubbed his body against them, the rubs were sometimes returned.

Two things this is not. First, it is not an adoption. The paper is titled Repeated Non-Agonistic Interactions — careful, unglamorous, exactly as far as the evidence goes. "Adopted" is the press's word, not the researchers'.

Second, we do not know why. The suggestion that other dolphins had harassed him and he sought refuge is a plausible guess offered to reporters — not a finding. It may be that a curved spine made sperm whale pace easier to keep than dolphin pace. We don't know.

What is in the record is eight days, and the rubs coming back. That is enough. It does not have to mean anything about us to be worth knowing.

In 1979, in The Hitchhiker's Guide to the Galaxy, Douglas Adams made this exact joke and made it better. On Earth, he wrote, humans had always assumed they were the most intelligent species on the planet — whereas in fact they were the third. Mice first. Dolphins second.

Look at the mechanism. He does not overturn the ranking. He keeps its form perfectly intact and moves us down two rungs, and the whole apparatus collapses into absurdity, because a ladder that puts mice on top was never measuring anything.

The dolphins, knowing what was coming, left. Their last message, mistaken for a stunt: So long, and thanks for all the fish.

Then the bill arrived. In 1988 Adams travelled with the zoologist Mark Carwardine to find endangered species for Last Chance to See, and one of them was the baiji, the Yangtze river dolphin — a freshwater cetacean going deaf and blind in a river turned into an industrial corridor.

In December 2006, a six-week survey of the Yangtze ended without a single sighting. Turvey and colleagues published it as, possibly, the first human-caused extinction of a cetacean species.

A species actually said so long. Not as a punchline.

The ranking was never a description of the world. It was a permission slip — and the thing it gave us permission to do, we did.