The distinguished biologist Prof Richard Lewontin is one of science's superb curmudgeons. Never prepared to go with the crowd, he has made a career of asking awkward questions of his colleagues and their theories - often in a fantastically bad-tempered way.
It was he and the late Stephen J Gould, for example, who introduced the concept of spandrels into evolutionary thinking in 1979. The term was borrowed from architecture, where it means the space between an arch and a rectangular feature around it. These are often beautifully ornate, but they serve no purpose in terms of support. They are a byproduct.
The analogy was meant to caution evolutionary biologists about a tendency towards believing that natural selection is an all-powerful force that shapes every minute feature of living things. Many bits are simply byproducts of other evolved structures, he and Gould argued.
Lewontin was on fantastic curmudgeonly form at the American Association for the Advancement of Science annual meeting in Boston on Sunday in a session on the evolution of human cognition. His campaign against pan-selectionism was in evidence. "Evolution is not the evolution of traits but the evolution of organisms," he said.
But he had an even more sobering message, summed up in the title of his talk - "Why we know nothing about the evolution of cognition". He systematically dismissed every assumption about the evolution of human thought, reaching the conclusion that scientists are still completely in the dark about how natural selection prompted the massive hike in human brain size in the human line.
The main problem is the poor fossil record. Despite a handful of hominid fossils stretching back 4m years or so, we can't be sure that any of them are on the main ancestral line to us. Many or all of them could have been evolutionary side branches.
Worse, the fossils we do have are difficult to interpret. "I don't have the faintest idea what the cranial capacity [of a fossil hominid] means," Lewontin confessed. What does a particular brain size tell us about the capabilities of the animal attached to it?
He is even sceptical that palaeoanthropologists can be sure which species walked upright and which dragged their knuckles. Upright posture is crucial for freeing up the hands to do other useful things.
He is also not convinced that we can use current selective forces to infer what natural selection was doing to our ancestors. He used the example of the butterfly wing. The smallest wings provide no lift at all and so could not have been selected originally for flight. One idea is that they started off as structures to regulate body temperature and were later adapted by natural selection for lift. Maybe something like that happened for human brain size.
All in all, despite thousands of scientific papers and countless National Geographic front covers, we have not made much progress in understanding how our most complicated and mysterious organ came about.
"We are in very serious difficulties in trying to reconstruct the evolution of cognition," said Lewontin. "I'm not even sure what we mean by the problem."
