Paths to the rise of intelligence are of general interest to us. The paucity of direct testing on cetaceans makes the topic of peculiar interest to this website.
How Clever are Octopuses?
It’s more difficult to gauge the intelligence of octopuses than any other gifted group. By rights, this post should focus on the two different occasions in which experiments have supported social learning in the common octopus. As a solitary species, it is unlikely that such behaviour could be helped by hard wiring (eg. through the presence of mirror neurons). As such being able to copy the actions of a more trained octopus after visual observation, would suggest a level of intelligence close the that of mirror self-recognition (for details of how high the almost-MSR level is see my earlier post HERE). Those results have caused much controversy, precisely because it is so difficult to explain in a solitary animal – without invoking a super high intelligence. Other cognitive tests confirm that octopuses are likely to be the most intelligent invertebrate group, but not necessarily at that stellar level. Unfortunately, my prime concern today is not to pin down the exact level of octopoid cognition, but to look at two great issues that this research raises.
How Difficult is the Path to Intelligent Life?
By way of Earth’s example we can assume that the evolutionary drive to high intelligence might not be rare, else it would not have occurred in multiple unrelated groups. By contrast, the prevailing Rare Earth Hypothesis, posits that life is common throughout our universe, but intelligent life is very rare. Rare Earth is particularly difficult to explain given that Darwin’s entire principle was that no new biological function ever arises except by a slight modification of an already existing function. The first of our issues starts here: in all biology, the function whose complexity is such that we have the least idea of how to construct it, is (nontrivial) self-replication. Unfortunately this is the function which is needed for evolution to be possible in the first place. We would expect to be alone in the observable universe if we stayed faithful to Darwin.
Given the fundamental nature of this problem, it would be more rational to flip the Rare Earth Hypothesis, whereby simple life inevitably leads to complex intelligent spacefaring life, that eventually begins spreading from its home planet at a significant fraction of the speed to light. It seems more likely then, that complex life is common in the universe, but simple life forms would prove exceedingly rare. If this were so, you may well ask why hasn’t intelligent life arisen here long ago.
One powerful reason for the delay comes from biogeochemistry. For the first 80% of life’s history, oxygen levels were too low to sustain a vigorous nervous system. That changed shortly before the sudden appearance of complex animals. Among paleontologists this is know as the Cambrian explosion. From that time, average brain size increased in many different groups. Oxygen levels continued to rise, peaking about the prime age of the dinosaurs.
So… why didn’t technological intelligence arise much earlier on Earth?
Since vertebrates have a far better fossil record than soft-bodied cephalopods, and terrestrial animals are far better studied than marine ones, let’s look first at those dinosaurs. The only subgroup of dinosaurs alive today are the birds, yet paleontologists traditionally ignore their direct descendants, and use non-dinosaurs reptilians, such as crocodilians, to interpret their endocasts. If we do so, the brain of Troonodon, one of the most encephalised dinosaurs, would be about 100g. More likely, it had one closer to twice that size. Now, let us try to calculate its expected intelligence.
As this website is keen to point out, all large objective datasets place raw brain size as a better indicator of intelligence than any where a standard adjustment is made for body size. Furthermore, birds have a far higher neuron density than mammals. Interestingly, if we try to compare the studied intelligence of their smartest species with that of primates (the only cognitively well studied group), the best match-up tends to be with primates with their equivalent forebrain neuron count. Extrapolating that to their feathered relative we would calculate an expected Troonodon intelligence equivalent to a primate with a brain of 500g to 1kg. That is, great ape to human level intelligence. If we try the same exercise for Tyrannosaurus our result, we would be an order of magnitude higher, but the uncertainty would also be greater since they are not as closely related to birds. Mammals, by contrast, would not come anywhere close to matching this until 50 million years, or so, after the K-T strike made the dinosaurs relict. Given that some of their species were dexterous, what held them back during their long reign?
To me, one of the strangest differences between avian and mammalian intelligence, is that birds that are good at one task, such as singing, tend to be worse at other tasks, such as nest building. By contrast mammals tend to manifest intelligence as if it were a single factor, that we now know of as general intelligence. Dinosaurs might have been much smarter than mammals, but that comet strike changed the environment so dramatically, that superlative specialist cognitive skills were no longer as valuable as lower skills of a more general type. Could it be that, without the K-T disaster, animals far smarter than us might have might never have gained the skills to make boats and steam engines?
Could there Really have been a Past Technological Civilisation?
This brings me to a preprint by Dr. Wright entitled Prior Indigenous Technological Species. Wright argues that any evidence that we find for a past technological species, is far more likely to have an endogenous origin than an extraterrestrial one. So, in the unlikely event that we do find such an artifact, where should we look for that bygone civilisation.
Wright argues that any evidence for previous cities would have crumbled with time, but I disagree. True, all obvious signs would be long gone, but some strange concentrations of elements would remain, and a small number of materials just don’t degrade with time. A billion years later a gold and diamond ring is still as clear proof of its artisan, as the day they made it. Likewise, geological turnover covers much, yet large patches of those bygone continents would lie not too far beneath our current surface as to remain hidden to date. If cities had ever been built on land, we should have found the signs of several of them by now.
Another point Wright misses, is that it would be neigh impossible for a technological civilisation to arise without generating their own mass extinction, without use of fossil fuel, and without use of monoculture. I have been surprised to learn that we actually do see all three signals in the fossil record at a single point. Wright’s paper not being the main focus of this post, I will look at only one of those factors today: the use of fossil fuels.
The biosphere contains an isotopic ratio of carbon 13 to carbon 12 that varies slightly according to the total mass of the biosphere, as well as in other complex ways. This gives a natural limit to the change in that ratio, equivalent to all life on Earth being destroyed in a firestorm. A change greater than this in the δ13C, within a few thousand years, would be a clear sign of fossil fuel use. A signal almost that large occurs several times, including in the rock layer at the K-T boundary, but it clean exceeds that limit during the end Permian extinction. Furthermore, the signal goes the wrong way to be due to coal. Such would be the signal expected from burning massive quantities of methane hydrate, which happens to be more plentiful than coal, and doesn’t have to be seam mined. You may well wonder what sort of civilisation uses methane hydrate in preference to coal, and leaves no cities on land? To find out we should visit the only time when Wright’s scenario could be set: 251.9 million years ago.
In the end Permian there were no candidates for high intelligence on land. Some groups now known for their intelligence, such as mammals and dinosaurs, didn’t even exist. The only candidates for that level of intelligence were the cephalopods. Now, as then, prime among them were the octopuses. Unfortunately, their fossil record is very poor, even if it is sufficient to indicate that they existed in a form seemingly as advanced as those living today.
Why is the Octopus Brain so complex?
Today, octopuses live short and solitary lives, yet their brains are the most complex among the invertebrate. One indication that the drive to higher intelligence in their group is ancient, is that their genome is larger than ours, and a greater proportion of it is dedicated to brain function. Take the example of protocadherins, which allow the formation of different types of nervous tissue. Excepting octopuses and a few close relatives, no other animals have more types of protocadherins than mammals. All other invertebrates have less than half that number, yet octopuses have more than double the mammalian count.
Another indication is that this greater complexity is not just confined to their DNA. This year it became the only group known, in which alternative gene editing adds significant complexity to their nervous system. At the moment, the advantage this confers is not clear – though the retention of multiple instances, and their neigh exclusive use within nervous tissue, indicate that that gain is significant. If I had to guess, I would pick that it helped alter their specialist forms of intelligence to suit a changed environment.
And, at last, we are ready to conclude. I sometimes find myself wondering how would the degenerated descendants of truly intelligent forebears appear. If they had a specialist nature to their intelligence, what otherwise inexplicable traits might they have retained millions of years after their fall? How about these city building tendencies discovered last month. I’m really not sure what to make of that.