How Language Began

by Daniel L. Everett

  To summarise what we have learned about brains, humans’ larger brains could evolve only by overcoming three major disadvantages.3 First, it has already been seen that brain tissue is among the most metabolically expensive in the human body. The second problem is that larger brains take longer to mature. Human children are not able to defend, feed, clothe and shelter themselves for a minimum of twelve years and sometimes much more, depending on the culture. Finally, the third major disadvantage of larger brains is that there is a conflict in bipeds between the benefits of narrow hips to aid movement and the need for a birth canal large enough to accommodate increasingly larger-brained infants. Big brains can kill mothers in childbirth because the birth canal is small. This is so that the mother can walk, while the brain is large, so that the child can think.

  This raises the question of how big brains need to be to support human intelligence. Many palaeoneurologists use what is called the EQ or Encephalisation Quotient. This is a ratio of a species’ brain size to the average brain size of a mammal with the same body size. The theory behind the EQ is that intelligence grows not so much with the absolute size of the brain (a sperm whale’s is around 8,000cm3), but with the ratio of the species’ brain size to its body size. And this idea does seem to be a fairly reliable predictor. Tom Schoenemann from Indiana University has made the case that absolute brain size also matters, because it leads to specialisation in the brain that smaller brains are unable to achieve. Schoenemann lists several advantages to having larger brains for Homo erectus, other members of the genus Homo, and all other creatures.

  First, ‘highly encephalized species … tend to forage (or hunt) strategically, taking into account the habits of their food (or prey), while less encephalized species tend to graze (or hunt) opportunistically’. Further, ‘As brain size increases, different areas of the cortex become less directly connected with each other.’ The consequence of this changed connectivity is that ‘as brains increase in size, areas are increasingly able to carry out processing independent of other regions … Such independence makes parallel processing increasingly possible and this has significant consequences because it leads to great sophistication in behavioral response.’4

  Suzana Herculano-Houzel, in her 2016 book The Human Advantage: A New Understanding of How Our Brain Became Remarkable, has made the case that human brains are superior due in part to much greater neuronal density – we have more neurons per cubic centimetre overall and more connections between them.

  The idea that culture affects behaviour, appearance, intelligence and other aspects of one’s phenotype, leads to the conclusion that the most important question about our brains is not, ‘What in the brain makes language possible?’ The right question is, ‘How do brains, cultures and their interactions work together to produce language?’ The answer is that, over time, each has helped the other to improve. One cannot understand the evolution of language, therefore, without understanding the evolution of the brain. The brain likewise cannot be understood without understanding the evolution of culture.

  The challenge in understanding the evolution of the hominin brain once hominins diverged from other primates some 6 million years ago (whether via Ardipithecus, Sahelanthropus, or Orrorin), is not only how the human brain got bigger, but why. It is known that the brain grew from the time of Australopithecus from approximately 500cm3 to nearly 1,300cm3 in a relatively short space of 125,000 generations, or 3 million years. To understand why this increase occurred, it is necessary to understand the brains of contemporary Homo sapiens and to work out methods for understanding brain evolution in light of fossil and cultural evidence. There were several changes in the environment that pressured human brains to enlarge in order to support greater intelligence. Fortunately, we know the starting point for these changes to be Australopithecus, and much is known about the end point of hominin evolution – Homo sapiens. It only remains to determine how we got from one to the other. This entails figuring out the stages along the way. It is necessary, therefore, to examine the evidence for brain evolution in the fossil record and changes in the environment that might have exerted selective pressure upon human brain evolution.

  One aspect of the brain’s growth and development, its encephalisation, is easy. Creatures with bigger bodies tend to have bigger brains. As fossils indicate an increase in the overall body size of hominins, they also show increases in brain size. The formula seems simple enough – grow the body, grow the brain. So did the brain just come along for the ride with the growth of the body? Maybe not. In fact, the relationship of encephalisation to body growth may have been the reverse. It is possible that the external pressures leading to the growth of the brain also caused hominins to have larger bodies. Brain size and body size are controlled by some of the same genes. As Mark Grabowski puts it,

  results suggest that strong selection to increase brain size alone played a large role in both brain and body size increases throughout human evolution and may have been solely responsible for the major increase in both traits that occurred during the transition [from Australopithecus] to Homo erectus. This switch in emphasis has major implications for adaptive hypotheses on the origins of our genus.

  And, he continues,

  It may simply be that a larger brain requires a larger body to meet its increasing energetic demands and evolutionary constraints due to brain-body co-variation are one way of maintaining this relationship.5

  This all means that the evolution of brain and body size is a chicken vs egg problem. Either the brain evolved and brought the body with it or vice-versa. But whichever came first, the enduring question is to understand the pressures that led to growth in human intelligence. I think that the best way to look at the problem is, like so many aspects of biological development and existence, as a case of symbiosis – where two or more creatures or parts of creatures (like brains) developed or evolved in tandem, each needing and affecting the other.

  Considering, then, the implications of brain anatomy and functioning as part of a human body and culture for the understanding of brain and language evolution, it is worth returning to further discussion of palaeoneurology. What one needs to consider seriously is what it meant and still means for the understanding of hominin evolution that the human brain grew so quickly and reached such a large size relative to the rest of their bodies. As neurolinguist John Ingram has stated, this represents in evolutionary time a ‘runaway growth of the brain’.

  A fascinating discussion on brain size and growth is given by palaeoanthropologist Dean Falk, who compares the Taung baby fossil discovered by Raymond Dart to the discovery of the ‘Hobbit’, a tiny variety of Homo erectus whose fossils were unearthed on the island of Flores by Australian palaeontologists Peter Brown and Michael Morwood.6 It had been known for some time that Homo erectus arrived on Flores and developed a robust cultural outpost some 900,000 years ago. But the Hobbit was unexpected. The first question about them was, ‘Why were they so small?’ Another was, ‘How did they survive so long in co-existence with Homo sapiens?’ Apparently, the Hobbit had lived until as recently as 18,000 years ago, perhaps even as recently as 14,000 years ago. Since most researchers believed that all non-sapiens species of the genus Homo aside from neanderthalensis had died out roughly 200,000 years ago, this was a shock.§

  The brain of these occupants, now known as Homo floresiensis, was much smaller than that of their erectus ancestors. In fact floresiensis’s brain was smaller than the brain size of many australopithecines, coming in at around 426cm3. What does this surprising reduction in brain size in the erectus line mean for the understanding of the development of human intelligence? Does the smaller brain size of the Hobbit creatures indicate that they lost intelligence? That would represent a fascinating step backwards in evolution. Both had similar brain sizes and both were around 3'11" (1.19m) tall, but was the Hobbit as smart as an australopithecine? Or was it smarter? Or was it not as smart? Was floresiensis as intelligent as any other Homo erectus, in spite of having a brain less than
half the size of the Homo erectus that left Africa hundreds of thousands of years earlier?

  Based on their use of tools and other archaeological evidence, it seems that floresiensis was smarter than australopithecus. There is evidence that it had culture, at least in the use and manufacture of tools, as well as its ancestors’ initial voyage to Flores, discussed earlier. It is possible that the Hobbit had lost the culture of its ancestors, but this is unconvincing speculation, because we know it used fire and stone tools that were polished and shaped for working on softer materials such as wood and bone. This means, however, that intelligence is not and cannot be simply a function of brain size. There is no evidence other than cranial size to suppose that Homo floresiensis was any less intelligent than Homo erectus. If they were, in fact, equally smart then this raises the question of whether erectus, with its brain roughly two-thirds the size of a modern human’s, might have even been as intelligent as Homo sapiens. The point of asking this question is that if one is looking for evidence concerning the intelligence of a fossil human, the cultural evidence may be more important than the physical. And the failure of brain size per se to reflect intelligence means that to understand the brains of our hominin predecessors we need accurate information about their cytoarchitecture, neural density, their culture and their languages. None of this is possible to achieve given current data and methods.

  Summarising to this point, the archaeological record supports the thesis that general intelligence underwrites language, not some hypothesised language-specific, innate portion of the brain. And no innate language-dedicated areas of the brain have been found. If this thesis is correct, then one might suppose that reliance on large-scale, non-innately specialised neurological connections ensures greater plasticity. The specialisation of regions of the brain is largely due to the cytoarchitecture of the relevant brain areas coupled with the ontological development of the individual (their life path), including their biology, culture and personal psychology. Yet overall, the brain draws on all of its forces largely simultaneously as its possessor moves through the world.

  One lesson to be drawn, therefore, regarding the Hobbits of Flores is that inferring intelligence from brain endocasts is a risky business. There are certainly signs that can be interpreted, such as the evidence for development of different areas of the brain that we know to be associated with intelligence, language, planning and problem solving in modern humans, but the knowledge we derive from examining skulls is still inadequate for understanding the growth of human intelligence and must take a back seat to the cultural evidence. It would have been easy, in the absence of evidence about their villages, sailing, tools and so on, to claim that erectus was a dumb brute compared with modern humans because of its 950m3 brain size. But the cultural evidence, to the contrary, suggests that such speculation is unwarranted. What we see from the cultural evidence is that Homo erectus was intelligent, capable of human language and the master of its environment.

  Robin Dunbar, a British anthropologist, claims that the main force driving hominins to develop greater intelligence was increased social complexity. Dunbar argues that it wasn’t so much the problem solving required by ecological change that favoured the growth of human intelligence, but rather that the pressure for intelligence and encephalisation came from the increasing size of human societies. Humans were settling in larger and more complex groups. These surpassed in size and complexity those of any other primate. Dunbar’s argument, then, has to do with the exponential growth in the number of social relationships that arises from even the modest increases in overall group size. Whereas humans’ closest living relatives, chimpanzees, live in social groups of about fifty individuals, human hunter-gatherer societies average about 150, adding huge stress to the brain to keep track of the much greater number of social relationships that this 300 per cent larger group size entails. Individual members of a society are like neurons in the brain. The more there are, the more connections between them. In other words, just as it is the relationships between neurons that make the brain so complex, so it is the exponential growth in relationships as societal numbers increased arithmetically that required more intellectual horsepower in order to keep track of those relationships, at least according to Dunbar. In other words, as group sizes grew, the human cortex grew as well.

  To support this hypothesis, Dunbar noted that cortex size co-varies with group size across several species. Of course, one could reply that Dunbar has the cart before the horse here. Perhaps it was brain growth and greater intelligence that enabled the growth in social relationships among humans, rather than the other way around? But the causality seems more likely to go Dunbar’s way: social size → brain size, rather than brain size → social size. If one had a larger brain first, before social change, then one might have preferred to become a hermit. That is, the brain growing first could have led to any number of social models. But if society grew first then it would have indeed pressured the brain to be able to keep track of the new relationship sizes.

  Another socially induced pressure for intelligence growth is the growth in cooperation. As humans banded together, they worked together. The first human bands were made viable through cooperative work. Of course, in any group effort there will usually be a passenger or two, those who are content to reap full benefits of the efforts of others while failing to provide full efforts themselves. For group relationships to work more effectively, therefore, natural selection would have favoured improved intelligence in order to detect cheaters.

  As we have seen, sexual selection was noticed early on by Darwin as a major force in evolutionary change, responsible for beauty (such as male peacock feathers), physical attributes such as larger breasts in human females relative to other primates (apparently even early hominin males preferred their women to look pneumatic) and longer penises for human males.¶

  An additional consideration favouring increased intelligence could be that its possessors are more likely to survive diseases of the mind or nervous system (such as meningitis) that have the side effect of reducing intelligence in the survivors. This could have in turn fed sexual selection in that males and females would prefer mates who survived diseases with less impairment or long-term effects.

  It is most likely that all of the above reasons contributed to the natural selective pressures for greater human intelligence. And yet none seems to be the primary contributor to major leaps in our cognitive powers. In fact, it doesn’t seem prudent to suppose that ‘smarts’ can be understood exclusively through brain-case size of fossils or in overall size of the brain, or even in evidence that this or that area of the brain was better or less developed. Intelligence is not simply a function of brain size or brain component size. If it were, then among modern humans men would almost always be more intelligent than women because their brains are almost always larger, often much larger. Some modern European women have brains of only around 950cm3, nearly identical in size to the brains of Homo erectus. Yet they certainly seem as intelligent as modern males with larger brains.

  So what might have selected humans for greater intelligence as a function of brain size, cytoarchitecture, synaptic complexity, white matter, glial cells# and so on? The strongest force for the evolution of human intelligence was in all likelihood a combination of language and culture, as it is manifested through the use of symbols, grammar, pitch and gestures. As people began to use these methods of communication they were able to think more together, enhance each other’s ability to know the world around them and to predict its future forms. Questions began to occupy the minds of sapiens ancestors: ‘Where will that animal be in a few seconds?’ ‘In what direction will that fire burn?’ ‘When will the rain return?’ ‘To where does that river flow and what will I find if I go up vs if I go down it?’ And while asking these questions, humans needed to use language to bring order to their social interactions, naming kin and other relationships, leading to general improvement in their cognitive endowment.

  Now that we have some id
ea of how the brain evolved in general terms, we need to ask the next question. What are the specific features of the human brain that underwrite our language ability? And are these features unique to language or do they serve other roles beyond language? That is the crux of a several-decades long debate in the cognitive sciences and palaeoanthropology.

  * This is an area of the brain often identified with language. We will be discussing it in detail in chapters 6 and 7 below.

  † White matter is named for the white (because fatty) material (technically, myelin sheaths) that surrounds the nerve fibres connecting parts of the brain used for higher cognitive functions. Timothy A. Keller and Marcel Adam Just, ‘Altering Cortical Connectivity: Remediation-Induced Changes in the White Matter of Poor Readers’, Neuron 64 (5), 2009: 624–631; doi:10.1016/j.neuron.2009.10.018

  ‡ Philosopher Robert Brandom has made this point in his own work, in books such as Making it Explicit (Cambridge, MA: Harvard University Press, 1998), in which he offers some convincing reasons to believe that we acquire concepts only by using them to draw inferences – that is one can say that humans have a concept only after they have learned it well enough to use it in reasoning. I have made similar points from a very different perspective in Dark Matter of the Mind.