by Meave Leakey
If you don’t believe me, consider this. An amateur runner with little training can sustain an average speed of five metres per second and can easily cover ten kilometres on a daily basis without tiring overmuch. With more fitness training and more innate talent, a world-record-holding Kenyan can run at 6.5 metres per second and sustain this speed over a marathon distance for two to three hours (although not every day). In October 2019, Eliud Kipchoge achieved what had always been considered beyond the realm of human possibility by running a full marathon of 42.195 kilometres at a blistering pace of 1 hour, 59 minutes, and 40.2 seconds. In contrast, a dog of similar body mass can sustain a gallop at 7.8 metres a second but can only do this for ten to fifteen minutes. Dogs such as huskies are the elite athletes of their species. These specially bred dogs can run up to fifty kilometres. However, there are two provisos: they can only do so at a trot, and they need frigid conditions to keep them cool.
The way that nonhuman mammals keep cool is by panting, which is rapid shallow breathing. Although an animal takes ten times as many breaths panting as they would breathing normally, the exchange of air is limited to a space in the upper pharynx, so there is little or no intake of new oxygen into the lungs and bloodstream to fuel the aerobic exercise of the muscles. This makes extended galloping impossible while panting. The world’s premier sprinter, the cheetah, gets so hot that it has to stop after just one kilometre. It can reach a staggering speed of 110 kilometres per hour (70 miles) in seconds, but a chase usually lasts only 20 seconds and rarely exceeds a minute. A human generates ten times more heat while running than walking, so you can imagine what hopeless athletes we would be if we relied on panting to keep us cool. Instead, humans have evolved an efficient alternative cooling system that allows us, in Dan’s words, to “dissipate copious quantities of heat while running in hot, arid conditions.” A key way we do this, of course, is by sweating profusely when we run. We also shed our fur sometime in the past, which makes sweating still more efficient because fur traps air and moisture in a layer above the skin and reduces convection.
This combination of efficient sweating and no fur is what singles out humans. Horses are the mammal that we immediately think of as endurance runners that sweat. And Dan and Dennis talk about horses. Horses can outrun humans with a maximum galloping speed of 8.9 metres per second for a 10 km race. But for longer distances, a horse’s pace must slow to about 5.8 metres per second. At this speed, a horse can run for about 20 km a day—anything more, and the horse will suffer irreparable damage to its muscles and skeleton. It turns out that Dan and Dennis were not the first to question a man’s prowess compared to that of a horse. Two inebriated men in a Welsh pub were overheard by the landlord debating this very point in 1980, and a challenge was born, lubricated by several additional pints. The landlord turned the private argument into a public challenge, and ever since, man and horse race in an annual marathon event over the hilly moors of Wales from that very pub. Dan points out that, on occasion, the man outruns the horse in this unlikely race. In 1999, the fastest horse beat the runner by only eighty seconds, and in 2004, the horse was beaten by a runner by two minutes. But horses have to carry riders in this race, which unfairly handicaps them for our comparison. But these particular horses have also been carefully bred over generations to improve their endurance and speed, and the race is in cold, wet, windy Wales, which tips the advantage away from humans.
Other long-distance runners that come to mind are wild dogs and wolves. The wild dog, which hunts in the hot African savannah, is perhaps the most apt comparison for the purposes of looking at H. erectus. These dogs are impressive. They hunt in packs that range in size from two to as many as sixty individuals and reach speeds of up to a staggering 18.3 metres per second (64 kph or 40 mph), and they can sustain a slightly slower hunt speed of about 50 kph for at least 5 kilometres in hunts that can last up to an hour. Their prey includes gazelle, wildebeest, and zebra, and range in size between 20 and 126 kilos (44 to 277 lbs.) compared to the wild dogs’ body weight of 17 to 30 kilos (37 to 66 lbs.). One pack that scientists followed in the Ngorongoro Crater in Tanzania made an average of two kills per day with a hunting success rate of 85 percent, and another pack in the Serengeti achieved success 70 percent of the time. These figures far exceed those of other predators. These dogs hunt only in the cool of early morning and late evening to avoid getting heat exhaustion, so their high success rate during hunting is all the more impressive compared to those of other African plains predators.
We alone are capable of endurance running at midday in hot climates, and it turns out that traditional human societies in tropical arid habitats still practice persistence hunting. Societies where this technique has been documented include the Bushmen of the Kalahari, the Tarahumara of northern Mexico, the Navajo and Paiutes of the American Southwest, and the Australian aborigines. In all these cases, the hunters prefer the hottest possible time of day and year. The favoured speed in all these groups is between the prey’s preferred trotting and galloping speeds—an awkward speed for prey that cannot gallop for sustained periods and don’t have long enough to cool down in between galloping bursts before the hunters catch up again.
What is remarkable about this hunting strategy is that it unites the benefits of low risk, low metabolic cost, no need for sophisticated technology, and a relatively high success rate. Of the hunts documented by one researcher in the Kalahari, 50 percent were successful, which is higher than the rate achieved using a bow and arrow. Two further elements contribute to the high success of this type of hunting. First, tracking is a critical and difficult skill that has to be learnt. The better the tracker—which entails both reading the tracks on the ground and anticipating the animal’s movements—the faster the hunter can track the prey, and the sooner it will tire and collapse from overheating.
Second, it requires a high degree of cooperation and a cohesive social group with strong bonds. When a hunter returns to camp empty-handed, he or she can depend on other members of the group to help replenish the additional calories expended in the hunt as well as their normal metabolic requirement. When the hunt is successful, the highly caloric meat is sufficient to feed the whole group. Like modern humans who use this strategy, the wild dogs that have been studied in the Serengeti are notable for the extreme amity that exists between members. Even with a whole pack crowding around a single kill, there is little strife. And even more remarkable, dominant adult males will wait while the pups take their fill.
For modern humans—with their big brains and capacity for speech —planning and executing a successful hunting strategy and passing complicated tracking skills down from one generation to the next are relatively straightforward to master. The crux of Dan and Dennis’s argument is that H. erectus could have done this too. They point to its physique, its brain size, and its ability to make symmetrical tools that require the maker to work from a preformed mental template.
There are two possible holes in their argument that need to be dealt with, however. First, could early hominins have run over rocky, thicketed, and uneven ground without sustaining serious injuries? When I confronted Dennis with this objection, he provided a surprising and compelling response. He pointed out that in the United States “very long endurance races (50 to 100+ miles) are popular and are generally run over rough, hilly and mountainous terrain with plenty of rocks, thickets, and the like. There is a surprising lack of serious injury among the participants—exhaustion and dehydration are the actual killers.” Dennis also mentioned another fascinating fact about these cross-country ultramarathons: at this distance over this terrain, middle-aged women are much more competitive with men. So presumably for H. erectus, every member of the group could participate in long-distance hunting with equal success—grandmothers included. Dennis’s answer satisfactorily dispenses the first problem but leads us directly to the second, potentially bigger flaw in the argument. How did H. erectus avoid death from dehydration? The only solution will never be provable—but surely a m
anually dexterous creature capable of fashioning stone tools and carrying these from the factory site would have been capable of making a receptacle from a gourd or the stomach of prey to carry some water on the marathon hunt. It is also worth noting that modern desert-dwelling people are notable for how much less water they need to survive than we soft town-dwellers do. Quite probably, H. erectus was also efficient with its water metabolism.
Alan’s calculation of what the Turkana Boy’s brain would have reached at adulthood (909 cc) is not far off from that of a one-year-old human baby. One of the interesting developments in early childhood learning that I experienced after I had my own children is the use of signing to help babies to communicate before they can speak. If a mother is diligent at teaching her baby the signs, and begins between the ages of six and eight months, the child will be using the basic signs—I am hungry, I am sleepy, my tummy hurts, more, and enough—within only six to eight weeks. By age one, the child will be mimicking and learning new signs almost as fast as the mother can teach them and will have quite a big repertoire of signing vocabulary. As any mother or grandmother will testify, babies understand a great deal even though they cannot speak.
My point is that one-year-olds are really smart. At age one, my granddaughter, confronted with the sight of a carcass of a lamb hanging in a cool room off our kitchen, pointed and said, “Baaa!” At aged two, Seiyia could identify the birds in the bushland around the house. Once, when we were on a walk together, she pointed at a lamb frolicking playfully in the grass, and said, “Babu [her grandfather] cooks really tasty chops!” Not long after this, while watching the DVD of Finding Nemo, she announced how much she liked eating fish. From a very tender age, Seiyia has always had a clear appreciation of where her dinner comes from. I don’t think it is a big stretch to hypothesize that H. erectus would have already developed some effective form of communication that it used to hunt and gather food cooperatively even though it only had the brain of a one-year-old.
In any case, because of our secondarily altricial development, human babies are not as good an example as other members of the primate family. Other primates give a better minimum expectation of what H. erectus would have been capable of. During the dry season, the local troop of baboons that frequents the valley beneath our house stealthily approaches the vegetable garden every Sunday at one p.m. I don’t know how they know it is Sunday lunchtime, but they do—and they also know that this is the time when they are least likely to be caught when they breach the fence and start ripping up our precious vegetables. It is really remarkable and truly inconvenient at times! But more formal observations confirm how intelligent other primates are and how highly evolved primate social systems can be.
The celebrity of the chimpanzee world, Kanzi, and his younger sister, Panbanisha, offer a fascinating glimpse into how versatile the cognitive skills of chimpanzees are. Kanzi is a bonobo chimpanzee orphan that was an infant in the 1980s when scientist Sue Savage-Rumbaugh was engaged in trying to teach his adopted mother, Matata, basic words and symbols at the Great Ape Trust near Des Moines, Iowa. Like any child, Kanzi loved to mess up the teaching aids that Sue was working with or jump on the keyboard of lexical symbols or partake in other amusing games. Sue had little inkling of how much Kanzi was absorbing passively as he played until Matata was taken away for breeding when Kanzi was just two and a half years old. Kanzi, completely bereft and devastated at the loss of his second mother, turned in desperation to the closest social bond he had left—Sue. To Sue’s astonishment, Kanzi turned to the keyboard more than three hundred times the very day his mother was taken away and asked Sue for food, affection, and help finding his mother. Kanzi had learnt these symbols the way human babies learn language—by being around those who were using them. But Kanzi had never before had the motivation to use them, and Sue realised the implications of this serendipitous revelation. Chimps, like humans, are sponges for information as infants. But she was barking up the wrong tree trying to teach words and sentences out of context to adult chimpanzees.
The research program was completely overhauled. Before long, Kanzi was defying all preconceived notions of what linguists believed to be possible for a nonhuman. He was talking about places and objects that were out of sight, referring to the past and the future, and he was able to understand new sentences made up of familiar words. Linguists were in an uproar and required still more convincing. They asserted that Kanzi could have been responding to body language and facial expressions rather than words and that language included the ability to use metaphors and figures of speech. So Sue donned a welder’s mask to completely obscure her face and stood stock still while she asked Kanzi the unlikeliest of questions. He passed with flying colours. And Panbanisha, on one occasion when she disliked the behaviour of a visitor, used the symbol for “monster” to refer to the offender. Even more astounding is another story that Great Ape Trust researcher Bill Fields tells. Kanzi wanted to refer to a Swedish scientist named Pär Segerdahl, who Kanzi knew would be bringing bread with him. Not having a symbol to refer to the Swede, Kanzi pointed to the symbols for bread and the pear fruit. When asked, “Kanzi, are you talking about pears to eat or Pär?” Kanzi replied by pointing at the man.
The study with the bonobos is part of a growing body of work that has upturned a basic tenet of linguistic theory held since the 1950s: that the key to language is a uniquely innate human understanding of the grammar or the rules of language. But there is increasing evidence that language also depends on an ability to imagine the world from another’s perspective. This “theory of mind” holds that the meaning of words depends on the social context that produced them. For this reason, autistic people struggle enormously with social interaction although they have a perfect grasp of the mechanics and rules of language. But to understand the true meaning of what is being said, they also have to consciously learn a whole set of signals and nonverbal cues—such as eye contact and body posture—to help them figure out what people mean: skills that nonautistic people don’t even know they are using. Neuroscientists have a biological explanation for this. We have specialised brain cells called mirror neurons that reflect physical actions as well as emotions. These neurons fire when you stick your tongue out and when you watch somebody else stick their tongue out. But they also fire when you feel pain or when you see another person in pain, so they are essential for empathy. It is the ability to empathise—to put ourselves in another’s shoes—that allows us to understand the social context in which the words are being used. Without this, a lot of what can be accomplished with language simply disappears. It turns out that people with autism have mirror neurons that don’t always fire correctly.
Not only could Kanzi and Panbanisha comprehend spoken language that included metaphor (the monster and the pear) and the past and future tenses, but the evidence that they also showed empathy is very convincing. Kanzi’s friend Bill Fields is missing a finger on one of his hands. On one occasion, Kanzi was grooming Fields, and when he got to the hand with the missing finger, he used the keyboard to point to the lexicon symbol for “Hurt?” In much the same way, Richard had a bizarre bonding experience with an untrained adult chimpanzee. He was in a room with many others including the chimpanzee’s adopted human family. The chimpanzee entered the room, looked around, and made a beeline for Richard, who was sitting with his legs extended in front of him. Richard was rather alarmed to be singled out by this rather large strong adult chimp. But the chimpanzee merely settled itself at Richard’s feet, rolled up the leg of his trousers, and experimentally tapped his artificial leg. The chimpanzee then looked at him directly in the eye before exiting the room without further ado. How the chimpanzee divined that Richard had no legs I simply don’t know—but it clearly did, and it cared enough to investigate.
Empathy is the ingredient that allows us to use language for the purposes of social bonding. Obviously, it is impossible to see from the fossils whether H. erectus would have been already using language for both practical communications
and social bonding. But based on the evidence of modern bonobos and very young children, it seems likely that at least rudimentary bonding by language had begun. H. erectus probably had the brains to track and hunt as well as the social bonds to cooperate in the hunt and support the hunters when they were not successful. But did they also have the physique to persistence hunt using endurance running?
The focus on early bipedalism was previously centred on the ability to walk, not run, and there are more than a dozen skeletal features found in H. erectus that improve both running and walking. Dan and Dennis had another look at the skeletal features found in H. erectus to see if there are any characteristics that specifically pertain to running. The biomechanical differences between running and walking were the obvious place to start. At this point, it all gets horribly technical. One of the differences between running and walking is the way that the kinetic and potential energy are stored during each new step. When we walk, as we extend each leg, the body’s centre of mass rises as potential energy for half the stride and is then released as kinetic energy during the second half much like the way a pendulum keeps swinging. During running, the model is like a coiled spring rather than a pendulum. Elastic energy is stored in collagen-rich tendons and ligaments in the leg in the first half of the stance and then released in the second half of the stance that propels the body into an aerial phase. This means that any derived features pertaining to spring-mass mechanics are direct evidence for improving running capabilities.