Meat eating made us human. The anthropological evidence strongly supports the idea that the addition of increasingly larger amounts of meat in the diet of our predecessors was essential in the evolution of the large human brain. Our large brains came at the metabolic expense of our guts, which shrank as our brains grew.

In April 1995 an article appeared in the journal Current Anthropology that was an intellectual tour de force and, in my view, an example of a perfect theoretical paper. “The Expensive-Tissue Hypothesis” (ETH) by Leslie Aiello and Peter Wheeler demonstrated by a brilliant thought experiment that our species didn’t evolve to eat meat but evolved because it ate meat.

The ETH is an example of the kind of scientific detective work I love. In fact, this paper is one of my all time favorites. (An amazing bit of trivia about this paper is that it almost didn’t get published. I had the opportunity to talk with Leslie Aiello at a meeting a few months ago, and she told me the journal was reluctant to publish the paper because the editors thought it too technical for their readers. I suspect they also found it too controversial. Now I’m sure they’re glad they published because I would imagine it is the most cited of all the papers ever published in Current Anthropology.) The authors methodically lay the scientific foundation for their experiment, then, like Sherlock Holmes, progress step by step, accumulating little pieces of data until they reach the ineluctable conclusion that meat eating made us human. I would like to walk us all through their thought processes as laid out in their brilliant paper.

Let’s start with the problem.

For years anthropologists have speculated about why humans developed such large brains so quickly – from softball size to what we have now in just a short 2 million years. Below is a graphic showing hominid/human brain growth over time.

    

A number of hypotheses have arisen to answer this question. Some say that humans developed large brains because they had to contend with problems involving group size, others posit that large brains came about as a consequence of developing complex foraging strategies, others yet say the development of a social or Machiavellian intelligence was the driving factor. And even others say that the complexities of learning to hunt expanded brain size.

Any or all of these hypotheses may be valid, but the problem isn’t really as much a matter of why as it is a matter of how. Other primates deal with groups and have complex foraging strategies; and many deal with social problems within their groups, and some even hunt. Yet they still have small brains. (Granted, their brains are larger for their size than those of other mammals, but primates sport small brains as compared to humans.) How did the human brain grow?

This isn’t an easy question to answer because of the thermogenics involved. Brains consume a large amount of fuel and, consequently, throw off an enormous amount of heat for their size. The metabolic rate of brain tissue is nine times that of the average of the metabolic rate of the rest of the body.

So what? you may say. So we’ve got a big, hot-running, energy-burning brain. What difference does that make? It’s reflected in our overall metabolic rate, right? Well, sort of, and therein lies the crux of the problem. As we will see below, our total metabolic rate – even with our huge brains – is the same as that of any other animal our size. Or to say it another way, animals our size with much smaller brains have the same metabolic rate that we do with our huge brains. This fact was the starting point for the authors of the ETH. So let’s start there as well.

    

In keeping with a great scientific tradition, Aiello and Wheeler were able to see what they saw because they stood on the shoulders of giants who came before them. In their case the giant was Max Kleiber, an animal physiologist working at the University of California at Davis, who published a groundbreaking paper in 1947 and a scholarly text titled The Fire of Life in 1961. Kleiber’s work involved the meticulous measurement of the metabolic rates of numerous animals, including humans. As he plotted the various metabolic rates, he discovered an extremely strong correlation between the mass of an animal and its metabolic rate. Kleiber found that this relationship held constant across numerous species. His October 1947 paper in Physiological Reviews simply titled “Body Size and Metabolic Rate” was a classic. By using the equations Kleiber worked out, the metabolic rate of virtually any animal could be determined simply by knowing the animal’s body size. Or, as Kleiber put it in the paper:

Does a horse produce more heat per day than a rat or do some rats produce more heat than do some horses? Almost anybody who understands what is meant by “heat production per day” will not hesitate to give the correct answer and will even be convinced that the daily rate of heat production of men or sheep is greater than that of rats, but smaller than that of horses. Thus most people (among those who understand the question) are convinced that in general the bigger homeotherms produce more heat per day than the smaller homeotherms, that, in other words, the metabolic rate of homeotherms is positively correlated to body size.

The answer to the next question: “does a horse produce more heat per day per kilogram of body weight than a rat?” requires some biological training. Most biologists, however, will not hesitate to answer that the rate of heat production per unit body weight of the big animal is less than that of the small animal.

The positive correlation between metabolic rate and body size, and the negative correlation between metabolic rate per unit weight and body size, establish two limits between which we expect to find the rate of heat production [basal metabolic rate] of a horse if we know the rate of heat production of a rat. We expect the metabolic rate of the horse to be somewhere between that of the rat, and that of the rat times the the ratio of horse weight to rat weight, provided of course that we do not regard these two correlations as simply accidental.

If we are firmly convinced that the metabolic rate of horses, and other homeotherms of similar size, is never outside these two limits, then we admit to recognize a natural law between body size and metabolic rate.

This natural law, carefully calculated by Kleiber, is now known as Kleiber’s law. Below is Kleiber’s law graphed out by him as it appeared in his seminal paper. And this is exactly as it appeared in the journal, but with the addition here of colors for better legibility. Since their was no Excel nor graphics software in Kleiber’s time, the graph was hand drawn and appeared in the pages of Physiological Reviews as such. How times have changed.

    

As you look along the line running from lower left to upper right, you can find rats and horses and a host of other mammals including humans. Over the years, mammals that Kleiber didn’t have the opportunity to work on have been measured, and they all fit nicely along Kleiber’s line, following Kleiber’s law. Because of this tight correlation, Kleiber’s equations can be used to precisely estimate the metabolic rate of any animal just by knowing its size.

Aiello and Wheeler used Kleiber’s law as the jumping off point for their grand thought experiment.

Since all animals measured have conformed to Kleiber’s law, Aiello and Wheeler postulated that animals now extinct – including our human and pre-human predecessors – would have fallen along the same line. Using skeletal remains paleontologists have been able to calculate body sizes of extinct animals along with pre-Homo and early-Homo species. Then using Kleiber’s law, it is possible to closely estimate the metabolic rates of these creatures. And here’s where it gets interesting.

According to Kleiber’s law, an australopithecine weighing 80 pounds would have the same metabolic rate as a human weighing 80 pounds despite the disparity in brain size between the two. The much larger brain of the human would have 4-5 times the metabolic rate of the brain of the australopithecine, yet would have the same overall metabolic rate. What gives?

That’s precisely what the authors of “The Expensive-Tissue Hypothesis” wondered.

Because the human brain costs so much more in energetic terms than the equivalent average mammalian brain, one might expect the human BMR [basal metabolic rate] to be correspondingly elevated. However, there is no significant correlation between relative basal metabolic rate and relative brain size in humans and other encephalized animals.

Where does the energy come from to fuel the encephalized brain?

The authors postulated a solution.

One possible answer to the cost question is that the increased energetic demands of a larger brain are compensated for by a reduction in the mass-specific metabolic rates of other tissues.

In other words, if one organ – the brain, for example – is chewing up a lot of energy and contributing a disproportionate amount of the basal metabolic rate for the animal as a whole, then maybe another organ or group of organs are consuming less energy to compensate. The heart, the kidneys, the liver, the skeletal muscles, the GI tract – all consume energy and contribute to metabolic rate. Maybe one of these organs became smaller as the brain became larger over time.

We can hone our analysis a little finer if we begin to look at an energy-balance equation, but an energy-balance equation of a different kind. I have written a number of times in this blog about the energy-balance equation that applies to weight loss: change in weight equals energy in minus energy out. That is not the equation we’ll be talking about here. The other energy-balance equation says that the total metabolic rate is the sum of all the metabolic rates of the various organs and tissues in the body. If you add the metabolic rates of the kidneys, the heart, the brain, the muscles, the digestive tract and so on together, you will get the total metabolic rate of the body, which makes sense because it is the sum of the parts.

Total BMR = brain BMR + heart BMR + kidney BMR + GI tract BMR + liver BMR + the remainder of the body’s tissues.

The authors of the ETH set out to look at the metabolic rates of the various organs. By a diligent search of the literature, they found that along with the brain, the the heart, the kidneys, the liver and the gastro-intestinal tract account for the vast majority of the total BMR. They dubbed these organs as ‘expensive tissues’ because they consume a large amount of energy as compared to their size. (Surprisingly, muscle mass doesn’t contribute all that much to the total metabolic rate (skin and bone contribute even less), which gives the lie to that old notion — that I, myself, have fallen prey to — that replacing fat with muscle increases metabolism significantly.)

Aiello and Wheeler reasoned that if the total metabolic rate stayed the same while the energy-expensive brain grew over time some other expensive tissue had to get smaller. There could be no other solution.

But which of the expensive tissues got smaller?

Aiello and Wheeler examined the data on the metabolic rates and sizes of the various expensive tissues and learned that for a 65 kg primate, the heart, the kidneys, and the liver were approximately the same size as those of a 65 kg (143 lb) human. The greater metabolic rate of the large human brain was compensated for by a GI tract significantly decreased in size. It turns out that the GI tract of a 65 kg human is just a little over half the size of the GI tract of a similar sized primate.

The combined mass of the metabolically expensive tissues for the reference adult human is remarkably close to that expected for the average 65-kg primate, but the contributions of individual organs to this total are very different from the expected ones. Although the human heart and kidneys are both close to the size expected for a 65-kg primate, the mass of the splanchnic organs (the abdominal organs) is approximately 900 g less that expected. Almost all of this shortfall is due to a reduction in the gastro-intestinal tract, the total mass of which is only 60% of that expected for a similar-sized primate. Therefore, the increase in mass of the human brain appears to be balanced by a almost identical reduction in size of the gastro-intestinal tract.

Below is a graphic from the ETH showing the sizes of the different organs as based on predictions from a 65-kg primate and the observed size in humans.

    

So we know that as humans evolved larger brains they simultaneously co-evolved smaller guts in order to maintain a set BMR. And this is where the story gets interesting. Why? Because

the logical conclusion is that no matter what is selecting for brain-size increase, one would expect a corresponding selection for reduction in the relative size of the gut.

Some researchers believe that increasingly complex activities drove the brain to enlarge. As the authors of the ETH summarized it:

The relationship between relative brain size and diet is often mentioned in the literature on primate encephalization and is generally explained in terms of the different degrees of intelligence needed to exploit various food resources. For example, [some] have argued that a relatively large brain and neocortical size correlates with omnivorous feeding in primates, which requires relatively complicated strategies for extracting high-quality foodstuffs. Alternatively, [others] have suggested that frugivores have relatively large brain sizes because they have relatively larger home ranges than folivores, necessitating a more sophisticated mental map for location and exploitation of the food resources.

But it doesn’t matter whether our brains got big because our predecessors were socialized, developed complex foraging strategies, lived in and had to deal with groups or were skilled hunters, in order to obey Kleiber’s law, something had to force our guts to get smaller at the same time. What could that be?

According to Aiello and Wheeler, it was increased diet quality that allowed the gut to get smaller while still absorbing the necessary nutrients to fuel the metabolism. As they put it

The results presented here [in the ETH] suggest that the relationship between relative brain size and diet is primarily a relationship between relative brain size and relative gut size, the latter being determined by dietary quality. This would imply that a high-quality diet is necessary for this encephalization, no matter what may be selecting for that encephalization. A high-quality diet relaxes the metabolic constraints on encephalization by permitting a relatively smaller gut, thereby reducing the considerable metabolic cost of this tissue.

What the authors are saying is that it doesn’t matter how much more brain power was required, the brain couldn’t enlarge without something else giving. What obviously gave was the size of the GI tract, and the only way a smaller GI tract could provide the fuel for the body was to have a higher-quality diet. How did the our most ancient relatives the early hominids increase the quality of their diets?

A considerable problem for the early hominids would have been to provide themselves, as large-bodied species, with sufficient quantities of high-quality food to permit the necessary reduction of the gut. The obvious solution would have been to include increasingly large amounts of animal-derived food in the diet.

Increasing the amount of easily-digested food of animal origin allowed us to shrink our guts while expanding our brains. Had we remained on a diet high in vegetation, we would no doubt not have been able to expand our brains irrespective of how much more thinking those brains would have needed to do. It just wouldn’t have been possible to do so without violating Kleiber’s law.

Take the gorilla, for example, almost pure vegetarians that spend their entire ‘working’ day foraging and eating, which they have to do to get enough calories to maintain their enormous bulk. They have large guts and pay for it by having small brains. Even smaller than that of our most primitive ancestors, the australophthecines.

Gorilla has one of the lowest levels of encephalization of any haplorhine primate, and the much higher level of encephalization of all the australopithecines suggests a diet of significantly higher quality than that of this genus.

Which makes sense when you consider that carbon 13 isotope analysis has shown that Australopithecus africanus (the species that came right after Lucy) consumed meat. As you go up the lineage from Australopithecus and through Homo, you find that more and more meat was consumed the higher up the tree you go.

It’s easy to see that, as compared to humans, chimps and gorillas have large, protuberant bellies, which supports the fact that they have larger GI tracts, but what about our ancient ancestors. All we have to go on are skeletal remains, which show nicely that their heads (and brains) were much smaller than ours, but what about their guts? How do we really know their guts were larger? According to Kleiber, they would have to be, but how to we really know they were?

The large gut of the living pongids gives their bodies a somewhat pot-bellied appearance, lacking a discernible waist. This is because the rounded profile of the abdomen is continuous with that of the lower portion of the rib cage, which is shaped like an inverted funnel, and also because the lumbar region is relatively short (three to four lumber vertebrae).

The drawing below from the ETH shows the inverted-funnel shape of the ribcage of the chimpanzee on the left. You can mentally draw the lines downward from these ribs and envision the pot-bellied look of the abdomen that these primates evidence. Looking at the image on the right, you can see that Australopithecus afarensis (Lucy’s species) has the same inverted-funnel shaped rib cage, indicating a large belly and a low-quality diet.

    

The drawing in the middle is of a modern human. If you extrapolate the lines down from the human rib cage, you can see that they lead to a more narrow waist. Makes you think more of a lean, rangy wolf or other slim-waisted carnivore, whereas the other two don’t.

The authors conclude:

If an encephalized animal does not have a correspondingly elevated BMR [which according to Kleiber, it can’t], its energy budget must be balanced in some other way. The expensive-tissue hypothesis suggested here is that this balance can be achieved by a reduction in size of one of the other metabolically expensive organs in the body (liver, kidney, heart of gut). We argue that this can best be done by the adoption of a high-quality diet, which permits a relatively small gut and liberates a significant component of BMR for the encephalized brain. No matter what was selecting for encephalization, a relatively large brain could not be achieved without a correspondingly [sic] increase in dietary quality unless the metabolic rate was correspondingly increased.

At a more general level, this exercise has demonstrated other important points. First, diet can be inferred from aspects of anatomy other than teeth and jaws. For example, an indication of the relative size of the gastro-intestinal tract and consequently the digestibility of the food stuffs being consumed is provided by the morphology of the rib cage and pelvis. Second, any dietary inference for the hominids must be consistent with all lines of evidence. Third, the evolution of any organ of the body cannot be profitably studied in isolation. Other approaches to understand the costs of encephalization have generally failed because they have tended to look at the brain in isolation from other tissues. The expensive-tissue hypothesis profitably emphasizes the essential interrelationship between the brain, BMR, and other metabolically expensive body organs.

I hope you are now armed with enough knowledge to be able to see through these articles and/or charts that are all too common showing how the GI tract of humans is closer to that of a gorilla than it is to that of a cat or some other carnivore. It seems to me that Aiello and Wheeler have pretty thoroughly demolished the notion that humans are actually designed by the forces of natural selection to be vegetarians. Based on the data and the argument they present, it is actually the opposite: we evolved to be meat eaters.

It was our gradual drift toward the much higher quality diet provided by food from animal sources that allowed us to develop the large brains we have. It was hunting and meat eating that reduced our GI tracts and freed up our brains to grow. As I wrote at the start of this post, the evidence indicates that we didn’t evolve to eat meat – we evolved because we ate meat.

Lierre Keith had it right in The Vegetarian Myth:

The wild herds of aurochs and horses invented us out of their bodies, their nutrient-dense tissues gestating the human brain.

If we evolved because we ate meat, why would we want to stop now?