Because the maximum safe level of protein intake for humans is around 50 percent of total calories, the rest must come from fat, such as blubber, or carbohydrates, such as in fruits and roots. Fat is an excellent source of calories in high-latitude sites like the Arctic or Tierra del Fuego, where sea mammals have evolved thick layers of blubber to protect themselves from the cold. However, fat levels are much lower in the meat of tropical mammals, averaging around 4 percent, and high-fat tissues like marrow and brain are always in limited supply. The critical extra calories for our equatorial ancestors therefore must have come from plants, which are vital for all tropical hunter-gatherers. During periods of food shortage, such as the annual dry seasons, fat levels in meat would have been particularly low, down to 1 percent to 2 percent. A carbohydrate supply from plant foods would then have been especially vital.
But if early humans had the same small guts as we do, they could not have obtained their plant carbohydrates without cooking. Recall the poor metabolic performance of the urban raw-foodists in the Giessen study. Those people ate very high-quality cultivated food processed with the aid of sprouting, blending, and even low-temperature ovens, yet still obtained so little energy that reproductive function was seriously impaired. If our early human ancestors indeed ate their plant food raw, they would have needed to find ways of processing it that were superior to our modern technology. But it is not credible that Stone Age people developed non-thermal methods of food preparation more effective than using an electric blender.
Hunter-gatherers living on raw food might sometimes have found plant foods of an exceptionally high caloric density, such as avocados, olives, or walnuts. But no modern habitats produce such foods in abundance all year. Perhaps a few lost places would have had highly productive natural orchards until they were replaced by agriculture, such as the fertile valleys of the Middle East. But occasional productive areas would not explain the wide geographical range of human ancestors across Africa, Europe, and Asia by 1.8 million years ago. Furthermore, seasonal scarcities occur in every habitat and would have forced people to use foods of lower caloric density, such as roots. The notion of a permanently superproductive habitat is unrealistic. People with an anatomy like ours today could not have flourished on raw food in the Pleistocene epoch.
Beyond reducing the size of teeth and guts, the adoption of cooking must have had numerous effects on our digestive system because it changed the chemistry of our food. Cooking would have created some toxins, reduced others, and probably favored adjustments to our digestive enzymes. Very little is known about how our detoxification system and enzyme chemistry differ from those of great apes, but studies should eventually provide further tests of the hypothesis that human bodies are adapted to eating cooked foods.
Take, for example, Maillard compounds, such as heterocyclic amines and acrylamide. These complex molecules are formed from a process that begins with the union of sugars and amino acids, particularly lysine. Maillard compounds occur naturally in our bodies and increase in frequency with age. They occur at low concentration in natural foods but under the influence of heat their concentration becomes much higher than what is found in nature, whether in smoke (from fires or cigarettes) or cooked items. Their presence is easily recognized in the brown colors found in pork crackling or bread crust. Maillard compounds cause mutations in bacteria and are suspected of leading to some human cancers. They can also induce a chronic state of inflammation, a process that raw-foodists invoke to explain why they feel better on raw diets. The cooking hypothesis suggests that our long evolutionary history of exposure to Maillard compounds has led humans to be more resistant to their damaging effects than other mammals are. It is an important question because many processed foods contain Maillard compounds that are known to cause cancer in other animals. Acrylamide is an example. In 2002 acrylamide was discovered to occur widely in commercially produced potato products, such as potato chips. If it is as carcinogenic to humans as it is to other animals, it is dangerous. If not, it may provide evidence of human adaptation to Maillard compounds, and hence of a long exposure to heated foods.
Evolutionary adaptation to cooking might likewise explain why humans seem less prepared to tolerate toxins than do other apes. In my experience of sampling many wild foods eaten by primates, items eaten by chimpanzees in the wild taste better than foods eaten by monkeys. Even so, some of the fruits, seeds, and leaves that chimpanzees select taste so foul that I can barely swallow them. The tastes are strong and rich, excellent indicators of the presence of non-nutritional compounds, many of which are likely to be toxic to humans—but presumably much less so to chimpanzees.
Consider the plum-size fruit of
Warburgia ugandensis
, a tree famous for its medicinal bark.
Warburgia
fruits contain a spicy compound reminiscent of a mustard oil. The hot taste renders even a single fruit impossibly unpleasant for humans to ingest. But chimpanzees can eat a pile of these fruits and then look eagerly for more.
Many other fruits in the chimpanzee diet are almost equally unpleasant to the human palate. Astringency, the drying sensation produced by tannins and a few other compounds, is common in fruits eaten by chimpanzees. Astringency is caused by the presence of tannins, which bind to proteins and cause them to precipitate. Our mouths are normally lubricated by mucoproteins in our saliva, but because a high density of tannins precipitates those proteins, it leaves our tongues and mouths dry: hence the “furry” sensation in our mouths after eating an unripe apple or drinking a tannin-rich wine. One has the same experience when tasting chimpanzee fruits such as
Mimusops bagshawei
or the widespread
Pseudospondias microcarpa
. Though chimpanzees can eat more than 1 kilogram (2.2 pounds) of such fruits during an hour or more of continuous chewing, we cannot. Some other chimpanzee foods taste bitter to us, such as certain figs. Still other fruits elicit in us unusual sensations, such as the fruits of
Monodora myristica,
whose sharp and lemony taste is followed by a numbing sensation at the tip of the tongue like that caused by novocaine. Of the scores of chimpanzee foods I have tasted, I could imagine filling my belly with only a very few species, such as a wild raspberry—but alas, one rarely finds more than a handful of these delicious fruits at a time. The shifts in food preference between chimpanzees and humans suggest that our species has a reduced physiological tolerance for foods high in toxins or tannins. Since cooking predictably destroys many toxins, we may have evolved a relatively sensitive palate.
By contrast, if we were adapted to a raw-meat diet we would expect to see evidence of resistance to the toxins produced by bacteria that live on meat. No such evidence is known. Even when we cook our meat we are vulnerable to bacterial infections. The U.S. Centers for Disease Control and Prevention state that at least forty thousand cases of food poisoning by
Salmonella
alone are reported annually in the United States, and as many as one million cases may go unreported. The estimated total number of cases due to the top twenty harmful bacteria, including
Staphylococcus, Clostridium, Campylobacter, Listeria, Vibrio, Bacillus,
and
Escherichia coli
(
E. coli
), is in the tens of millions per year
.
The best prevention is to cook meat, fish, and eggs beyond 140
o
F (60
o
C), and not to eat foods containing unpasteurized milk or eggs. The cooking hypothesis suggests that because our ancestors have typically been able to cook their meat, humans have remained vulnerable to bacteria that live on raw meat.
Anthropology has traditionally adopted the Man-the-Hunter scenario, proposing our species as a creature that was modified from australopithecines principally by our tendency to eat more meat. Certainly meat eating has been an important factor in human evolution and nutrition, but it has had less impact on our bodies than cooked food. We fare poorly on raw diets, no cultures rely on them, and adaptations in our bodies explain why we cannot easily utilize raw foods. Even vegetarians thrive on cooked diets. We are cooks more than carnivores. No wonder raw-foodism is a good way to lose weight.
CHAPTER 3
The Energy Theory of Cooking
“A man does not live on what he eats, an old proverb says, but on what he digests.”
—JEAN ANTHELME BRILLAT-SAVARIN,
The Physiology of Taste: Or Meditations on Transcendental Gastronomy
A
n obvious implication of animals and humans gaining more weight and reproducing better on cooked than raw diets is that when a food is heated, it must yield more energy. Yet authoritative science flatly challenges this idea. The U.S. Department of Agriculture’s
National Nutrient Database for Standard Reference
and Robert McCance and Elsie Widdowson’s
The Composition of Foods
are the principal sources for public understanding of the nutrient data for thousands of foods in the United States and the United Kingdom, respectively. They provide the data for our food labels. These references report that the effect of cooking on energy content is the same for beef, pork, chicken, duck, beetroot, potatoes, rice, oats, pastries, and dozens of other foods—on average, zero. According to these and similar compilations, cooking has important effects in changing water content and reducing the concentration of vitamins, but the density of calories supposedly remains unchanged whether food is eaten raw or is roasted, grilled, or boiled.
This conclusion is very puzzling. Obviously it conflicts with the abundant evidence that humans and animals get more energy from cooked foods. It also conflicts with various contrary conclusions from nutritional science. On the one hand, a widespread idea states that cooking is “a technological way of externalizing part of the digestive process,” a claim that seems to imply some kind of benefit such as accelerated digestion. On the other hand, cooking is sometimes claimed to have a negative effect on energy value. I recently spotted some small “fresh premium breakfast sausages” in my local supermarket. The food label gave their energy content in calories. With a curious nod to those who might want to eat raw sausages, it included values for both the raw and the cooked product. “Serving size 2 links. Raw 130 cals (60 from fat). Cooked 120 cals (60 from fat).” The claim might seem surprising, but cooking can reduce calories in various ways. Cooking can lead to the loss of nutrient-filled juices. It can generate indigestible molecules such as Maillard compounds, reducing the amount of sugar or amino acids available for digestion. It can burn carbohydrates. It can lead to changes in texture that reduce a food’s digestibility. Leading nutritionist David Jenkins judged such effects significant: “The predominant effect (of cooking) is . . . to reduce the digestibility of the proteins.”
Although different nutritionists say that cooking has no effect on the caloric content of food, or increases it, or decreases it, we can clear up this confusion. As indicated by the evidence from raw-foodists and the immediate benefits experienced by many animals eating cooked food, I believe the effects of cooking on energy gain are consistently positive. The mechanisms increasing energy gain in cooked food compared to raw food are reasonably well understood. Most important, cooking gelatinizes starch, denatures protein, and softens everything. As a result of these and other processes, cooking substantially increases the amount of energy we obtain from our food.
Starchy foods are the key ingredient of many familiar items such as breads, cakes, and pasta. They constitute almost all the world’s major plant staples. In 1988-1990, cereals such as rice and wheat made up 44 percent of the world’s food production, and together with just a few other starchy foods (roots, tubers, plantains, and dry pulses) accounted for 63 percent of the average diet. Starchy foods make up more than half of the diets of tropical hunter-gatherers today and may well have been eaten in similar quantity by our human and pre-human ancestors in the African savannas.
The most direct studies of the impact of cooking measure digestibility, meaning the proportion of a food our bodies digest and absorb. If the digestibility of a particular kind of starch is 100 percent, the starch is a perfect food: every part of it is converted into useful food molecules. If it is zero percent, the starch is completely resistant to digestion and provides no food value at all. The question is, how much does cooking affect the digestibility of starchy foods?
Our digestive system consists of two distinct processes. The first is digestion by our own bodies, which starts in the mouth, continues in the stomach, and is mostly carried out in the small intestine. The second is digestion, or strictly fermentation, by four hundred or more species of bacteria and protozoa in our large intestine, also known as the colon or large bowel. Foods that are digested by our bodies (from the mouth to the small intestine) produce calories that are wholly useful to us. But those that are digested by our intestinal flora yield only a fraction of their available energy to us—about half in the case of carbohydrates such as starch, and none at all in the case of protein.
This two-part structure means that the only way to assess how much energy a food provides is to calculate ileal digestibility, which samples the intestinal contents at the end of the small intestine, or ileum. The procedure requires scientists to conduct research on ileostomy patients, who have had their large intestine surgically removed and have a bag, or stoma, where the ileum ends. Researchers collect the ileal effluent from this bag.