Perfect Health Diet
1 Why We Start with an Evolutionary Perspective
Why understanding the big picture is so crucial to your health.
An ancient Indian story tells of eight blind men trying to discern the nature of an elephant. Each felt a different part and reached a different conclusion about the nature of the elephant.
The poet John Godfrey Saxe reported the outcome:
And so these men of Hindustan
Disputed loud and long,
Each in his own opinion
Exceeding stiff and strong,
Though each was partly in the right
And all were in the wrong.
Much the way experts quarrel about diet!
Why is it so hard to figure out the optimal diet?
Like the blind men in the fable, diet experts begin with no clear picture of what an elephant looks like and after lifelong investigations acquire only a partial grasp of the evidence. The biomedical database PubMed contains more than 22 million articles, and a million new papers are added each year. A typical scientist reads at most a thousand papers per year. No matter how
long a scientist’s career, it’s impossible to read more than 0.1 percent of the literature. Most of this reading has to be in the scientist’s specialty—a small part of the elephant.
Adding to the problem is the complexity of human biology. We need to get many nutrients, maybe hundreds, from our food. Food contains thousands of toxins. With so many different ways food can nourish or harm us and so many different ways to assemble foods into a diet, picking out which diet is healthiest is like answering a multiple-choice test that has a billion choices. It’s easy to go wrong.
Looking at all this research is like looking at a disassembled jigsaw puzzle with no picture of the completed puzzle. It’s hard to tell how to put the pieces together.
A Big-Picture View We Can Trust
What we really need is a big-picture view—a view of the whole elephant. We need a reliable guide to the optimal diet, a guide that gives us an approximation to the truth at the very beginning of our investigations. This approximate answer can be a lodestar that guides us through the labyrinth of details, preventing many a wrong turn.
This is where an evolutionary perspective comes in. We know that healthy people and animals are more likely to survive the vicissitudes of life and have children and grandchildren. This means that evolution selects for healthful behaviors—including healthful eating.
If we’re looking for a human diet that evolution guarantees is healthful, the place to start is with the diets of the Paleolithic. The Paleolithic was so long—2.6 million years—that Paleolithic man became highly optimized for the Stone Age environment. In the last 10,000 years, mutations have become much more common due to population growth,1
but most beneficial mutations have not had time to become widespread. The historical era has been a period of genetic diversification and emerging but incomplete adaptation to modern life. That means if we want an environment, diet, and lifestyle that will be healthful for all of us, we have to look back to the Paleolithic.
SCIENCE OF THE PHD
Why We Share a Paleolithic Heritage
The Paleolithic began 2.6 million years ago with the invention of stone tools and ended 10,000 years ago with the invention of agriculture. The Paleolithic lasted a hundred thousand generations and was characterized by small populations, typically, tens or hundreds of thousands; at the end of the Paleolithic the human population was 3 million. The modern era has a large population—7 billion today—but evolution has had little time, less than five hundred generations, to work its magic.
We can calculate how long it will take before every possible mutation appears in some person, somewhere. Every child has a similar number of mutations—about 175 new point mutations among the 3 billion base pairs of the human genome.2
• In the Paleolithic, with 10,000 children per generation, it would have taken 8,000 generations, or 160,000 years, for each possible mutation to occur once.
• Today, with more than a billion children per generation, every possible point mutation now appears about twenty times per generation, or almost yearly.
We can also calculate the time required for a beneficial mutation to reach “fixation,” or universal presence throughout humanity. This time is on the order of ln(N)/s, where N is the population size and s is the selection coefficient, a measure of how beneficial the mutation is in terms of expected number of children.3
• In the Paleolithic, a mutation that raised the probability of having an extra child by only 0.1 percent would have reached fixation in 460,000 years. So a mutation with selective advantage of 0.1 percent would have occurred within the first 160,000 years of the Paleolithic, then become
universal 460,000 years later—long before the Paleolithic was over.
• In the modern era, a similar mutation would occur every year but would require 200,000 years to reach fixation. The modern era is less than 10,000 years old, however, so few recently mutated genes have had time to become universal. As a result, our genetic adaptation to the new environment of modern life—agricultural foods, city living, the presence of governments and complex institutions—is incomplete. And human genetic diversity is greater than ever before.
Because mutations that would remove our adaptation to Paleolithic diets have had little time to spread through the population, it is likely that nearly everyone is extremely well adapted to Paleolithic diets. The same cannot be said for modern diets.
Perfect Health Diet
2 The Paleolithic Diet
• Eat real food: recently living plants and animals.
• Eat mostly plants—but low-carb!
• Among plant foods, favor in-ground starches.
• Don’t be afraid to eat fat! Hunter-gatherers flourished on a fat-rich diet.
The premise of “Paleo” diets is that foods hunted and gathered by our Paleolithic (“Old Stone Age”) ancestors represent the healthiest human way of eating, while agriculturally-produced foods may be dangerous to well-being.
There’s solid evidence backing this idea. Direct evidence for the superiority of Paleolithic diets comes from archaeological studies of ancient skeletons. These studies tell us that until the modern era, with our reduced rates of infectious disease, the Paleolithic was the healthiest epoch of human history.
Studies of animals also show that “wild” diets are the healthiest. For example:
• Thirty-two percent of pet cats and dogs are obese,1
but obesity is rare among wild wolves and tigers. It’s not only pets: feral rats living in cities and eating discarded human food have grown increasingly obese in parallel with the human obesity epidemic.2
• Zoo-born elephants live only half as long as elephants living wild in parks such as Amboseli National Park, Kenya.3
Zoo elephants also have much higher rates of obesity than wild elephants. Elephants make a great comparison animal, because they are rarely subject to predation in the wild.
What’s the “wild” human diet? Presumably, the diet obtained the same way wild animals obtain their food: by hunting and foraging in the manner of our Paleolithic ancestors.
READER REPORTS: A Cure for IBS
I’m 62 and have suffered, along with anyone who gets near me, with IBS for the past 25 or so years, and have tried just about every supplement to alleviate the condition without success. Since starting the PHD my symptoms disappeared in less than a week—and haven’t come back. As Billy Crystal would say, “UN beWEEV abo.” Thanks so much.
Paleolithic Health and Neolithic Decline
The tall stature and strong bones of Paleolithic skeletons indicate that Paleolithic humans were in remarkably good health. Paleolithic humans were tall and slender; cavities and signs of malnutrition or stress in bones were rare; muscle attachments were strong, and there was an absence of skeletal evidence of infections or malignancy.4
The adoption of farming in the Neolithic radically changed the diet, and with it came a dramatic loss of health. Farmers needed crops that yielded many calorie-rich seeds from each seed planted, so the harvest could feed the farmer’s family for a year and supply seeds for sowing in the spring. This required a turn of the diet to grains and legumes—foods that, as we shall see, are toxic.
After the adoption of agriculture, stature lessened; smaller tendon
attachments show that muscles weakened; bone and teeth pathologies, such as cavities and osteoporosis, became common; hypoplasias show that periods of malnutrition were common; and signs of infections and inflammation became common.
SCIENCE OF THE PHD
The Neolithic Decline
A large number of journal articles, anthropology Ph.D. theses, and books discuss the collapse of health that is visible with the adoption of cereal grain agriculture.5
A few tidbits:
• Average height dropped, bottoming out at about five feet, three inches for men, five feet for women around 3000 B.C.—about five inches shorter than in the Early Upper Paleolithic.6
• Bones from the Neolithic site of Ganj Dareh in Israel, studied by the anthropologist Anagnostis Agelarakis, showed hypoplasias on the teeth, indicative of malnutrition when young; signs of ear infections and gum inflammation; broken or fractured bones; and arthritis. Those who survived childhood struggled to reach middle age.7
• Nine of sixteen Bronze Age mummies—and seven of the eight of people who died after age 45—in the Museum of Egyptian Antiquities, Cairo, had atherosclerosis.8
The drop in stature persisted throughout the agricultural era until modern times. Only in the twentieth century, with rising wealth and the elimination of many infectious diseases, did humans regain Paleolithic stature.
So Paleolithic diets were quite healthful—agricultural diets, not so much.
We’d better look into what those healthy Stone Age hunter-gatherers were eating!
Paleolithic Plant Foods: Savanna Starches
Many people assume that our distant ancestors resembled chimps and gorillas—forest-dwelling apes who ate fruit. That’s a mistake.
Our ancestors had a long association with open woodlands and tree-spotted grasslands. Where the fossils of human ancestors have been found, tree cover was generally less than 40 percent, sometimes as low as 5 percent.9
Fossils testify that our Paleolithic ancestors lived in open, grassy terrain. Fossil hominids lack the stiff spines and long powerful arms of forest-dwelling apes, and appear to have spent much of their time walking bipedally as grassland dwellers do.10
Ape bipedalism has a long history. Ardipithecus ramidus, which dates from about 4.4 million years ago, spent a significant amount of time walking bipedally,11
as did Oreopithecus bambolii, whose fossils date from 10 to 7 million years ago.12
Another bipedal hominoid dates to 21.6 million years ago.13
Very possibly the common human-chimp ancestor was a bipedal ape living in open terrain, and chimps and gorillas adapted to the forest after they diverged from the human line.
Not only did our hominid ancestors live in wooded grasslands, their food came from grasslands too. This has been proven by a clever method—“isotope signatures” of fossilized bones. Combined with the structure of hominid teeth, this evidence tells us that our ancestors were eating savanna tubers, roots, and corms—foods similar to our modern potato and taro. They had invented the digging stick and were eating starch!
SCIENCE OF THE PHD
How We Know Paleolithic Hominids
Ate In-Ground Starches
Carbon comes in heavy (carbon-13) and light (carbon-12) forms, and grasses and sedges (“C4 plants”) incorporate relatively more carbon-13 than other plants. So the carbon-13 to carbon-12 ratio in a skeleton tells us what fraction of the creature’s food was obtained from grassland plants or animals that ate grassland plants.14
There is considerable variability, but in general grassland plants
predominated in the diet of Paleolithic and earlier hominids. This created a puzzle, known as the “C4 conundrum.” Hominids such as Australopithecus africanus and Paranthropus robustus did not have the right kind of teeth for eating grasses and were not thought to be major hunters of grazing animals, yet their bones show that they got their carbon from grasses. The resolution of the puzzle: those apes were getting their dietary carbon from C4 plant underground storage organs—tubers and corms similar to the modern potato and taro.15
This emphasis on starchy roots, tubers, corms, and rhizomes continued throughout the Paleolithic. Food residues from Upper Paleolithic sites dated to 30,000 years ago show that the grinding of starchy roots and rhizomes into flours and foodstuffs was a common practice.16
Microfossils on Neanderthal teeth from around 44,000 years ago show evidence of the consumption of many roots and tubers, some of which show evidence of cooking.17
Neanderthal consumption of starchy plants goes back at least 250,000 years.18
Modern hunter-gatherers who live in environments that lack starchy plants all trade for starches produced elsewhere. The anthropologist Thomas Headland proposed that it would not be possible for humans to survive in forest environments without such trade; this was debated as the “wild yam question.”19
READER REPORTS: Weight Loss, Improved Energy
I am in the middle of the wardrobe crisis that I’ve been waiting to have for ten years: all my clothes are too big. I don’t mean a little loose; I mean I perpetually look like I’m headed out to an M.C. Hammer costume contest.
Over the past few months I’ve lost 25 pounds. That’s a good thing, since the drop on the scale was a side effect of lifestyle changes that have left me with more stamina and energy than I had when I was 20.
It’s not an exaggeration to say that the Perfect Health Diet changed my life.
A final line of evidence—genetics—supports the idea that our Paleolithic ancestors ate starches. Chimps have two copies of the gene for salivary amylase, the enzyme that digests starches. Humans worldwide average seven copies of the gene; aboriginal peoples eating low-starch diets, such as the rain forest–dwelling BiAka and Mbuti pygmies of the Congo Basin, average 5.4 copies.20
A plausible interpretation is that our Paleolithic ancestors ate enough starch to reach 5 to 6 copies of the amylase gene and that subsequent evolution since the Neolithic invention of cereal grain agriculture has increased the amylase copy number a bit further.
Paleolithic Animal Foods
The Paleolithic began with the invention of stone tools about 2.6 million years ago. These tools were used to hunt animals, tear meat, and cut bones to reach the marrow. Bone marrow consumption is attested from 1.9 million years ago.21
The pursuit of marrow, which is nearly all fat, shows that animal fats were a sought-after part of the early Paleolithic diet.
By 1.75 million years ago, ancestral Homo had spread to northern latitudes, where plant foods are relatively scarce. It is likely these northern hominids were eating a meat-based diet.
By 40,000 years ago, we can tell that Neanderthals (hunting herbivores such as mammoths) and humans (hunting many species with an emphasis on fish) were top-level carnivores. Upper Paleolithic humans weren’t getting protein from plants—no beans for them!—and were higher-level carnivores than wolves and arctic foxes.22
SCIENCE OF THE PHD
Isotope Signatures of Protein Sources
Nitrogen is found in protein and comes in heavy (nitrogen-15) and light (nitrogen-14) forms. Whenever an animal eats protein, it tends to incorporate nitrogen-15 in tissues and exhale or excrete nitrogen-14, so the ratio of heavy to light nitrogen increases by 3 to 4 percent with every step up the food chain.
Unfortunately nitrogen-15 is unstable and is only preserved in bones and teeth from the last 50,000 years, so we have no idea how high on the food chain Australopithecus or Homo habilis were. But nitrogen isotope ratios show that both humans and Neanderthals were at the top of the food chain and getting nearly all their protein from animal food sources.
Another sign that Paleolithic humans were doing a lot of hunting is animal extinctions. The arrival of Paleolithic humans in Australia and the Americas was quickly followed by the extinction of large animal species. Earlier, in Eurasia and Africa, species such as mammoths and saber-toothed tigers were hunted to extinction.
Animal extinctions began at an early date. Between 1.9 million and 1.5 million years ago, Homo erectus appears to have caused the extinction of twenty-three of the twenty-nine known species of large African carnivores.23
The six species that survived were “hypercarnivores,” such as lions and leopards, which ate only meat; the twenty-three that went extinct were omnivores such as civets, which scavenged and ate a wide range of foods. It is thought that they went extinct because they were in direct competition for scavenged carcasses with hominids.24
Subsequent advances in human culture were often followed by new animal extinctions. The extinction of elephants from the Levant around 400,000 years ago was probably due to hunting by archaic humans.25
What Was the Proportion of Animal to Plant Food?
Anthropologists debate the relative proportions of plant and animal food in the diet of our Paleolithic ancestors. Unfortunately, for the earlier part of the Paleolithic there is no evidence that directly answers that question.
We do know that a great expansion of brain sizes occurred during the Paleolithic, and it was probably made possible by new calorie-rich food sources. There are two major theories:
1. Stone tools and cooperative hunting enabled our Paleolithic ancestors to obtain fatty animal foods.26
2. Control of fire enabled our Paleolithic ancestors to cook starchy plants, rendering them less toxic and more digestible. This greatly increased the calories obtainable from plant foods.27
The second theory has been popularized by Richard Wrangham in his book Catching Fire: How Cooking Made Us Human. However, most anthropologists favor the first. The use of stone tools coincided with the brain expansion; while the first known use of fire was 1 million years ago,28
routine use of fire may have begun only 300,000 to 400,000 years ago,29
and more sophisticated use of fire such as heat treatment of tools may have begun 164,000 years ago.30
So the foods driving the brain size expansion during the Paleolithic were probably fatty animal foods.
We do have solid evidence for the diets of modern hunter-gatherers, which probably closely resemble the diets of the Upper Paleolithic. They may be our most useful guide to what a “Paleo diet” for modern humans should look like.
Modern Hunter-Gatherer Diets
The first attempt by an anthropologist to quantify the diets of modern hunter-gatherers was the 1967 Ethnographic Atlas of G. P. Murdock, which was corrected in 1999 by J. P. Gray.31
This looked at 229 aboriginal groups still living in a way that resembled their traditional lifestyle.
The data were analyzed by Loren Cordain and colleagues.32
They found that hunter-gatherers obtained most of their energy from animal foods—meat, fish, and eggs:
• 46 hunter-gatherer groups obtained 85 percent or more of their energy from meat, fish, and eggs, but no groups obtained 85 percent of energy from plant sources. There were no vegetarian hunter-gatherers.
• 133 hunter-gatherer groups obtained 65 percent or more of their energy from meat, fish, and eggs; only 8 groups obtained 65 percent of energy from plants.
• The median group obtained 70 percent of their energy from animal foods, 30 percent from plant foods.
Plant foods contain both carbohydrates and fat. Tropical groups ate the most plant foods, and many of those plant foods, such as nuts, coconuts, and palm fruit, were rich in fat. So carbohydrate intake was well below 35 percent for the overwhelming majority of groups.
The data in the Ethnographic Atlas are dated, and some researchers consider them unreliable.33
Fortunately, detailed studies of the diets of authentic hunter-gatherers have been conducted very recently, and they confirm the results from the Ethnographic Atlas. On our blog, we looked at a study of nine hunter-gatherers—Onge of the Andaman Islands, Anbarra and Arnhem aborigines of northern Australia, Aché of eastern Paraguay, Nukak of south-eastern Colombia, Hiwi of Venezuela, !Kung Bushmen of the Kalahari desert of southern Africa, Gwi Bushmen of Botswana, and Hadza of north-central Tanzania—by anthropologists Hillard Kaplan, Kim Hill, Jane Lancaster, and Ana Magdalena Hurtado.34
Every group ate a substantial amount of meat. Animal foods provided 50 to 85 percent of calories. The !Kung ate the least meat but still averaged 0.57 pounds per day of meat.
Roots and other in-ground plants were the most important plant food. Seeds and nuts were a small contributor for every group but the !Kung, who ate mongongo nuts, a fatty food. “Fruits” were more often fatty nuts than the sugary fruits we are familiar with; for instance, the Nukak ate the palm oil–rich fruit of the palm tree, and the Hadza ate a number of fatty fruits. Only the Gwi consumed a significant amount of sweet fruits, chiefly melons.
In eight of the nine cultures, roots were a much more important source of calories than fruits. Among the Gwi, fruits and roots provided an equal share of calories.
Measured by calories, the diets were generally low in carbohydrates and high in fat. Seven of nine cultures—the Onge, Anbarra, Arnhem, Aché, Nukak, Hiwi, and !Kung—ate 10 to 20 percent carbs. For the Gwi San a majority of calories were carbs, and for the Hadza about 40 percent of calories were carbs. For most groups, fat intake ranged from 40 to 70 percent of calories.
Plant and Animal Food Balance
Although carbohydrates are a small part of calories for many hunter-gatherers, this does not mean they are unimportant. In fact, carbohydrates are a prized part of the diet among modern hunter-gatherers.
Indeed, the Mbuti pygmies of the Congo have two words for hunger: “protein hunger” (ekbelu) and “calorie hunger” (njala). In remote hunting camps on the Ituri plateau of the northeastern Congo, Mbuti generate very high hunting returns and dry large quantities of surplus meat for trade but have no access to starchy plants; in their camps they often complain of njala. Similarly, when the Maku hunters of the Amazon Basin run out of cassava in the forest, no matter how much meat they have, they “have no food.”35
“Drying Out” from Too Few Carbohydrates
I reached my weight loss goals by eliminating grains and limiting dairy to butter and cream and reducing fruit intake. That said, over the last month or so, I was wondering why my body seemed to be drying out from the inside out. I wanted to tweak my diet to optimum health and found your book. The information about the importance of mucin was helpful. What was missing in my diet were the carbs that you recommend. Sweet potatoes, white rice etc. Maybe less protein than I’ve been eating and more saturated fat. . . .
I’m having better results every day. I am fascinated that I have a laboratory of my own body to put your ideas to a test and have them show positive results. Thank you both so much.
—Doris Hames, Atlanta, Georgia
The natural inference is that a healthful diet needs a certain amount of plant foods to balance its animal foods. As we’ll see, starchy in-ground plants are so calorie poor that even obtaining a mere 15 percent of calories from
carbs means consuming more plant foods than animal foods by weight. The Paleolithic diet may have been low-carb, but it wasn’t low-plant.
Takeaway: The Diet of the Paleolithic
The Paleolithic diet was a fat-predominant, low-carbohydrate diet. Calories came mainly from fat-bearing animal foods, but plant foods were an essential part of the diet and comprised most of the weight. Typically:
• Carbohydrates made up 15 to 20 percent of calories, with excursions toward 50 percent depending on food availability. Most calories came from fatty animal foods.
• Plant foods consisted predominantly of starchy in-ground carbohydrate sources such as roots, rhizomes, tubers, and corms plus above-ground fat sources such as coconuts, palm fruit, and mongongo nuts. Sweet fruits were rarely a major part of the diet.
It was on a majority-fat, low-carb diet mainly composed of animal foods and in-ground plants that our ancestors evolved from a regional population of small-brained African apes numbering (probably) in the tens of thousands to a highly intelligent species at the top of the food chain and a global population in the millions.
As our Paleolithic ancestors who dominated the globe were characterized by tall stature and healthy teeth and bones and their health deteriorated as soon as their diet was altered, we think it’s safe to say that such a low-carb, high-plant, starch-meat-and-fat-based diet is a healthful human diet.