When I wrote the Overcoming the Curse of the Mummies chapter in Protein Power, I wrote mainly about the evidence of disease found in the mummies of ancient Egyptians and correlated this disease with their high-carbohydrate diet. Along with all the material on mummies, which is the part everyone seems to remember, I wrote about a study done in the United States in the 1970s that persuasively demonstrated the superiority of the hunter diet as compared to an agricultural diet, which no one seems to remember. I came across that study a couple of days ago and decided to present it in a little more detail than I was able to in Protein Power.
The anthropological record of early man clearly shows health took a nosedive when populations made the switch from hunting and gathering to agriculture. It takes a physical anthropologist about two seconds to look at a skeleton unearthed from an archeological site to tell if the owner of that skeleton was a hunter-gatherer or an agriculturist.
Unlike the Egyptian mummy data, there is usually no soft tissue material left when remains of early man are found. But the skeletal remains of hunter-gatherers show them to be much healthier than agriculturalists. Hunter-gatherers had better bones, had no signs of iron-deficiency anemia, no signs of infection, few (if any) dental cavities, fewer signs of arthritis and were in general larger and more robust than their agriculture-following contemporaries. One of the theories as to why postulates that hunter-gatherers lived in smaller, more mobile societies. Consequently, they weren’t as likely to get communicable diseases and were able to travel to find food, whereas agriculturists were rooted to one spot, lived in larger groups, making the spread of disease more likely, and they were subject to lack of food if a drought or other natural disaster decimated their crops.
The study we’re going to look at today is unusual in several respects. First, there is a large amount of data, i.e., a lot of skeletons of both groups. Second, it compares sedentary hunter-gatherers to sedentary agriculturalists. And it compares peoples who probably had the same genetic heritage to one another. Finally, it compares hunter-gatherers to agriculturalists living in the same general area. The only real difference between the two groups of people is the time in which they lived and diet.
The group of agriculturalists lived in an area called Hardin Village, which is a famous archeological site located in Kentucky on the bank of the Ohio River across from the current day city of Portsmouth, Ohio. These people farmed the area from about 1500 AD to 1675 AD. There is no indication in the archeological record of any European contact with these Hardin Villagers.
The hunter-gatherers lived in the same general area in an archeological site called Indian Knoll, which is a large midden (an ancient refuse heap) located on the Green River in western Kentucky. Carbon-14 dating dates the age of habitation of these hunter-gatherers to about 5000 years ago. Based on the excavation of the deep midden, these people lived at this site for a long period of time, i.e., they stayed in one spot instead of roving as most hunter-gatherers did.
Writes Claire Cassidy, Ph.D., author of the study:
Available fauna and flora, water, and climate were so similar in the two areas that it may be assumed that whatever natural stresses existed at one site were probably existent at the other also, and therefore, in themselves, these should not affect the health and nutrition differently.
Population size and degree of sedentarism affect disease spread. In the cases of the Hardin Village and Indian Knoll, since both are sedentary or semisedentary, this variable should be negligible in explaining differences in disease experience between the sites.
Archeological-reconstructable variability in material culture is also fairly small (though Indian Knollers used the spear-thrower and spear, while Hardin Villagers had pottery, permanent houses, and the bow and arrow). Thus, in all probability the most significant difference between these two populations is in subsistence technique, with agriculture at the later site, and hunting-gathering at the earlier.
So, we have two societies who both lived in the same area and didn’t move around much, if at all. One lived by agriculture and one lived by hunting and gathering. Genetically they were probably the same, although there is no way to tell for sure. Both groups had the same climate, weather, water, etc. Neither group had contact with Europeans, so there is no contamination that way. The groups are separated by only diet and time. The hunter-gatherers lived in the area approximately 3500 years before the farmers did and had a substantially different diet.
(When I first went through this paper, I went nuts trying to remember whether the Hardin Villagers were the agriculturalists or whether the Indian Knollers were. After whipping back to the start of the paper a half dozen times to check and recheck on it, I decided to come up with a mnemonic for it. I started thinking of the Villagers part of Hardin Village as being farmers. Village = farm. Farmers live in villages – at least they do in my mind, so it’s easy for me to remember that the Hardin VILLAGERS are farmers. Readers of this blog are probably smarter than I am and won’t have to go to such lengths, but if any do, this will help.)
What did these folks eat?
At Hardin Village, primary dependence was on corn, beans, and squash. Wild plants and animals (especially deer, elk, small mammals, wild turkey, box turtle) provided supplements to a largely agricultural diet. It is probable that deer was not a quantitatively important food source… At Hardin Village, remains of deer were sparse.
At Indian Knoll it is clear that very large quantities of river mussels and snails were consumed. Other meat was provided by deer, small mammals, wild turkey, box turtle and fish; dog was sometimes eaten ceremonially.
There are several other dietary differences. The Hardin Village diet was high in carbohydrates, while that at Indian Knoll was high in protein. In terms of quality, [some] believe that primitive agriculturalists got plenty of protein from grain diets, most recent [researchers] emphasize that the proportion of essential amino-acids is the significant factor in determining protein-quality of the diet, rather than simply the number of grams of protein eaten. It is much more difficult to achieve a good balance of amino-acids on a corn-beans diet than when protein is derived from meat or eggs. The lack of protein at the Hardin Village signaled by the archaeological data should prepare us for the possibility of finding evidence of protein deficiency in the skeletal material.
The Hardin Village site yielded 296 skeletons and the Indian Knoll site 285.
What did this skeletal data show? Let’s take a look.
Based on the ages of the people whose skeletons were found (anthropologists can easily tell age from skeletal remains), the life expectancies for people of both sexes and all ages were lower at Hardin Village as compared to Indian Knoll. And infant mortality was higher at Hardin Village as well.
Iron-deficiency anemia of sufficient duration to cause bone changes was present at Hardin Village but absent at Indian Knoll. And half the cases of serious iron-deficiency anemia occurred in children at Hardin Village.
Iron-deficiency anemia is a true deficiency disease, often an accompaniment of low-meat diets, long-term infection, or chronic disease. It is also frequently found in cases of protein-energy malnutrition. The classic sign of iron-deficiency anemia presents as a couple of conditions seen in the skull called porotic hyperostosis and cribra orbitalia. 8.2 percent of the Hardin Villagers had iron-deficiency anemia severe enough to cause one or both of these conditions. These conditions are extremely painful and those afflicted had to have been miserable, especially the children, most of whom were under five years old.
There were signs of malnutrition in both populations, but the signs differed between them.
There are a couple of ways anthropologists look for periods of malnutrition. One is by examining the tibias (lower leg bones) with X-ray looking for a finding called Harris lines (or growth arrest lines).
All these Harris lines indicate is that an episode of malnutrition occurred during childhood while the bones were developing, causing a period of growth arrest that lasted at least ten days or more. But since these lines appear after the period of malnutrition, they can’t provide information as to the total duration of the lack of food. The total number of lines found tells approximately how many episodes of dietary lack occurred that were serious enough to halt bone growth.
To determine the severity of periods of malnutrition, anthropologists look for enamel hypoplasia. Enamel hypoplasia derives from periods of ill-health or hunger lasting long enough to interrupt the deposition of enamel on the teeth. These defects, like Harris lines, represent periods of growth arrest in childhood, but unlike Harris lines, enamel hypoplasia quantifies the severity of the period of malnutrition. The worse the defect, the worse the malnutrition.
Interestingly, there were more Harris lines found in the specimens from Indian Knoll, but these lines were regularly spaced, “indicating that malnutrition occurred at periodic intervals, perhaps as a “normal part of life.” There were an equal number of jaws at both sites demonstrating teeth with enamel hypoplasia, “but the frequency of severe episodes of arrest was significantly higher at Hardin Village.”
The most parsimonious interpretation of this information is that mild food shortages occurred at regular intervals at Indian Knoll; perhaps late winter was a time of danger. [Researchers] using growth arrest lines [Harris lines] and … archaeological data, have similarly concluded that in the hunter-gatherer populations they studied, food shortages occurred regularly, probably on a yearly basis. At Hardin Village growth arrest was caused by illnesses or crop failure which resulted in long-lasting, but randomly-occurring episodes of growth arrest.
Bones can also exhibit signs of certain types of infection. Bone infections affected an equal number of people at both sites, but affected significantly more children at Hardin Village than at Indian Knoll.
A specific type of infectious disease showing up in skeletal remains and identified as the syndrome of periosteal inflammation was present at both sites, but was thirteen times more common at Hardin Village. No one knows for sure what causes this disorder, but it is thought to be caused by a treponematosis, a disease caused by a similar but not identical agent as that that causes yaws, pinta or even syphilis.
The author of this study attributes the greatly increased incidence of this disease in the Hardin Villagers to “lack of resistance in the host because of poor diet and general health.”
Teeth are often a window into the diet of ancient populations. Based on the wear patterns and number of caries (dental cavities), teeth can provide much information on the quality of the diet. Teeth ridden with decay are typically associated with poor quality diets, and the unhealthy teeth themselves can be a major factor in the overall poor health of an individual.
Tooth decay was rampant at Hardin Village, but uncommon at Indian Knoll. Adult males at Hardin Village had an average of 6.74 carious teeth per mouth, while at Indian Knoll the corresponding frequency was 0.73 per mouth. For women the rates were 8.52 and 0.91 per mouth respectively. No Indian Knoll children under twelve years of age had caries, whereas some Hardin Village children already had developed caries in milk teeth in their second year of life. Tooth decay is closely associated with sugar content and consistency of food, occurring with higher frequency in sweet or high carbohydrate diets which are soft and sticky.
Here is the summary of the findings of this analysis of skeletal data as tabulated by the author:
1. Life expectancies for both sexes at all ages were lower at Hardin Village than at Indian Knoll.
2. Infant mortality was higher at Hardin Village.
3. Iron-deficiency anemia of sufficient duration to cause bone changes was absent at Indian Knoll, but present at Hardin Village, where 50 percent of cases occurred in children under age five.
4. Growth arrest episodes at Indian Knoll were periodic and more often of short duration and were possibly due to food shortage in late winter; those at Hardin Village occurred randomly and were more often of long duration, probably indicative of disease as a causative agent.
5. More children suffered infections at Hardin Village than at Indian Knoll.
6. The syndrome of periosteal inflammation was more common at Hardin Village than at Indian Knoll.
7. Tooth decay was rampant at Hardin Village and led to early abscessing and tooth loss; decay was unusual at Indian Knoll and abscessing occurred later in life because of severe wear to the teeth. The differences in tooth wear and caries rate are very likely attributable to dietary differences between the two groups.
Her analysis based on this data:
Overall, the agricultural Hardin Villagers were clearly less healthy than the Indian Knollers, who lived by hunting and gathering.
The author raises a couple of interesting questions about the diet of early populations and the drive to eat carbohydrates in place of real food once the taste is acquired. Before we get to these interesting issues, however, I want to delve into a sad situation that obviously prevailed in the Hardin Villagers and continues to be present in some modern day agriculturalists.
Below is a chart from the paper showing the life expectancies by age of people living in Hardin Village and Indian Knoll. Look at the enormous increase in mortality in the agricultural Hardin Villagers between the ages of two to four.
Why this rapid increase in mortality in these young children. The author tells us:
The health and nutrition situation at Hardin Village may profitably be compared with that in modern peasant villages. In may of these, children are typically fairly healthy until weaned. At this time they are introduced to a soft diet consisting largely of carbohydrates (in much of Africa and Central America, a pap is made of sugar, water, and maize flour: in Jamaica green bananas replace maize). In many cases, within a few weeks or months these children develop diarrhea, lose weight, suffer multiple infections, and may eventually develop the form of protein-energy malnutrition called kwashiorkor. In this disorder caloric intake is usually adequate, but protein and other nutrient intakes are extremely limited; without modern hospital care many victims die.
At Hardin Village the highest rate of death occurs between the second and fourth years of life. This is typical for a population experiencing weaning problems. Considering the softness of the adult diet and the high caries rate of both children and adults, it is not unlikely that the children were weaned onto a corn pap of some type.
The high prevalence of childhood infection, severity of growth arrest in the first few years of life, and the existence of iron-deficiency anemia all point to a situation at Hardin Village analogous to those in modern peasant villages. In other words the evidence supports a hypothesis that malnutrition began with weaning at Hardin Village, sometimes resulted in kwashiorkor, and continued at low level – just enough to reduce the resistance of the population to infectious disease – throughout the life of the individual.
Think about this the next time you hear a pediatrician recommend that babies who are being weaned start out on some sort of Pablum or other processed cereal for infants. And they virtually all recommend it. Our grandchildren’s pediatrician recommended it, but MD and I interceded after the first feeding. From then on they all got pureed turkey, pork, chicken or other meat along with pureed vegetables. They had no grain.
Dr. Cassidy, the author of this fascinating paper, speculates in the discussion section about why a society would abandon hunting and gathering for agriculture when the diet quality provided by an agricultural subsistence is so inferior. She writes about the possibility of all the game being decimated by over hunting, and she mentions the possibility of inter tribal warfare reducing the male hunting population to the point that those remaining standing couldn’t provide enough food for all by hunting alone. Then she gets to the heart of the matter, and asks some questions that are pertinent not just to ancient agricultural societies, but to us today.
Thus population expansion, inefficient hunting techniques, loss of game from the area by migration and overkill, and warfare, all may have contributed to force the Hardin Villagers to become more and more dependent on a small number of high-carbohydrate agricultural foods of limited quality, and this may have been so even were they aware of an increase in physical ill-health in the group.
Finally, we must also wonder if people didn’t ultimately begin to prefer corn and beans to meats? There is some evidence that carbohydrates can become so palatable to humans that they eat them in preference to other foods; such a situation may have further limited the appeal of hunting.
If this is the case, the Hardin Villagers are not the only society in history who have chosen carbohydrates in preference to other foods. And they certainly aren’t the only ones to prefer corn and beans to meats. I would venture that most people today prefer carbs to meat, a notion that is confirmed in the nutritional data. Carbs play a far larger role in the American diet than do meats of all kinds. And if many so-called nutritional experts had their way, we would all eat even more.
The next time you may be tempted by the siren song of the high-carb pushers, remember what happened to the Hardin Villagers and do the Nancy Reagan: Just say no.
*Cassidy CM. Nutrition and health in agriculturalists and hunter-gatherers: a case study of two prehistoric populations. in Nutritional Anthropology. Eds Jerome NW et al. 1980 Redgrave Publishing Company, Pleasantville, NY pg 117-145