*Disclaimer - Sam is not suggesting nor does he recommend the consumption of any cooked food in the modern human diet, but is simply suggesting that the possible inclusion of cooked food in the diet historically may have had an impact on human anatomy. He argues, however, that human metabolism is still very much adapted towards raw food consumption, and thus terms humans as an "anomaly species."
“Over evolutionary history the adoption of cooking should probably be regarded as one of the largest ever improvements in dietary quality, and one of the largest ever changes in food distribution and availability” (Wrangham and Conklin-Brittain, 2003: 42). There has been much debate as to the diets of our human ancestors since our divergence from a common ancestor with chimpanzees. There is a consensus that while the diets of the proceeding Australopiths still included much of the same leaves, pith, and fruit that its ancestors ate, it also consisted of more meat and much harder plant material, which some suggest to be in the form of underground storage organs. The shift to eating starchy roots was likely a result of the climatic shift that occurred during the late Plioscene where much of the African rainforests became woodlands and savannahs, transitioning from humid, rainy environments to dry, arid environments, a breeding ground for a wide variety of carbohydrate-rich roots. It is suggested that Australopiths were a transition between the last common chimp-human ancestor, who was just starting to get its feet wet (or dried) in the savannahs, and that it wasn’t until early Homo that a full transition was made to the African savannahs. It is clear that a significant change occurred between the early and late Homo lineage, at the onset of Homo erectus, which occurred at around 1.9 million years ago, and it is hypothesized to be a result of the beginning of cooking, specifically that of underground storage organs (USOs) as a fallback food, and this resulted in anatomical and physiological adaptations for cooked starch.
Species, over time, develop anatomical foraging adaptations and specializations to better exploit their niche, and these specializations are made in order to survive periods of scarcity, or what’s eaten in the most dire of times in order to survive (Wrangham et al., 1999). The geographic shift of Australopiths marked a change in their diet, which was somewhere in-between a modern chimpanzee and a hunter-gather human. Thus, it included wild fruits, leaves, pith, possibly a little more meat, and more grass stems and USOs. Granted, during this time, all food was still consumed raw. The most successful of the Australopiths evolved into the Homo lineage. Throughout the Australopithecines and early Homo, oral cavity size and enamel thickness on the teeth began to increase, until late hominids had large, thickly enameled premolars and molars, compared to chimpanzees, who have smaller, thinly enameled cheek teeth. This, along with the Australopiths’ larger chewing surfaces and more heavily pitted teeth compared to extant African apes, suggests a major dietary shift (Laden and Wrangham, 2005). As time progressed, accompanied by the dental changes were, in Homo erectus, an increase in body mass (most notably in females) and brain size, indicating a reliable, high-energy diet (Wrangham et al., 1999).
While apes occupied humid forests and had a fallback food of abundant herbaceous vegetation and soft leaves, increasingly, Australopiths, and eventually entirely, early Homo, occupied an arid environment, and during the late Plioscene (1.8 million years ago [mya]), the start of the evolutionary time of H. erectus, there was a climatic shift that resulted in a large increase in aridity in eastern Africa, the same region that Homo is believed to have evolved in (Snodgrass et al., 2009). Thus, these environmental changes would be most advantageous to the creature that can best exploit the expanded regions, and Homo erectus did just that. Compared to non-human primates and other mammals, humans have a significantly higher quality diet than expected for body size, as a result of an evolutionary commitment to high-calorie foods (Wrangham and Conklin-Brittain, 2003; Snodgrass et al., 2009;). While some suggest that this high-quality diet resulted from increased fruit and/or meat consumption (Leonard et al., 2003; Milton, 2003;), there are multiple problems with these claims. Firstly, the calorie-dense diet of Homo erectus was spurred from the savannah life, where there is decreased access to fruit and leafy vegetation (Laden and Wrangham, 2005) because of the dry climate and the significantly decreased number of trees. An effect of the climatic shift was a marked morphological change as Australopithecus and Homo evolved toward a decrease in upper body arboreal adaptations and an increased use of bipedal locomotion (Wrangham et al., 1999). In other words, it is highly unlikely that H. erectus could have sustained itself on any significant quantity of fruit. The farther along the evolutionary path that the human lineage diverged from the chimpanzee lineage, the more their habitats differed, and as a result, the more their diets differed. Australopithecus bridged a gap between a chimpanzee and human-like creature as they moved out of the rain forests and into the woodlands and savannahs. Savannahs do not have enough herbaceous vegetation to sustain apes (Laden and Wrangham, 2005), ruling out any sort of leafy material as the fallback food of choice. Secondly, large quantities of animal protein without consumption of fat or carbohydrates can be functionally damaging to humans (Wrangham and Conklin-Brittain, 2003). Thirdly, meat has a much lower caloric yield during periods of climatic stress and is much leaner and tougher than the kind we would find in the grocery store (Wrangham et al., 1999). Finally, if modern hunter-gatherers still find it difficult to acquire meat and treat it as more of a delicacy, similar to the way chimpanzees hunt monkeys, (Marlowe and Berbesque, 2009), then it is highly unlikely that meat could have become the staple of H. erectus. What food is abundant and obtainable? The answer lies in underground storage organs, including corms, bulbs, rhizomes, tubers, and caudex (Laden and Wrangham, 2005). As the icing on the cake, cooking USOs would raise energy intake much more than replacing plants with meat (Wrangham et al., 1999).
While other large primates continually consume snacks of raw fruits and leaves, human feeding habits are more similar to carnivores in that we can consume what we would consider meals, instead of constant, all-day feeding. Unlike carnivores, however, instead of meat, plant food continued to be central to the ancestral human diet. While all other apes at the time continued to consume fruit sugars as their base source of energy, Homo likely switched to large starchy underground storage organs (USOs) and cooking improved the digestibility of these foods (Wrangham and Conklin-Brittain, 2003; Lucas et al., 2006;).
The anatomy of the mouth displays many indications supporting the transition to the USO-based diet. There has been identified a mutation that occurred two million years ago in the myosin muscle fiber of the jaw muscles that has significantly reduced the bite force potential of humans compared to all other apes (Lucas et al., 2006). A DNA sequence revealed the human sarcomeric myosin gene responsible for expressing large muscles of mastication, called MYH16, which codes for heavy chain myosin in the jaw (Stedman et al., 2004). A frameshift deletion of an ACC codon was found at codon 660 of the messenger RNA that causes an inactivation of the gene, reducing the size of individual muscle fibers, reducing masticatory muscle size (including temporalis muscle reduction) and causing an overall weakening in jaw muscle strength (Stedman et al., 2004). As a result, humans are forced to rely on genes MYH1 and MYH2, which are less abundant than MYH16. This mutation is only found in humans, and in all humans worldwide (Stedman et al., 2004), meaning that it must have occurred sometime while man was still evolving in Africa. All non-human primates still contain this ACC codon in the coding of the MYH16 gene, and thus, still have the myosin heavy chain and the stronger jaw muscles. Macaques, which still have the intact MYH16 gene, have type II (fast twitch) masticatory muscle fibers eight times larger than humans, and this is a direct result of Homo sapiens’ reliance on the fewer number of MYH1 and MYH2 genes (Stedman et al., 2004). Thus, Homo erectus must have experienced lower forces on its jaw muscles by eating less tough food, likely cooked (Lucas et al., 2006). The advent of cooked food in the human diet resulted in the loss of selective constraints for strong chewing forces that, in the past, would have made MYH16 necessary for the survival of the human ancestral species (Stedman et al., 2004). What is the benefit of decreased chewing strength? Some of the metabolic energy diverted to chewing could now be used in other places, such as towards increasing cranial capacity (Stedman et al., 2004). It is estimated that the MYH16 gene inactivation first occurred 2.4 million years after the Homo lineage diverged from chimpanzees, but was not a hallmark trait until the appearance of Homo erectus in the fossil record (Stedman et al., 2004). Hence, it is no coincidence that the decrease in chewing strength and the increase in brain size coincide, relieving our human ancestors of yet another evolutionary burden (Stedman et al., 2004).
The changes accompanied by the transition to cooked food caused changes in the stress-strain relationship of the tooth-food impact so that there was less work involved in eating (Lucas et al., 2006; Milton, 2006). The incisors are responsible for physically taking the bite out of the food, and thus, they control bite size (Ang et al., 2006). The broad edge of the incisors that are characteristic of anthropoids, including humans, are often referred to as “spatualte” in shape (Ang et al., 2006). That said, frugivorous primates have wider incisors than folivorous primates (Ang et al., 2006). For example, chimpanzees have larger and wider incisors than Howler monkeys. Natural selection tends to favor efficiency in one form or another, given that trade-offs are always made, but it is likely that natural selection acts to make incision into food as efficient as possible in order to minimize the work done during the food fracturing, or biting (Ang et al., 2006). Thus, it is a good guess that, given evolutionary adapted circumstances, teeth will erupt out of the gum in their ideal orientation (Ang et al., 2006). This concept may not be applicable to the westernized world, given the unparalleled strains that modern society, culture, and western diets have placed on humanity, and this may have altered the natural eruption patterns of human teeth. Compared with apes, human incisors erupt vertically oriented, while ape incisors erupt much more procumbently (Ang et al., 2006). Furthermore, the upper incisors of chimpanzees are much larger than those of humans (Ang et al., 2006). Natural selection would likely favor the evolution of incisors in frugivorous primates like chimpanzees that create the most efficient fracturing of food for piercing through fruit skins and peels (Ang et al., 2006). It is clear that apes require more robust incisors than humans, and it is expected that not only are their incisors broader, but that their incisal apex angles are larger to accommodate the necessity of accessing the pulp of many fruits (Ang et al., 2006). However, the drawback of having wider incisal apex angles and the consumption of harder foods is that the incisors will become very blunt and worn with age (Ang et al., 2006). Humans, having characteristically smaller incisors than apes, likely have smaller incisal apex angles, and although the incisors have less strength, they don’t wear as easily and stay sharper for a longer period of time (Ang et al., 2006). This means that they continue to maintain a minimal amount of work effort to process food when worn (Ang et al., 2006). Human and ape comparisons of incisal apex angles, to current knowledge of this paper, have not been tested so this theory is speculative. The difference in not only the size, but in the orientation and definition of the incisors between apes and humans display a clear movement away from fruit as the dietary staple in early Homo. There was also a reduction in the size of the premolar and molar teeth, which are used in chewing (Lucas et al., 2006), and this, as well as the incisal evolution, may be linked to cooking food, which has the effect of easing mechanical particle size reduction, or in other words, easing chewing stresses.
Going hand in hand with these adaptive changes was the recent triplication of human salivary amylase genes, which may point the humans toward an evolutionary dietary history of high starch intake (Lucas et al., 2006). Amylase is produced by both the parotid gland (for use in saliva) and the pancreas (for use in the small intestine), and both aid the breakdown and digestion of carbohydrates. There has been proven a positive correlation between the number of salivary amylase gene copies and the amount of protein expression (Perry et al., 2007). The mean copy number of the salivary amylase gene, AMY1, was found to be greater in populations that consumed high-starch diets (Perry et al., 2007). The fact that both high and low-starch samples were taken from both African and Asian populations and that the resulting data showed that high-starch consuming populations also had the greatest number of salivary amylase gene copies indicates that diet, rather than geographic location, more strongly influences the AMY1 copy number (Perry et al., 2007). However, it should not be ignored that in pre-agricultural times, geographic region and environment impacted and, to a great extent, imposed dietary constraints on animal species. It is also important to note that a significant amount of starch digestion occurs in the mouth (Perry et al., 2007), and that an increased AMY1 copy number further increases the efficiency by which starch foods are digested in the mouth, stomach and small intestine (Perry et al., 2007). The reason that salivary amylase continues to impact digestive capabilities past the mouth is because of its ability to survive the stomach’s acidic environment and its fully intact passage into the intestines. This results in enhanced digestion by augmenting the enzymatic activity of pancreatic amylase in the small intestine (Lucas et al., 2006; Perry et al., 2007). In essence, the increased amylase release by the parotid salivary glands helps to predigest the starchy carbohydrates in USOs (Laden and Wrangham, 2005). Cooking also substantially increases amylase’s activity on starch, especially if the food is eaten while hot, and duplicating the amylase gene would increase the amount of enzyme available to aid in the breakdown of the food (Lucas et al., 2006).
A study done on chimpanzees found that among chimps there are only two different variations in the number of AMY1 copies. Furthermore, bonobos have a disrupted AMY1 coding sequence and, as a result, those genes may be non-functional and bonobos may not produce any salivary amylase at all (Perry et al., 2007). While the number of AMY2 gene copies, which influence the amount of pancreatic amylase produced, is the same in both humans and chimpanzees, on average, humans have three times more AMY1 copies than do chimpanzees. This makes it more likely that AMY1 copy number was gained in the human lineage rather than lost in chimpanzees (Samuelson et al., 1990; Perry et al., 2007). Because AMY1 gene copy number is directly correlated with the amount of salivary amylase protein produced, this explains why salivary amylase levels are approximately six to eight times higher in humans than in chimpanzees (Perry et al., 2007). This amplification of the human salivary amylase gene occurred post human-chimpanzee divergence (Samuelson et al., 1990), and this makes sense because both chimpanzees and bonobos are largely frugivorous and ingest relatively little starch compared with most human populations. Increased salivary amylase production may also result in increased fitness with regards to the ability to live through diarrheal and other intestinal diseases by enhancing energy absorption, making energy loss via diarrhea less taxing (Perry et al., 2007). Therefore, it’s possible that this gene duplication increased fitness post-Neolithic revolution with the onset of high population-density living where infectious diseases were a problem and where grain consumption became widespread. However, one problem with this theory is that, while on average, high starch consuming populations have just under seven AMY1 gene copies, low starch consuming populations have over 5 gene copies. This general similarity in the proximity of average gene copies between high and low-starch populations indicates that this genetic mutation occurred far before the advent of modern settlements, and more than likely on the African savannah. Thus, the drastic difference in the number of AMY1 copies may suggest a clear divide in our evolutionary history with all other apes, directing evidence toward and highlighting the importance of the consumption of starch-rich USOs (Perry et al., 2007).
USOs are adapted to adverse growing conditions, which means that they are extremely diverse and abundant in woodlands and savannahs because they have long dry seasons and/or little rainfall, and the USOs store food and/or water during these dry, stressful periods (Wrangham et al., 1999; Laden and Wrangham, 2005). Conversely, rainforests have a lower biomass of USOs where they are also less diverse because rainforests have only short dry periods and a significant amount of rainfall (Laden and Wrangham, 2005). Furthermore, USOs in the savannah have increased edibility compared to rainforest USOs (Laden and Wrangham, 2005).
Most living examples of tropical/subtropical hunter-gatherer tribes eat mainly plant foods (Wrangham et al.,1999), and there are many tribes in east and southern Africa that consume USOs as fallback food. The G/wi, who live in the Kalahari desert, consume USOs when sweet fruits, their preferred food, is not available. During this time they are forced to survive on four species of tuber, and as a result their body weights drop, illness is more frequent, and some people starve, all aspects characteristic of fallback food consumption in survival mode (Laden and Wrangham, 2005). A second living example is the !Kung San, who also live in the Kalahari desert in parts of Nambia and Botswana. They consume tubers during the winter dry season when preferred foods are not available (Laden and Wrangham, 2005). Similarly, the Hadza of northern Tanzania also consume tubers during the late dry season as well as during the main rainy season (Laden and Wrangham, 2005). The Hadza are a prime candidate for a Homo erectus comparison because, not only do they live in part of the region where humans spent as much as two million years of their evolutionary history, but for the Hadza, tubers are most continuously available (many are available all year) out of all the foods they consume, but they are least preferred and have the lowest caloric density of all of the Hadza’s foods (Marlowe and Berbesque, 2009). Furthermore, Hadza women’s BMI and body fat percentage drop when more tubers are taken, another characteristic of USOs as the Hadza’s fallback food (Marlowe and Berbesque, 2009).
In contrast, arctic populations, who consume extremely high proportions of, if not exclusively, animal foods, do not have any genetic adaptations to aid in the processing of such a diet, making it highly unlikely that humans evolved consuming a diet high, animal-based diet (Milton, 2000).
The Hadza are in serious competition with mole rats for deep tubers, and mole rat C3/C4 ratios overlap with both Australopithecines and Homo erectus (Laden and Wrangham, 2005; Marlowe and Berbesque, 2009). From 6.0-1.5mya, mole rats consistently appear in the fossil record with hominids, suggesting that they all shared not only the same habitat, but also had similar diets, including significant tuber consumption (Laden and Wrangham, 2005). Other than mole rats (and warthogs) (Marlowe and Berbesque, 2009), there were likely hardly any other animals competing to eat USOs, making them a trustworthy food source (Wrangham et al., 1999).
Beyond anatomical and technological adaptations to food consumption, digestive physiology is a major indicator of specific dietary adaptations. Human digestive anatomy is distinct from that of great apes in a few key ways. Firstly, humans have a small digestive tract for an animal of comparable size (Milton, 2006). More specifically, we have a smaller gut volume, a longer small intestine, a smaller cecum and colon, and a faster gut passage rate (Wrangham and Conklin-Brittain, 2003). This differs from apes in that in humans the small intestine makes up the largest portion of the digestive tract, while in apes the colon is the largest (Milton, 2006). This suggests an element of refinement in the human diet when compared with apes, such as a history of consuming “predigested” or processed food, a decreased ability to handle and process fiber, and a greater ability to uptake and absorb more nutrients. During scarce times when humans would be forced to consume lower-quality, less calorically dense foods, the digestive tract moves the food through more quickly to enable the feeder to consume a larger volume of food (Milton, 2006). Since modern humans have retained this evolutionary trait, this suggests that Homo remained dependent on fallback foods throughout its entire evolution. Even with the ability to increase metabolism, the human gut still has a distinct, finite capacity on the maximum amount of food that can be processed in a single day (Milton, 2000), further indicating limits on Homo’s ability to tolerate excessive amounts of fiber. While there is clear evidence to support that human digestive morphological adaptations still greatly resemble that of plant eaters (Milton, 2006), the high-quality hominin diet has often been attributed to a large increase in meat consumption (Milton, 1999; Leonard, 2002). The problem with this theory is that carnivorous animals display specific all-meat dietary adaptations, like extremely high protein requirements for maintenance and growth, strange gluconeogenesis patterns, and lack of ability to synthesize vitamin A and niacin (Milton, 2000). Therefore, it is more likely that H. erectus continued to focus the mainstay of its diet around plants.
Some humans purport that a raw food diet is what humans, like all other animals, are adapted to consuming. Putting health factors aside, it appears that raw food diets are only possible on the high-energy dense, low-fiber, domesticated fruits and vegetables provided by modern agriculture. A similar diet in the wild would not only be near impossible to meet human caloric needs, but the immense amount of fiber in wild fruits and herbs that would be necessary to consume for sustenance would surely exceed the human colon’s ability to process raw fiber (Wrangham, 2003; Milton, 2006).
While there is a lot of evidence to support the human adaptation to cooked starch hypothesis, there are some significant holes in the theory. One problem with the idea of USOs as early Homo’s fallback food is that it is necessary to have a very effective digging tool to unearth most tubers, which lie two feet underground (Marlowe and Berbesque, 2009). The Hadza use fire-hardened, sharpened branches as their digging sticks to dig up tubers, although it is unknown what the Australopithecines and earlier Homo lineage might have used (Marlowe and Berbesque, 2009). While humans may be lacking anatomical adaptations for accessing food underground, nonetheless, based on parallels in digging adaptations between bears, pigs, and humans, all are adapted to exploiting plant roots and tubers as food (Laden and Wrangham, 2005). While bears have non-retractable claws, pigs a cartilaginous snout, and humans a digging stick, they all functionally do the same thing (Laden and Wrangham, 2005). Another problem with this hypothesis is that there is no strong evidence of controlled use of fire before 250,000 years ago, although there are sites in South Africa and Kenya possibly dating back as early as 1.6mya (Wrangham and Conklin-Brittain, 2003). However, regardless of the time when cooking began, a typical speciation event generally takes only 15,000-25,000 years, and some mammalian species can evolve in 5,000 years, but if humanity’s adaptation to cooked food came very recently then this leaves question as to what prompted all of the previously stated anatomical and physiological changes found in Homo erectus (Wrangham and Conklin-Brittain, 2003).
Every species has a specific niche of which its specialized adaptations allow it to exploit. The sharks and fish exploit the oceans, the monkeys and apes exploit the rainforests, and man? Man exploits the savannahs. In an age where humans are so far removed from an evolutionary past time, it is so difficult to unveil exactly what happened in the past six million years that everything, at best, is a guess with the evidence available. With that in mind, some of us are better guessers than others, and some are better researchers than others. Placing reliance on the accuracy and accountability of the anthropological field, for now, man will eat cooked starch.
Literature Cited
Ang KY, Lucas PW, Tan HTW. 2006. Incisal orientation and biting efficiency. J Hum Evol 50:663-672.
Leonard WR. 2002. Food For Thought. Sci Am 287(6):106-115.
Lucas PW, Ang KY, Sui Z, Agrawal KR, Prinz JF, Dominy NJ. 2006. A brief overview of the recent evolution of the human mouth in physiological and nutritional contexts. J Physiol Behav 89:36-38.
Marlowe FW, Berbesque, JC. 2009. Tubers as Fallback Foods and Their Impact on Hadza Hunter-Gatherers. Am J Phys Anthropol 140:751-758.
Milton K. 1999. A Hypothesis to Explain the Role of Meat-Eating in Human Evolution. Evol Anthropol 8(1):11-21.
Milton K. 2000. Hunter-gatherer diets – a different perspective. Am J Clin Nutr 7(3):665-667.
Milton K. 2006. Diet and Primate Evolution. Sci Am 128(2):22-29.
Perry GH, Dominy NJ, Claw KG, Lee AS, Fiegler H, Redon R, Werner J, Villanea FA, Mountain JL, Misra R, Carter NP, Lee C, Stone AC. 2007. Diet and the evolution of human amylase gene copy number variation. Nat Genet 39(10):1256-1260.
Samuelson LC, Wiebauer K, Snow CM, Meisler MH. 1990. Retroviral and Pseudogene Insertion Sites Reveal the Lineage of Human Salivary and Pancreatic Amylase Genes from a Single Gene during Primate Evolution. Mol Cell Bio 10(6):2513-2520.
Snodgrass JJ, Leonard WR, Robertson ML. 2009. The Energetics of Encephalization in Early Hominids. In: Hublin J-J, Richards MP, editors. The Evolution of Hominin Diet: Integrating Approaches to the Study of Paleolithic Subsistence. New York: Springer. p 15-30.
Stedman HH, Kozyak BW, Nelson A, Thesier DM, Su LT, Low DW, Bridges CR, Shrager JB, Minugh-Purvis N, Mitchell MA. 2004. Myosin gene mutation correlates with anatomical changes in the human lineage. Nature 428:415-418.
Wrangham R, Conklin-Brittain NL. 2003. Cooking as a biological trait. Comp Biochem Physiol, A 136:35-46.
Wrangham RW, Jones JH, Laden G, Pillbeam D, Conklin-Brittain NL. 1999. The Raw and the Stolen: Cooking and the Ecology of Human Origins. Curr Anthropol 40(5):567-594.
Wrangham R, Laden G. 2005. The rise of the hominids as an adaptive shift in fallback foods: Plant underground storage organs (USOs) and australopith origins. J Hum Evol 49(4):482-489.
