Verified by 
Humans can eat just about anything. This is, of course, not because we have the physiology to consume anything, but because we have the intellectual capacity and technology to make almost anything organic consumable. The origins of this ability date back to when our ancestors started making stone tools about 2.5 million years ago. These tools, mostly formed from the sharpened edges of stone flakes, were used to process meat, leaving cutmarks on animal bones that can still be seen on fossil remains. The development of stone tools may have signaled the beginning of a change in our ancestors' diets, with an increasing reliance on non-plant foods.
Substantial increases in brain size in our ancestors began around 2 million years ago. Researchers have long surmised that there was a relationship between brain expansion and meat-eating. However, the fossil and archaeological evidence for an increased reliance on meat-eating may represent just one aspect of an increased omnivory in our ancestors. Richard Wrangham (Catching Fire, Basic Books, 2009) has been a recent advocate for the idea that the controlled use of fire and cooking were critical factors in increasing the availability of protein and calories to our ancestors, allowing them to support an enlarged and energetically expensive brain. Although Wrangham emphasizes the importance of meat and hunting in this scenario, he also stresses that cooking increases the palatability and digestibility of plant materials as well, especially energy-dense roots and tubers.
Evidence for the controlled use of fire is rather sketchy prior to one million years ago. Well before this time, human ancestors such as Homo erectus, whose brain size fits between ourselves and the great apes, had started the expansion of the hominid range out of Africa and into other parts of the Old World. Peter Ungar and his colleagues (2006, Annual Review of Anthropology 35:209-228) argue that more than meat-eating, dietary versatility would have been essential for a species like erectus to be able to expand its range into new environments. Some of these environments were more temperate and seasonal than those in which the species originally involved. The use of stone tools for meat and plant processing was probably essential for such an expansion.
Stephen Cunnane and Michael Crawford (2003, Comparative Biochemistry and Physiology Part A 136:17-26) point out that there is a basic chicken-and-the-egg dilemma in understanding human brain and dietary evolution: large human brains require an enriched diet that we are able to obtain due to our increased intelligence; but our human ancestors were not cognitively advanced when this process started. Cunnane and Crawford ask, what change in dietary behavior could support the evolution of a larger brain without initially requiring a great increase in intelligence?
Crawford, Cunnane, and their colleagues (1999, Lipids 34:S39-S47) have for several years argued that it was the addition of certain foods rich in fatty acids, especially docosahexaenoic acid (DHA) and arachidonic acid (AA), essential components of developing mammalian nervous systems, that allowed relatively small-brained hominids to make the nutritional leap whereby they could support a larger brain. Although AA is available from egg yolks, organ meat, and muscle meat from land animals, the best sources for DHA are fish and shellfish (AA is also present in aquatic animals). Crawford and his colleagues hypothesize that early Homo species exploited the shallows of African lakes and rivers, where a potential abundance of fish and shellfish could be obtained. They argue that this would not require a technological advance but rather should be seen as an expansion of traditional gathering into an aquatic dietary niche. Thus aquatic foods provided a jumpstart for cognitive evolution without requiring a cognitive revolution.
The aquatic food hypothesis has been criticized on several fronts, most notably with the observation that the availability of essential fatty acids is neither ecologically nor metabolically constrained to such an extent that aquatic foods were essential for brain growth or evolution (see my book The Lives of the Brain, Chapter 7, for an overview of the debate). Another implicit argument against it is that there is very little archaeological evidence for the exploitation of aquatic food in the pre-modern-human archaeological record. One could argue that ancient low-density populations might not leave much of an aquatic mark on the landscape. But in contrast (for example) to the vast shell middens that have been left by modern human populations along various seashores, the African archaeological record of 1-2 million years ago is not supportive of extensive aquatic food consumption, even if it is granted that the absence of evidence is not the evidence of absence.
The archaeological evidence may be changing, however. Two recent studies suggest that non-modern-human hominids may have indeed exploited marine resources. Chris Stringer and his colleagues (2008, Proceedings of the National Academy of Sciences 105:14319-14324) have reported on a Neandertal cave site in Gibraltar, which dates to over 40,000 years ago. They have identified an ash layer containing a hearth, Mousterian tools (almost always associated with Neandertals) and knapping flakes, and an abundance of mussel shells derived from a nearby estuary. This short occupation site provides us with a nice snapshot of Neandertal life. As Stringer and colleagues write (p. 14320): "This occupation level...records several activities in the life of the Neandertal occupants. These activities consisted of selection and collection of mollusks, transport of the gathered mussels to the cave shelter, fire making in the cave, the use of heat to open the shells, consumption of these mollusks, knapping on the hearth embers, and subsequent abandonment of the site." At a lower occupation level, further evidence of marine animal exploitation was found in the form of mammals such as seals and dolphins, which were deposited along with terrestrial mammals more typically associated with Neandertal hunting. A few fish remains were also found in these deposits.
The Neandertal evidence from Gibraltar demonstrates once and for all, that modern humans are not the only hominid species who ate from the sea (or lake or river). But this finding does not add much to the antiquity of seafood use-- the Gibraltar Neandertals lived well into the time when modern humans were the dominant hominid species and remaining Neandertals were relegated to the periphery of their original range. Stringer and colleagues even speculate that these Gibraltar Neandertals may have been able to hang on longer than their conspecifics because of their access to both terrestrial and aquatic resources.
A much deeper antiquity for aquatic food use is suggested in another recent study, by Jose Joordens and her colleagues (2009, Journal of Human Evolution 57:656-671). In the early 1890s, the Dutch army surgeon Eugene Dubois discovered the first remains of the species we now call Homo erectus, in Java, at a site called Trinil located on the Solo River. Dating of this site is somewhat controversial, but the timeframe for it is in the range of 900,000 to 1.5 million years; as it is today, Trinil then constituted a riverine environment, not all that distant from lakes, deltas, and the sea. Joordens and her colleagues made a thorough examination of the extensive faunal remains collected by Dubois and others at Trinil. Although the site is most famous for the human ancestors found there, they are vastly outnumbered by fish, mollusks, mammals, birds, and reptiles.
Joordens and her colleagues were concerned first with the issue of whether or not the aquatic environment of Trinil could nutritionally support a hominid species with little technology. Their analysis leaves little doubt that a hominid species could do quite well there, with at least 11 edible mollusk species and 4 fish species manually obtainable from shallow water. But did Homo erectus take advantage of this aquatic larder? Joordens and colleagues suggest that indeed they may have. The genera Pseudodon and Elongaria were the two most abundant of the mollusks in the faunal collections, however their distributions were somewhat unusual. First, rather than being distributed evenly throughout the site, their remains were concentrated in one layer and area. Second, almost all of the specimens were larger adults with juveniles noticeably absent. This was not because the original fossil collectors ignored small or fragmentary material: they were quite meticulous in this regard in general. Rather, Joordens and colleagues hypothesize that Homo erectus may have been the ones choosing the larger adults, eating them, discarding their shells in a limited location, and creating a shell midden to be discovered a million years later. To further test this hypothesis, Joordens and her colleagues are looking at the shells more carefully for signs of handling, such as consistent breakage processing patterns or stone tool cut marks.
The discovery of aquatic food exploitation by Neandertals and the possible systematic consumption of shellfish by Homo erectus certainly expand the temporal and geographical horizons of seafood eating in hominids. The evidence doesn't take us all the way back to the origins of genus Homo in Africa, but it does demonstrate that as brain size increased, there was very likely an expansion of the hominid diet that long predated our appearance on the scene. The development of sophisticated technology has been just one facet of human cognitive evolution. Behavioral plasticity and flexibility are also manifestations of increasing intelligence. The abilities of our ancestors to explore the environment, to productively test novel foods, and to communicate information about food to members of their social group were just as important as any advances in tool-making or -use. Aquatic foods were probably not necessary to make a bigger brain, but our evolution was undoubtedly abetted by our ancestors' willingness to try new foods, including those found near and under water.
