Reconstruction
FwJj 20 lies in the Koobi Fora Formation east of Lake Turkana in northern Kenya, and it is dated securely to 1.95 million years ago. What makes it exceptional is the combination: faunal remains carrying direct evidence of butchery, found in place with Oldowan stone artefacts, in a well-watered habitat. Most sites of this age offer tools or bones, and inferring the connection between them is the perpetual difficulty. Here the association is in situ, and it is the oldest such evidence for hominins predating Homo erectus. The diet it records is broader than expected. Alongside terrestrial mammals, the assemblage contains butchered turtles, crocodiles and fish. This is the earliest definitive evidence of aquatic animals in the hominin diet, and it changes the picture of early Homo from a scavenger contesting savannah carcasses to a forager also working shallow water. The practical logic is compelling. A large antelope carcass is guarded by lions, hyaenas and vultures, and reaching one early means real physical danger. A turtle is slow, defenceless and can simply be picked up. Fish concentrated in drying pools require no pursuit. Crocodiles are dangerous but their nests and young are not, and a stranded or dead one is straightforwardly available. These are low-risk, low-skill foods, accessible to juveniles, the elderly and pregnant women as readily as to the strongest adults, which matters for a species whose offspring are dependent for years. Nutritionally, aquatic animals carry fats that terrestrial game largely lacks, and the paper notes they are rich in nutrients relevant to brain growth. This connects to a long-running hypothesis that aquatic resources supported the encephalisation of the human lineage. The site's evidence is real and the nutrient chemistry is not in question, but the causal step from availability to brain expansion remains a hypothesis rather than a demonstrated mechanism, and it should be read that way. There is a further reason this matters. Early Homo is conventionally described through meat, with the tool-and-carcass story running from Gona and Bouri through Olduvai. FwJj 20 shows that the same toolkit was applied to whatever a landscape offered, including resources that required no confrontation with a predator. The dietary breadth already visible in australopithecine tooth chemistry appears to persist into early Homo, now extended by technology rather than replaced by it. Flexibility, not specialisation in big game, is what the record actually supports at this date, and it is flexibility that carried the genus out of its original habitats and eventually out of Africa.
Historical context
Around 1.95 million years ago, early Homo shared eastern Africa with Paranthropus boisei, and Homo erectus had not yet appeared. Fire was not in controlled use. The Turkana basin held a large lake fed by rivers, with deltas, floodplains and gallery woodland supporting hippopotamus, crocodile, turtle and fish alongside antelope and elephant. Oldowan technology was in use across the region. No hominin had left Africa, though the first would reach the Caucasus within about a hundred thousand years.
Evidence
Archaeological
StrongOldowan artefacts occur in situ with butchered faunal remains, an association rarely this secure at this age.
Zooarchaeological
StrongCut marks and breakage on turtle, crocodile, fish and mammal remains directly evidence hominin processing rather than carnivore or natural damage.
Food identification
StrongAquatic taxa are identified specifically, making this the earliest definitive evidence of turtles, crocodiles and fish in the hominin diet.
Dating
StrongThe site is well dated to 1.95 million years by tephrostratigraphy and magnetostratigraphy within the well-studied Koobi Fora Formation.
Historical interpretation
InterpretiveThe link between aquatic nutrients and hominin brain expansion is a hypothesis. The nutrient content of these foods is established; their causal role in encephalisation is not demonstrated by this site.
Sources
- 1.David R. Braun, John W. K. Harris, Naomi E. Levin, et al. (2010). Early hominin diet included diverse terrestrial and aquatic animals 1.95 Ma in East Turkana, Kenya. Proceedings of the National Academy of Sciences 107(22), 10002-10007. doi:10.1073/pnas.1002181107Scientific literature
- 2.Briana Pobiner (2013). Evidence for Meat-Eating by Early Humans. Nature Education Knowledge. www.nature.com/scitable/knowledge/library/evidenModern synthesis
Limitations
- moderatehistorical interpretation
Aquatic foods are often linked to hominin brain expansion. The nutrient content is established but the causal role in encephalisation is a hypothesis, not something this site demonstrates, and it should not be presented as settled.
- minorarchaeological attribution
The card refers to early humans. The hominins here predate Homo erectus and no fossil at the site is associated with the tools; Paranthropus boisei was also present in the region at this date.
- minordate
The card records 1,900,000 BCE. The published date for FwJj 20 is 1.95 million years, a small discrepancy within the card's stated precision.
