Reconstruction
Brain tissue is metabolically expensive: per gram it burns energy roughly sixteen times faster than skeletal muscle at rest. A human brain is about two per cent of body weight and takes something near a fifth of resting energy expenditure. The awkward part is that human resting metabolism is unremarkable for a primate of our body mass, so the surcharge never appears in the total. Something else must have been cut back to pay for it, and the obvious candidate is the other expensive tissue: the gut.
That is the expensive-tissue hypothesis, set out by Leslie Aiello and Peter Wheeler in 1995. In humans the digestive tract is substantially smaller than body mass predicts, and the energy saved is close to the brain's excess cost. A short gut, though, cannot extract a living from bulk fibrous plants. It needs food that is dense, soft and already partly broken down, which is what processing delivers.
The experimental case for that half is good. Feeding trials show cooked diets yield more net energy than raw diets of the same composition, with heating and mechanical processing each contributing. Slicing meat and pounding tubers, with no fire involved, cut the number of chews needed by roughly a sixth and the force per chew by about a quarter — enough on its own to account for the reduced human jaw and teeth. Living humans spend under five per cent of the day feeding, where a primate of our size should spend closer to half.
The trade-off itself, however, did not survive testing. Compared across roughly a hundred mammal species, gut and brain masses show no negative relationship; within primates the two scale together. The tissue that does trade against brain size is fat. And when total daily expenditure was measured directly in humans and all four great apes, humans came out highest — the budget was enlarged, not reallocated. The reframing that followed treats brain size as limited jointly by energy intake and by competing demands, with growth and reproduction as much in play as digestion.
Whichever mechanism is right, the dietary consequence is identical. A short gut and a raised throughput both demand more calories per mouthful and less work to release them. Modelling of raw-food feeding time suggests a body and a neuron count like ours are simply unaffordable without processing.
Sangiran supplies the bodies rather than the meals. The Central Java dome has produced more Homo erectus fossils than anywhere on earth, and zircon dating now places the earliest near 1.3 million years rather than the 1.66 million once claimed. It holds no hearth, no ash and no cut-marked bone. Its one line of direct dietary evidence, isotopes banded through a single molar, records seasonal switching between grass-based and mixed foods: flexibility, not energy density. The expensive body is a physiological argument that the archaeology of this date cannot yet corroborate.
Historical context
Brain size in the genus Homo was climbing but not racing: early Homo erectus crania fall between roughly 600 and 950 cubic centimetres, well above the australopiths and well below the modern human average near 1,350. The lineage had already spread from the Caucasus to Java, and the Acheulean handaxe was several hundred thousand years old in East Africa. Sunda was periodically dry land — falling sea levels joined Java to Sumatra, Borneo and the mainland, letting large mammals and hominins walk east across grassland, woodland and gallery forest. Crucially, nothing in that world had fire in any demonstrated sense. The oldest microstratigraphic evidence for in-situ burning, at Wonderwerk Cave in South Africa, is about a million years old; the earliest strong case for repeated hearth use, at Gesher Benot Ya‘aqov in the Jordan Valley, is roughly 790,000 years old; and fire becomes archaeologically routine only after about 400,000 years ago. If cooking paid for the expensive body, it did so long before the archaeology can see it.
Evidence
Historical interpretation
InterpretiveThe expensive-tissue hypothesis (Aiello & Wheeler 1995) is a physiological argument, not a finding from any site. Its central prediction, a negative brain–gut relationship, was not recovered across ~100 mammal species and is positive within primates (Navarrete et al. 2011); the negative correlation is with adipose tissue. Pontzer et al. (2016) show human total energy expenditure exceeds that of all four great apes, indicating an enlarged rather than reallocated budget. Isler & van Schaik's expensive-brain framework replaces the single trade-off with joint limits on intake and allocation.
Preparation method
ModerateControlled feeding trials establish that cooking raises net energy gain over raw diets of identical composition, and that thermal and non-thermal processing contribute separately (Carmody & Wrangham 2009; Carmody et al. 2011). Zink & Lieberman (2016) measured ~17% fewer chews and ~26% less force per chew from slicing meat and pounding tubers alone. The energetics of processing are well demonstrated; the antiquity of cooking is not.
Dating
StrongFission-track and U–Pb dating of zircon grains from tuffs above, below and within the hominin-bearing Sangiran and Grenzbank beds gives a first appearance datum of about 1.3 Ma, with a maximum near 1.45 Ma (Matsu'ura et al. 2020) — younger than the card's 1.5 Ma start.
Food identification
LimitedThe only direct dietary data for Javanese Homo erectus are laser-ablation δ13C and Sr/Ca series through the enamel of very few teeth (Kubat et al. 2023). One individual documents a C4-to-mixed shift within childhood. Sample size forbids generalising to the population, and nothing in it measures energy density.
Zooarchaeological
Insufficient evidenceNo butchery assemblage, cut-marked bone or percussion-broken long bone at Sangiran has been shown to result from hominin activity; the vertebrate fauna is a fluvially reworked accumulation. There is no local evidence for the meat-rich, high-return diet the energetic argument requires.
Archaeobotanical
Insufficient evidenceNo preserved plant foods, starch residues or phytolith assemblages are associated with these hominins, so the fibrous-plant baseline the hypothesis argues away from is itself unobserved here.
Geographic attribution
StrongProvenance is secure: the Sangiran Dome sequence is mapped bed by bed and the hominin specimens are tied to named lithostratigraphic units (Sangiran Formation, Grenzbank, Bapang Formation).
Visual reconstruction
LimitedDepicting a hearth, cooking, or roasted food at this date and place would illustrate the hypothesis rather than the record: no fire evidence exists at Sangiran, and the oldest in-situ burning anywhere is ~1.0 Ma.
Sources
- 1.Aiello, L.C.; Wheeler, P. (1995). The Expensive-Tissue Hypothesis: The Brain and the Digestive System in Human and Primate Evolution. Current Anthropology 36(2): 199–221. doi:10.1086/204350Scientific literature
- 2.Navarrete, A.; van Schaik, C.P.; Isler, K. (2011). Energetics and the evolution of human brain size. Nature 480(7375): 91–93. doi:10.1038/nature10629Scientific literature
- 3.Pontzer, H.; Brown, M.H.; Raichlen, et al. (2016). Metabolic acceleration and the evolution of human brain size and life history. Nature 533(7603): 390–392. doi:10.1038/nature17654Scientific literature
- 4.Isler, K.; van Schaik, C.P. (2009). The Expensive Brain: A framework for explaining evolutionary changes in brain size. Journal of Human Evolution 57(4): 392–400. doi:10.1016/j.jhevol.2009.04.009Scientific literature
- 5.Carmody, R.N.; Wrangham, R.W. (2009). The energetic significance of cooking. Journal of Human Evolution 57(4): 379–391. doi:10.1016/j.jhevol.2009.02.011Scientific literature
- 6.Carmody, R.N.; Weintraub, G.S.; Wrangham, R.W. (2011). Energetic consequences of thermal and nonthermal food processing. Proceedings of the National Academy of Sciences 108(48): 19199–19203. doi:10.1073/pnas.1112128108Scientific literature
- 7.Zink, K.D.; Lieberman, D.E. (2016). Impact of meat and Lower Palaeolithic food processing techniques on chewing in humans. Nature 531(7595): 500–503. doi:10.1038/nature16990Scientific literature
- 8.Organ, C.; Nunn, C.L.; Machanda, et al. (2011). Phylogenetic rate shifts in feeding time during the evolution of Homo. Proceedings of the National Academy of Sciences 108(35): 14555–14559. doi:10.1073/pnas.1107806108Scientific literature
- 9.Fonseca-Azevedo, K.; Herculano-Houzel, S. (2012). Metabolic constraint imposes tradeoff between body size and number of brain neurons in human evolution. Proceedings of the National Academy of Sciences 109(45): 18571–18576. doi:10.1073/pnas.1206390109Scientific literature
- 10.Antón, S.C.; Potts, R.; Aiello, L.C. (2014). Evolution of early Homo: An integrated biological perspective. Science 345(6192): 1236828. doi:10.1126/science.1236828Scientific literature
- 11.Matsu'ura, S.; Kondo, M.; Danhara, et al. (2020). Age control of the first appearance datum for Javanese Homo erectus in the Sangiran area. Science 367(6474): 210–214. doi:10.1126/science.aau8556Scientific literature
- 12.Kubat, J.; Nava, A.; Bondioli, et al. (2023). Dietary strategies of Pleistocene Pongo sp. and Homo erectus on Java (Indonesia). Nature Ecology & Evolution 7(2): 279–289. doi:10.1038/s41559-022-01947-0Scientific literature
- 13.Grabowski, M.; Hatala, K.G.; Jungers, et al. (2015). Body mass estimates of hominin fossils and the evolution of human body size. Journal of Human Evolution 85: 75–93. doi:10.1016/j.jhevol.2015.05.005Scientific literature
- 14.Will, M.; Stock, J.T. (2015). Spatial and temporal variation of body size among early Homo. Journal of Human Evolution 82: 15–33. doi:10.1016/j.jhevol.2015.02.009Scientific literature
- 15.Berna, F.; Goldberg, P.; Horwitz, et al. (2012). Microstratigraphic evidence of in situ fire in the Acheulean strata of Wonderwerk Cave, Northern Cape province, South Africa. Proceedings of the National Academy of Sciences 109(20): E1215–E1220. doi:10.1073/pnas.1117620109Scientific literature
- 16.Goren-Inbar, N.; Alperson, N.; Kislev, et al. (2004). Evidence of Hominin Control of Fire at Gesher Benot Ya‘aqov, Israel. Science 304(5671): 725–727. doi:10.1126/science.1095443Scientific literature
- 17.Roebroeks, W.; Villa, P. (2011). On the earliest evidence for habitual use of fire in Europe. Proceedings of the National Academy of Sciences 108(13): 5209–5214. doi:10.1073/pnas.1018116108Scientific literature
- 18.Hilgen, S.L.; Hilgen, F.J.; Adhityatama, et al. (2022). Towards an astronomical age model for the Lower to Middle Pleistocene hominin-bearing succession of the Sangiran Dome area on Java, Indonesia. Quaternary Science Reviews 297: 107788. doi:10.1016/j.quascirev.2022.107788Scientific literature
Limitations
- majordate start is superseded by current dating
The card begins at 1,500,000 BCE. Matsu'ura et al. (2020, Science, doi:10.1126/science.aau8556) dated zircons from tuffs bracketing the hominin-bearing Sangiran and Grenzbank beds by combined fission-track and U–Pb methods and placed the first appearance of Javanese Homo erectus at about 1.3 Ma, with an absolute maximum of roughly 1.45 Ma — explicitly at least 200,000 years younger than the 1.66–1.8 Ma ages published in the 1990s. A start of about 1,300,000 BCE matches the literature.
- majorthe title asserts a mechanism the comparative data do not support
'The Expensive Body' names the brain-for-gut trade of Aiello & Wheeler (1995, doi:10.1086/204350). Navarrete et al. (2011, doi:10.1038/nature10629) tested it against roughly 100 mammal species and found no negative brain–gut correlation, with positive scaling inside primates; the tissue that trades against brain size is fat. Pontzer et al. (2016, doi:10.1038/nature17654) measured total daily expenditure in humans and all four great apes and found the human budget enlarged, not reallocated. The hypothesis remains a live framework — see the expensive-brain reformulation, doi:10.1016/j.jhevol.2009.04.009 — but its headline trade-off should not be narrated as established fact.
- moderatecooking as the cause is a hypothesis whose archaeological timing does not fit
The energetic gain from cooking and from mechanical processing is well demonstrated experimentally (doi:10.1016/j.jhevol.2009.02.011; doi:10.1073/pnas.1112128108; doi:10.1038/nature16990). What is not demonstrated is that hominins were cooking at this date. The oldest in-situ burning anywhere is Wonderwerk Cave at about 1.0 Ma (doi:10.1073/pnas.1117620109), the earliest strong case for repeated hearth use is Gesher Benot Ya‘aqov at roughly 790 ka (doi:10.1126/science.1095443), and fire is only routine after about 400–300 ka (doi:10.1073/pnas.1018116108). Non-thermal processing needs no fire and is the more defensible mechanism for the early part of this range.
- moderateno fire, hearth or butchery evidence exists at this site, yet it sits in an epoch called The Fire Revolution
Sangiran has produced no ash lens, no hearth, no burnt bone and no hominin-attributed cut marks. Any visual reconstruction placing a fire, a hearth or cooked food in a Sangiran scene invents its central element, and any text implying that cooking is evidenced here overstates the record by a wide margin.
- moderate'larger bodies' is not securely established for Javanese Homo erectus
The 'expensive body' premise needs a body that actually got bigger. Grabowski et al. (2015, doi:10.1016/j.jhevol.2015.05.005) revised hominin body-mass estimates downward after replacing large-bodied modern reference samples, and Will & Stock (2015, doi:10.1016/j.jhevol.2015.02.009) found body size in early Homo spatially and temporally variable rather than a single step increase. The Sangiran sample is overwhelmingly cranial and dental, so local body mass is poorly constrained in any case.
- minordate end stretches beyond the securely dated hominin beds
The youngest well-dated Sangiran hominin horizons lie in the Bapang Formation around 0.8 Ma. Running the card to 700,000 BCE extends past the dated fossil material; see the Hilgen et al. (2022) astronomical age model, doi:10.1016/j.quascirev.2022.107788.
- minorthe framing omits the best-evidenced dietary fact about these hominins
The strongest direct dietary evidence for Javanese Homo erectus is the intra-tooth isotope and trace-element record of Kubat et al. (2023, doi:10.1038/s41559-022-01947-0): a C4-to-mixed δ13C shift within a single individual and damped Sr/Ca seasonality relative to Pongo, i.e. seasonal dietary flexibility. That is a different claim from energy density, and it is what the enamel actually shows.
