Reconstruction
Tabun Cave does not preserve a single meal or even the remains of one population. Its roughly 25-meter sedimentary sequence records changing use of a Mount Carmel cave across hundreds of thousands of years. Dorothy Garrod’s excavations from 1929 to 1934 established the principal layers, while later campaigns refined their stratigraphy. Stone tools document transitions from Acheulean and Acheulo-Yabrudian industries to successive Mousterian technologies. The sequence therefore became a reference point for Levantine prehistory, although its chronology remains under revision.
Human fossils add to that importance. Tabun C1, a substantially complete adult female skeleton recovered near the upper part of Layer C, is commonly classified as Neanderthal. A recent dating synthesis placed her at approximately 170,000 years ago, with a wide uncertainty of about 44,000 years, while an alternative anatomical interpretation has linked her to other Middle Pleistocene populations. The robust C2 mandible from lower Layer C is also taxonomically disputed. Tabun consequently records population complexity rather than a simple, securely demonstrated succession of named human species.
Food remains are abundant but require equally careful interpretation. Gazelle, Mesopotamian fallow deer, aurochs, red and roe deer, wild boar, equids and occasional very large animals occur in the assemblages. Cut marks, fresh-bone breakage and burning show that people butchered carcasses, removed flesh and sometimes opened bones for marrow. Layer C appears to contain substantial human food refuse, whereas many fallow deer in Layer B probably entered through the cave’s chimney as naturally trapped animals. Cut and burned specimens indicate that people sometimes processed these conveniently available carcasses, but the entire accumulation cannot be treated as a hunting record.
Bird evidence adds a smaller but unusually vivid food story. A modern taphonomic study found that rock doves dominate parts of Layer C associated with relatively intensive occupation. Some dove bones have differential burning consistent with roasting, making human collection and consumption probable. Raptors and natural cave deaths produced many other bird remains, so not every recovered bird was food.
Fire transformed the cave’s food practices. Burned-flint frequencies suggest increasingly habitual fire use in the region around 350,000–320,000 years ago. In Layer C, ash and phytoliths largely derived from wood and bark support repeated domestic fires. Meat and birds could have been roasted, although no fixed recipe or cooking installation survives. Flint flakes supplied cutting edges; joints were separated, flesh removed and marrow-bearing bones broken. Plant foods were probably available in the varied Carmel landscape, but Tabun has not yielded secure identification of particular edible plants. Its defensible modern connection is therefore not a surviving dish, but a deep record of enduring practices: carcass selection, butchery, marrow extraction and cooking with controlled fire.
Historical context
No farms, towns or food markets existed anywhere. Mobile hunter-gatherer populations occupied a Levant that formed a major ecological and geographic passage between Africa and Eurasia. Across Tabun’s immense span, glacial cycles repeatedly altered rainfall, vegetation, animal communities, sea level and the width of the nearby Mediterranean coastal plain. Homo sapiens had emerged in Africa by roughly 300,000 years ago, while Neanderthal-related populations developed across western Eurasia. At Mount Carmel, caves, wooded slopes, open country, cliffs and intermittent wetlands placed large ungulates, small game, birds, fuel wood and plant resources within a relatively compact foraging landscape.
Evidence
Archaeological
StrongApproximately 25 meters of stratified deposits contain large lithic, faunal and sedimentary assemblages spanning Lower and Middle Palaeolithic industries.
Zooarchaeological
StrongIdentified mammal and bird remains, including specimens with cut marks, fresh fractures and thermal alteration, directly demonstrate carcass processing and probable consumption in parts of the sequence.
Residue chemical
ModerateSediment mineralogy, micromorphology and phytolith analysis identify substantial wood-derived ash and repeated burning, especially in Layer C, but do not identify specific plant foods.
Food identification
ModerateGazelle and other ungulates were demonstrably butchered, and roasted rock dove is probable from assemblage composition and burning patterns. Some fauna, particularly Layer B fallow deer and many birds, accumulated naturally.
Dating
ModerateMultiple ESR, thermoluminescence, uranium-series and luminescence programs support great antiquity, but diagenesis, broad uncertainties and complicated layer correlations prevent a single exact date range for the entire record.
Preparation method
ProbableCut marks and fracture patterns support skinning, dismemberment, defleshing and marrow extraction. Differentially burned dove bones probably reflect roasting, while the precise placement of food relative to fires is reconstructed rather than observed.
Geographic attribution
StrongThe excavated cave is securely located at Nahal Me’arot on the western face of Mount Carmel, overlooking the Mediterranean coastal plain.
Historical interpretation
ModerateLayer C is reasonably interpreted as containing repeated domestic occupations. Layer B includes substantial natural-trap deposition with more intermittent human exploitation.
Visual reconstruction
InterpretiveAny depiction of people roasting meat or birds around a hearth would combine archaeological evidence with assumptions about implements, gestures, clothing, containers and food service that are not directly preserved.
Sources
- 1.Dorothy A. E. Garrod and Dorothea M. A. Bate (1937). The Stone Age of Mount Carmel, Volume I: Excavations at the Wady el-Mughara. Clarendon Press. books.google.com/books/about/The_Stone_Age_of_MoPrimary archaeology
- 2.Rosa M. Albert, Ofer Lavi, Lara Estroff, et al. (1999). Mode of Occupation of Tabun Cave, Mt Carmel, Israel During the Mousterian Period: A Study of the Sediments and Phytoliths. Journal of Archaeological Science 26(10): 1249–1260. doi:10.1006/jasc.1999.0355Scientific literature
- 3.A. B. Marín-Arroyo (2013). Palaeolithic Human Subsistence in Mount Carmel (Israel): A Taphonomic Assessment of Middle and Early Upper Palaeolithic Faunal Remains from Tabun, Skhul and el-Wad. International Journal of Osteoarchaeology 23(3): 254–273. doi:10.1002/oa.1241Scientific literature
- 4.Ron Shimelmitz, Steven L. Kuhn, Arthur J. Jelinek, et al. (2014). ‘Fire at Will’: The Emergence of Habitual Fire Use 350,000 Years Ago. Journal of Human Evolution 77: 196–203. doi:10.1016/j.jhevol.2014.07.005Scientific literature
- 5.David E. Friesem, Ruth Shahack-Gross, Mina Weinstein-Evron, et al. (2021). High-Resolution Study of Middle Palaeolithic Deposits and Formation Processes at Tabun Cave, Israel: Guano-Rich Cave Deposits and Detailed Stratigraphic Appreciation of Layer C. Quaternary Science Reviews 274: 107203. doi:10.1016/j.quascirev.2021.107203Scientific literature
- 6.M. Richard, N. Mercier, M. Weinstein-Evron, L. Weissbrod and R. Shimelmitz (2024). Chronology of the Late Lower and Middle Palaeolithic at Tabun Cave (Mount Carmel, Israel) with Insights into Diagenesis and Dose Rate Variation Using Post-IR IRSL Dating and Infrared Spectroscopy. Quaternary Geochronology 84: 101611. doi:10.1016/j.quageo.2024.101611Scientific literature
- 7.Linda Amos, Reuven Yeshurun, Mina Weinstein-Evron and Ron Shimelmitz (2025). Birds from the Oven: The Middle Palaeolithic Avifauna of Tabun Cave, Mount Carmel, Israel. Archaeological and Anthropological Sciences 17: 90. doi:10.1007/s12520-025-02197-4Scientific literature
Limitations
- moderatedate start
The supplied start date of approximately 350,000 BCE truncates Tabun’s earlier record. ESR and uranium-series analysis placed the base of Layer Ed at about 390,000 ± 50,000 years ago, and modern syntheses describe the full sequence as extending roughly 500,000 years. The precise age of the deepest deposits remains debated.
- minordate end
The supplied endpoint of approximately 45,000 BCE is a conventional estimate for the close of the late Mousterian sequence rather than a securely dated final occupation. Direct dates for upper Layer B are generally older and carry broad uncertainty, while Layer A includes later mixed material.
