Reconstruction
Spy Cave, overlooking the Orneau valley in present-day Belgium, became one of the defining sites of Neanderthal research after Marcel de Puydt and Maximin Lohest recovered two fragmentary adult skeletons there in 1886. Their authenticated association with Pleistocene deposits helped persuade nineteenth-century scholars that Neanderthals represented an ancient human population rather than diseased modern people. The discovery therefore mattered both as archaeology and as evidence for human evolution.
Dating the individuals has proved difficult. Early excavation removed material without modern spatial recording, and some bones were later treated with animal-derived conservation glue. Compound-specific radiocarbon work published in 2021 showed that several earlier measurements were as much as 10,000 years too young. The best current model places the disappearance of Neanderthals in northwestern Europe, including the reliably dated Spy material, between approximately 44,200 and 40,600 calibrated years before present—roughly 42,250–38,650 BCE. Attribution of every fragment to Spy I or Spy II, and the precise relationship between the skeletons and particular artifact layers, remain uncertain.
Food evidence comes from several independent but imperfect archives. Carbon and nitrogen isotope measurements indicate that the Spy Neanderthals obtained much of their dietary protein from terrestrial resources, especially large herbivores. Compound-specific isotope research has also allowed a possible plant-protein contribution, although its magnitude is methodologically debated. More direct evidence survives in dental calculus: starch grains and phytoliths show that the individuals placed starchy plants, including grass seeds, in their mouths. Damage to some starches is consistent with heating or cooking, demonstrating that their diet cannot defensibly be reduced to raw meat alone.
Spy Cave also contained thousands of animal bones. Horse, mammoth, woolly rhinoceros, reindeer, cave hyena and bear are prominent, but this assemblage accumulated over many millennia through human activity, carnivore scavenging and natural deposition. It is not a single Neanderthal pantry. Cut marks and other modifications on mammoth and deer remains nevertheless document human carcass processing, while the unusually high proportion of young mammoths has been interpreted as probable evidence of hunting. Stone tools provided cutting edges for skinning, dismemberment and meat removal; fire could cook food, improve starch digestibility and provide heat in a cold, unstable environment.
The Spy evidence presents adaptable foragers rather than rigid specialists. Large mammals supplied dense packages of meat and fat, while gathered plants added carbohydrates and dietary variety when season and habitat allowed. Exact recipes, meal proportions and preservation practices cannot be reconstructed. Spy's defensible connection to food today is therefore not a continuous culinary tradition, but the enduring human strategy it records: combining hunted and gathered foods, using tools to process carcasses, and applying heat to make plant foods more digestible.
Historical context
Northwestern Europe was passing through Marine Isotope Stage 3, an interval marked by rapid climatic oscillations rather than a single unbroken deep freeze. Open and patchy steppe-tundra environments supported mammoths, woolly rhinoceroses, horses and reindeer. Human populations were sparse and mobile. Early Homo sapiens were already present elsewhere in Europe more than 45,000 years ago, including at Ranis in Germany, while the last documented Neanderthal populations were approaching their disappearance. The authorship of some transitional stone-tool traditions once associated with Spy remains contested, and recent genetic work identifies the Lincombian–Ranisian–Jerzmanowician at Ranis with Homo sapiens.
Evidence
Archaeological
StrongFragmentary remains representing two classic adult Neanderthal individuals were authenticated after their 1886 recovery. Stone artifacts, combustion features and faunal remains are documented at the site, although nineteenth-century excavation limits precise contextual association.
Archaeobotanical
ModerateStarch grains and phytoliths recovered from Spy dental calculus directly show that plant material entered the individuals' mouths. Identification is generally limited to broad plant groups rather than exact species or dishes.
Zooarchaeological
ModerateThe large faunal collection includes cut-marked or otherwise human-modified mammoth and deer remains. Because the assemblage accumulated over a long period through human, carnivore and natural agencies, not every animal can be associated with the Neanderthals.
Residue chemical
StrongDental-calculus microremains and bone-collagen isotope analyses independently support consumption of plants and terrestrial animal protein. Quantifying the plant contribution from compound-specific nitrogen isotopes remains debated.
Food identification
ModerateLarge-herbivore protein and starchy plant consumption are supported, but the animal species contributing to each individual's diet and the precise taxonomic identities of most plant foods cannot be determined securely.
Dating
ModerateAdvanced compound-specific radiocarbon dating indicates that older bulk-collagen measurements were contaminated and too young. The revised regional disappearance model is robust, but assignment of every dated fragment to Spy I or Spy II and reconstruction of their original stratigraphic positions remain uncertain.
Preparation method
ProbableAlteration of starch grains is consistent with cooking, and cut marks document carcass processing. Exact heating methods, recipes, food combinations and preservation techniques are reconstructions rather than direct observations.
Geographic attribution
StrongThe remains and associated collections derive from Spy Cave at Jemeppe-sur-Sambre in Namur Province, Belgium.
Historical interpretation
StrongContemporary documentation and subsequent histories agree that the 1886 discovery played an important role in scientific acceptance of Neanderthals as an ancient human population.
Visual reconstruction
InterpretiveAny depiction of clothing, food portions, cooking equipment, body decoration, landscape season or group activity combines general Neanderthal evidence with artistic reconstruction and should not be treated as a direct view of Spy.
Sources
- 1.Thibaut Devièse, Grégory Abrams, Mateja Hajdinjak, et al. (2021). Reevaluating the timing of Neanderthal disappearance in Northwest Europe. Proceedings of the National Academy of Sciences 118(12), e2022466118. doi:10.1073/pnas.2022466118Scientific literature
- 2.Philip Van Peer (2021). The stratigraphic context of Spy Cave and the timing of Neanderthal disappearance in Northwest Europe. Proceedings of the National Academy of Sciences 118(26), e2106335118. doi:10.1073/pnas.2106335118Scientific literature
- 3.Amanda G. Henry, Alison S. Brooks and Dolores R. Piperno (2011). Microfossils in calculus demonstrate consumption of plants and cooked foods in Neanderthal diets (Shanidar III, Iraq; Spy I and II, Belgium). Proceedings of the National Academy of Sciences 108(2), 486–491. doi:10.1073/pnas.1016868108Scientific literature
- 4.Yuichi I. Naito, Yoshito Chikaraishi, Dorothée G. Drucker, et al. (2016). Ecological niche of Neanderthals from Spy Cave revealed by nitrogen isotopes of individual amino acids in collagen. Journal of Human Evolution 93, 82–90. doi:10.1016/j.jhevol.2016.01.009Scientific literature
- 5.Mietje Germonpré, Mircea Udrescu and Evelyne Fiers (2013). The fossil mammals of Spy. Anthropologica et Præhistorica 123/2012, 298–327. biblio.naturalsciences.be/library-1/rbins-staff-Scientific literature
- 6.Arev P. Sümer et al. (2025). Earliest modern human genomes constrain timing of Neanderthal admixture. Nature 638, 711–717. doi:10.1038/s41586-024-08420-xScientific literature
- 7.Royal Belgian Institute of Natural Sciences (n.d.). De Puydt & Lohest (1885–1886). Anthropology and Prehistory Collections: Spy Cave. collections.naturalsciences.be/ssh-anthropology/Modern synthesis
Limitations
- moderatedate start/date end
The supplied 44,000–36,000 BCE range reflects older dating interpretations. Compound-specific radiocarbon research found that several earlier Spy measurements were up to 10,000 years too young because of contamination and modeled Neanderthal disappearance in northwestern Europe at 44,200–40,600 cal BP, approximately 42,250–38,650 BCE. The supplied end date is therefore too recent for the best-supported current chronology.
- moderatedescription
The faunal collection cannot be treated as one contemporaneous Neanderthal hunting landscape or menu. Roughly 11,600 studied bones accumulated over many millennia through human activity, hyena scavenging, bear denning and natural processes, while early excavation weakened associations between the two skeletons, artifacts and particular animal remains.
