Siberian tomb offers oldest direct evidence of psychoactive cannabis use, study finds

A tomb built nearly 2,800 years ago for a high-ranking member of early Scythian society in southern Siberia has produced what an international team of archaeologists describes as the oldest directly dated physical evidence of cannabis being used for its psychoactive effects. The claim rests on two overlapping lines of evidence recovered from the same burial chamber: roughly 1,000 cannabis seeds and strands of preserved human hair carrying chemical signatures consistent with the plant's active compounds.

The findings, published in the journal Science, come from the Tunnug 1 burial mound in Tyva, a republic in southern Siberia, a site associated with early horse-riding Scythian elites. The paper, titled "Direct evidence for psychoactive cannabis use 2800 years ago," is led by Gino Caspari of the University of Bern's Institute of Archaeological Sciences and the Max Planck Institute of Geoanthropology, with collaborators in Germany, Russia, Sweden and Switzerland.

A princely tomb and its seeds

Of the roughly 1,000 cannabis seeds recovered from the tomb, six were directly radiocarbon dated. Combined with other chronological evidence from the site, the dating places their deposition in the early eighth century B.C. The researchers argue the seeds were placed deliberately rather than arriving by accident, noting their concentration and position within the burial.

The hair evidence, drawn from the tomb's central burial chamber, was analyzed using high-resolution mass spectrometry and other molecular techniques. Several samples carried clear signals consistent with cannabinol, or CBN, a compound that forms when THC — the primary psychoactive ingredient in cannabis — breaks down over time. Weaker signals consistent with THC itself were also detected. Because the hair and the seeds were recovered from the same grave, the authors say the combination points to intoxicant use rather than purely ceremonial deposition of the plant.

Tunnug 1 was first identified in 2017, when Caspari spotted a circular structure on satellite imagery of the Uyuk Valley and led a trial excavation with Russia's Hermitage Museum and the Russian Academy of Sciences; the Swiss National Science Foundation reported at the time that wooden beams from the mound dated to the ninth century B.C., making it among the earliest known Scythian princely tombs in the region. Later excavation seasons turned up dozens of sacrificed horses and scattered human remains, evidence the researchers describe as consistent with elaborate funerary ritual, including what they call likely human sacrifice.

How the case was built

The study's strength, according to Caspari, lies in combining two kinds of evidence that are rarely found together at an archaeological site of this age: botanical remains and human tissue carrying the chemical traces of what a person actually consumed.

"It's extremely rare that we have both preserved plant parts and human soft tissue with the active compounds in it in the same context," Caspari said, according to reporting on the study.

Before this find, the earliest widely cited direct chemical evidence for cannabis use as an intoxicant came from the Jirzankal Cemetery in the Pamir Mountains of western China, where researchers reported in 2019 that wooden braziers from roughly 2,500 years ago carried THC residue consistent with ritual burning, with THC levels higher than typically found in wild plants. The new Siberian findings, if accepted, would push the dated record of psychoactive cannabis use back roughly 300 years earlier and shift its geography west, onto the Eurasian steppe associated with early Scythian culture. Cannabis has long been tied to the Scythians through the writings of the Greek historian Herodotus, who described steppe peoples inhaling vapors from heated cannabis seeds, but that account had not previously been matched to dated physical and chemical evidence from a burial.

A contested piece of evidence

Independent specialists are divided on how much weight the hair analysis can bear, and the disagreement centers on a genuinely open methodological question rather than the broader plausibility of ancient cannabis use. Volker Auwaerter, head of forensic toxicology at the University Medical Center in Freiburg, Germany, who was not involved in the research, said the hair testing underlying the study's central claim "is scientifically weak." He argued that the analysis could not distinguish actual consumption from external contamination, since hair buried for millennia alongside cannabis plant material could absorb chemical traces without the compounds ever having been ingested or inhaled, and said a dedicated forensic toxicology protocol would have strengthened the result.

Elisa Guerra Doce, an archaeologist at the University of Valladolid in Spain who also was not part of the study, took a more supportive view, noting that toxicological hair analysis is an established method in archaeology for identifying past use of psychoactive substances and that the paired seed-and-hair evidence makes a reasonable case that cannabis was consumed by people connected to the burial rite rather than simply offered as a grave good.

The authors themselves are explicit that several questions remain unresolved. The paper does not establish how the cannabis entered the body — whether through inhalation, as later Central Asian sites with heated stones and cannabinoid residue on burner vessels suggest, or through ingestion. It also cannot determine dosage, how much plant material was involved, or whether the cannabis came from cultivated plants selected for higher psychoactive potency or simply from wild stands growing near the site.

What comes next

Caspari has said that genetic sequencing of some of the recovered plant material is already underway, work that could clarify whether the Tunnug 1 cannabis was domesticated or wild, and potentially trace its genetic relationship to other ancient cannabis populations across Eurasia. The team also says preservation conditions at the site, including excellent DNA survival, leave open the possibility of further biomolecular work on both the plant remains and the human remains recovered from the mound.

For now, the finding stands as a well-documented but not universally accepted data point: a physically and chemically consistent signal of cannabis intoxication from a specific, dated burial, bundled with caveats its own authors acknowledge and with at least one independent toxicologist publicly questioning whether the hair evidence meets the standard the paper's headline claim requires. Coverage of the dispute reflects a broader pattern in archaeological chemistry, where ancient biomolecular claims often outrun the field's consensus on what counts as sufficient proof.