GC-MS of braziers from 2,500-year-old graves revealed high-THC cannabis residues — the oldest proof of psychoactive cannabis use.
The Backstory
High in the Pamir Mountains of western China, at an elevation of over 3,000 meters where the air is thin and the ground freezes solid, archaeologists uncovered a cemetery unlike anything they had seen before. Stone circles marked the graves. Inside them, alongside the dead, sat wooden braziers — bowls with small heated stones inside, designed to burn something and release its smoke into the frigid air.
When Meng Ren and an international team of scientists analyzed the residue inside those braziers using gas chromatography-mass spectrometry, they found cannabinoids. Not just any cannabinoids — the chemical signatures showed THC levels far above what wild cannabis plants produce. These people, 2,500 years ago, weren't just burning hemp. They were deliberately using cannabis for its psychoactive properties.
This was the earliest clear chemical evidence of humans getting high.
The Jirzankal Cemetery
The cemetery sits in Tashkurgan Tajik Autonomous County, Xinjiang, in the far west of modern China — a place that 2,500 years ago was a crossroads of civilizations. The Pamir Mountains sit along what would become the Silk Road, connecting Central Asian, East Asian, and Western Asian cultures.
Process
What Was Found at the Jirzankal Cemetery
Surface graves with stone patterns
The cemetery contained graves marked by distinctive stone arrangements on the surface. Radiocarbon dating placed them at approximately 500 BCE — the first millennium before the common era.
Wooden braziers in burial chambers
Inside multiple graves, archaeologists found wooden bowls containing small stones that had been exposed to intense heat. These were identified as braziers — vessels for burning plant material and releasing smoke.
Silk Road cultural connections
The cemetery also yielded glass beads typical of Western Asia, angular harps with cultural ties to the broader region, and silk from eastern China. This was a population connected to extensive trade networks.
Mixed population
Skeletal analysis revealed both local and immigrant individuals. Only 1 of the 10 immigrant burials contained a brazier, suggesting cannabis burning was a local practice, not something visitors brought with them.
Ren et al. (2019), Science Advances 5(6):eaaw1391
The Chemistry
How They Did It
How GC-MS Confirmed Ancient Cannabis Smoking
Sample collection
Researchers extracted organic residue from the interior and exterior surfaces of 10 wooden brazier fragments and 4 burnt stones recovered from burial contexts.
Sampling both interior (where material was burned) and exterior (control) surfaces was critical for ruling out environmental contamination
Gas chromatography-mass spectrometry
GC-MS breaks organic material into its molecular components, allowing identification of specific chemical compounds. The technique can detect cannabinoids even in 2,500-year-old residue.
This is the same analytical method used by forensic laboratories and drug testing facilities — applied to archaeological samples
Cannabinoid detection
CBN (cannabinol) was detected on the interior surfaces of all but one brazier and on two of the heated stones. CBD and CBL (cannabicyclol) were also identified. THC degrades to CBN over time, so high CBN levels indicate the original material was high in THC.
CBN is the degradation product of THC — finding CBN in ancient samples is the chemical fingerprint of THC that has decomposed over centuries
Comparison to wild cannabis
The cannabinoid levels were significantly higher than what wild cannabis plants produce. Wild Cannabis sativa typically has very low THC — often below 1%. The Jirzankal residues showed concentrations far above this baseline.
This was the key finding: not just cannabis, but high-THC cannabis, indicating either selective cultivation or deliberate selection of naturally potent plants
Ren et al. (2019)
~500 BCE
the approximate date of the Jirzankal cannabis residues, making this the earliest direct chemical evidence of cannabis being burned for its psychoactive properties. Previous evidence of psychoactive cannabis use was roughly 500 years more recent.
For context, 500 BCE is the era of Confucius in China, the early Roman Republic, and the construction of the Parthenon in Athens. Cannabis was being used to alter consciousness during funeral rites while the foundations of Western and Eastern philosophy were being laid.
Ren et al. (2019)
The Herodotus Connection
The discovery at Jirzankal resonated with one of the most famous passages in ancient literature. Writing around 440 BCE — roughly contemporaneous with the Jirzankal burials — the Greek historian Herodotus described the funeral customs of the Scythians, a nomadic people of Central Asia:
“The Scythians take the seed of this hemp and, creeping under the felt coverings, throw it upon the red-hot stones; immediately it smokes, and gives out such a vapour as no Grecian steam-bath can exceed; the Scythians, transported by the vapour, shout aloud.”
— Herodotus
The Histories, Book IV, circa 440 BCE — describing Scythian cannabis rituals remarkably similar to the archaeological evidence found at Jirzankal
Archaeological excavations of Scythian-associated sites in the Altai Mountains of southern Russia (the Pazyryk culture) had already found cannabis seeds in burial contexts. But Ren's team provided the first direct chemical evidence confirming the practice in a contemporaneous Central Asian population along the Silk Road corridor.
Wild Cannabis vs. Cultivated Cannabis
One of the most important implications of the study is what it reveals about early cannabis cultivation.
Wild Cannabis sativa is primarily a fiber and seed plant. Its THC content is typically very low — well below the threshold for noticeable psychoactive effects. For the Jirzankal braziers to contain high-THC residues, one of two things had to happen: either these people were actively cultivating cannabis and selecting for potency, or they were identifying and harvesting wild populations that naturally produced higher THC.
Deliberate cultivation for potency
The elevated THC levels suggest intentional selection pressure — choosing plants with stronger effects and propagating them over generations, just as early farmers selected for larger grain or sweeter fruit.
The consistent presence of high-THC residues across multiple braziers in the cemetery suggests a reliable, not accidental, supply of potent cannabis.
Selection from naturally variable wild populations
Cannabis plants at higher altitudes, under greater UV stress, naturally produce more cannabinoids as a defense mechanism. The Pamir population may have simply recognized that mountain cannabis was stronger and preferentially harvested it.
The Ren team noted that altitude and UV exposure can increase cannabinoid production in wild cannabis. The Jirzankal Cemetery sits above 3,000 meters.
Both processes simultaneously
The two explanations aren't mutually exclusive. People may have noticed that high-altitude cannabis was more potent, then begun deliberately cultivating the strongest plants — combining environmental effect with artificial selection.
This mirrors how most crop domestication actually worked: humans noticed natural variation, then amplified it through selective cultivation.
Ren et al. (2019)
Two Discoveries, One Story
Ren's Jirzankal finding pairs with a slightly older discovery: Ethan Russo's analysis of a 2,700-year-old cannabis cache from the Yanghai Tombs in Xinjiang, China, found approximately 789 grams of dried cannabis in a shaman's grave. Together, these discoveries paint a picture of a Central Asian corridor — from the Turpan Depression to the Pamir Plateau — where cannabis was cultivated, traded, and used ritually for millennia before the common era.
The broader implication is that cannabis is not a modern drug. It is one of humanity's oldest psychoactive companions — woven into funeral rites, shamanic practices, and trade networks since before recorded history. Every debate about cannabis policy is, in a sense, the latest chapter in a story that began at least 2,500 years ago on a frozen mountaintop along the Silk Road.
Key Takeaways
The origins of cannabis smoking: Chemical residue evidence from the first millennium BCE in the Pamirs
Ren M, Tang Z, Wu X, Spengler R, Jiang H, Yang Y, Boivin N (2019) · Science Advances
Related Research
Key studies in this area
The 2,700-year-old cannabis stash
Russo et al. (2008)
The even older discovery: 789g of dried cannabis in a shaman's tomb in the Yanghai Tombs of Xinjiang — the companion archaeological find that, together with Jirzankal, establishes Central Asia as the cradle of psychoactive cannabis culture
History of cannabis and the endocannabinoid system
Crocq (2020)
A comprehensive historical review placing these archaeological discoveries in the broader timeline of human cannabis use
The isolation of THC
Mechoulam & Gaoni (1964)
The modern chemical identification of the same compound these ancient people were producing — THC was only isolated in a laboratory 2,500 years after people were deliberately burning it in funeral rites
Cannabis prohibition origins
Musto (1999)
The historical irony: a plant used ritually for millennia was criminalized in the 20th century based on racial politics, not pharmacology
Is this the oldest cannabis ever found?
No — Russo's 2008 analysis of cannabis from the Yanghai Tombs dates to approximately 700 BCE, roughly 200 years older. But the Yanghai cannabis was a dried cache (stored plant material), while the Jirzankal evidence is residue from cannabis that was actually burned. Ren's finding is the oldest evidence of cannabis smoking specifically — the deliberate combustion of cannabis to inhale its psychoactive smoke.
How do we know they were getting high and not just burning hemp?
The chemical analysis is the key. Wild hemp has very low THC content — typically below 1%. The residues from the Jirzankal braziers showed cannabinoid profiles consistent with high-THC cannabis, particularly elevated CBN (the compound THC degrades into over centuries). If they were burning ordinary hemp fiber, the cannabinoid signature would be different.
Why was cannabis used in funeral ceremonies?
The study doesn't definitively answer this, but the ritual context — braziers placed alongside the dead in grave chambers — suggests the cannabis smoke served a spiritual or ceremonial function. Herodotus described Scythian cannabis rituals as communal experiences of altered consciousness. The Jirzankal evidence aligns with cross-cultural patterns of using psychoactive substances to mark death, facilitate communication with the deceased, or ease grief.
Does altitude really make cannabis more potent?
Yes. Cannabis plants under greater UV stress (found at higher altitudes) tend to produce more cannabinoids, likely as a biochemical defense mechanism. The Jirzankal Cemetery sits above 3,000 meters. Whether the elevated THC in the brazier residues reflects altitude effects, deliberate cultivation, or both remains debated, but the altitude-potency connection is well-documented in modern cannabis botany.
What the researchers found
GC-MS analysis of 2,500-year-old funeral braziers revealed high-THC cannabis residues — the earliest chemical evidence of cannabis smoking for psychoactive purposes.
Why it matters
Establishes that deliberate psychoactive cannabis use dates back at least 2,500 years along the ancient Silk Road.
The numbers in context
10 brazier fragments and 4 stones analyzed. CBN detected in 9/10 brazier interiors. Cemetery at 3,060-3,080m. Date: ~500 BCE.
How the study worked
GC-MS of organic residues from archaeological wooden braziers and heated stones from burial contexts at Jirzankal Cemetery, Pamir Mountains, western China.
Who was studied
Archaeological site: Jirzankal Cemetery (~500 BCE), Pamir Mountains, western China.
What this study cannot tell us
Cannot determine exact original THC concentration. Cannot distinguish deliberate cultivation from harvesting naturally potent high-altitude plants.
Read the original research
The origins of cannabis smoking: Chemical residue evidence from the first millennium BCE in the Pamirs.
Science Advances, 5(6), eaaw1391
Open-access multidisciplinary journal from AAAS.
Citation
Ren, Meng; Tang, Zihua; Wu, Xinhua; Spengler, Robert; Jiang, Hongen; Yang, Yimin; Boivin, Nicole. (2019). The origins of cannabis smoking: Chemical residue evidence from the first millennium BCE in the Pamirs.. Science Advances, 5(6), eaaw1391. https://doi.org/10.1126/sciadv.aaw1391