Up in smoke - the early history of Cannabis
An introduction to the world of plants that were smoked throughout history
While it’s in the decline today, at least in the Western world, tobacco smoking is still an exceptionally widespread and practiced activity. Approximately 1.3 billion people, a fifth of the global population, are tobacco smokers today.1 The number of people who, instead, smoke other plant-derived substances, most notably “weed” is however constantly increasing, as a 2022 poll found that for the first time, more Americans smoked weed than tobacco.2 Despite the stunning success of modern “smoking” as an accessible means of consuming recreational drugs, it was only introduced to the broader world from the Americas (where it originates) in the late 16th century. Did people in the “old world” smoke other products such as marijuana before that time? Would their usage be recogniseable to us today?
The basics
Two obvious questions to elucidate before we begin with the history are: “how does smoking work”, and “why are some substances enjoyable to smoke but not others?” As it happens, the answer to the first question will also help to answer the second.
The first question, put more clearly, would read “How does smoke deliver biologically active chemicals into the bloodstream?” Since the topic of cigarettes is the most well-researched, we will focus on tobacco and its main active compound, nicotine. Other types of plant matter have other active compounds whose properties may differ. For example the main active compound of Cannabis, THC, will be discussed later on.
Contrary to what one might expect, the smoke produced by burning plant matter isn’t just a gas, but rather an aerosol mixture of gaseous volatile compounds and a non-volatile particulate phase of droplets between 0.1 and 1µm in diameter. Approximately 10^9 particles are present per milliliter of smoke, and the average cigarette produces 400-500 milliliters of smoke in total. The volatile portion is mostly composed of the chemicals which are found in air, but also organic compounds such as benzene, toluene, acetic acid, acetonitrile… Meanwhile the particulate phase of cigarettes contains at least 3500 compounds, and many more such as pesticides or flavour additives are found in some brands of cigarettes and not others and not listed.3
Particles of non-volatile organic compounds (nicotine in this case, but applicable to THC) are inhaled deep into the alveoles of the lung, where oxygen is absorbed by the blood in order to be transported throughout the body, thanks to their small size. Nicotine is especially well-suited to smoking as in this form (under physiological conditions) approximately 23% of the nicotine is in an un-ionized form able to cross biological barriers unimpeded. The rest loses a proton to become un-ionized in the very short term. Essentially all the inhaled nicotine will eventually make it into the bloodstream. It takes as little as 7 seconds for the compound to go from the lungs to the brain, where it binds to the nicotinic cholinergic receptors and leads to a reaction cascade, bringing about a quick nicotine “hit”. Inhalation is the most efficient delivery mode of such compounds to the bloodstream.4
These molecules are produced by the plant secondary metabolism, which deals with all molecules that aren’t strictly necessary for the growth or survival of the plant, but instead participate in signalling, communication or reaction to stress. Some plant families or genuses are known for their particularly active secondary metabolism which produces volatile and very aromatic compounds (in the literal, not necessarily chemical, sense). It’s the secondary metabolism that makes plants poisonous, medicinal or psychotropic, etc.
Thus, it is likely that plants which are smoked (or were smoked historically) are those which contain bioactive compounds whose properties allow them to quickly cross from the alveolar region to the bloodstream and interact with neural receptors in the brain in a non-toxic fashion. One can guess at the trial-and-error process which led to these discoveries, which could be analogous to the processes involved in the discovery of new sources of food, or recipes and cooking processes to make food edible and safe. As the smoke from wood fires naturally has a certain fragrance, over a long enough timescale the odds of someone throwing already-pungent plants in the flames, to try and perfume the space or just by idle curiosity, seem like they’d be non-negligible.
Cannabis, an ancient human companion
As smoking is a low-tech practice, its origins lie shrouded in the mists of history. Only archaeology and the oldest written records offer us a clue as to the behaviour of the first smokers. This is because so little is involved besides soft, degradable organic matter, which is rarely preserved through vast expanses of time, so one could only hope to find dried leaves or resins in the most peculiar circumstances. Tough and inert smoking paraphernalia such as pipes, which survive the ages far better, were mainly a post-Columbian exchange technology and won’t help us here.
Cannabis sativa, or hemp, is one of the oldest cultivated plants, and thus an excellent candidate to start this series on smokable plants. This annual plant of the family Cannabaceae (closely related to the stinging nettle family Urticaceae) has the peculiarity of being dioecious, meaning it only has male or female flowers, but not both. It’s believed to be originally central Asian, or maybe East-Asian, adapted to a temperate and steppe environment, where it can grow between 1 and 5 metres in height depending on local conditions. Like many annuals, it thrives on bare soil disturbed by human activities, which allowed it to spread widely alongside humans throughout Eurasia and later to the Americas.
Originally, the plant was likely harvested from the wild and used for three simultaneous purposes: the stem provides a type of sturdy fiber called bast, used for example to braid horse bridles (crucial in the steppes), hempseed produces a type of edible seed oil for cooking or livestock feeding, and its female flowers contain the potent drug THC. Over time, these three usages led to selected domesticated strains, so that today “hemp” refers mainly to fiber plants with negligible THC content. This ancient human selection has caused some confusion in taxonomy, with variable numbers of species and subspecies recognised over time. Hemp is one of the oldest sources of textile fiber, with remains of clothes found dating back to 6000 BC at least. It is also one of the oldest crops, probably grown in China as an oil-seed crop since around 4000 BC. Unfortunately indications of ancient use as a drug or medicine are more contentious.
One theory of domestication is that this crop was a “camp follower”. While being generally common in the central Eurasian landscape, it preferred disturbed and rich soils near rivers, and other well-irrigated areas. These would have been areas frequented by herds of mammals but also potentially nomadic human encampments. As the humans caused soil disturbance, created middens of discarded organic matter near the camp (enriching the soil), and collected plants including hemp from their vicinity, potentially spreading seeds around them accidentally, they would have created the perfect conditions for the plant to spread and thrive around them.5
As for “smokability”, burning this plant can be described as the “least bad” method of absorbing bioactive compounds into the organism. The main bioactive molecule of Cannabis is a cannabinoid known as Delta-9-Tetrahydrocannabinol, or THC for short. In the plant, it is part of a panoply of dozens of similar cannabinoids which play roles in mitigating environmental stresses such as grazing by herbivores and insects (by becoming less appetizing or downright toxic) but also fungi and bacteria, or drought.6 This compound is nearly exclusively found in the female flowers and their floral bracts (smaller leaves which subtend individual flowers or whole inflorescences to support and protect them).
Unlike with nicotine, the majority of THC is denaturated by pyrolysis (destruction by heat) before it reaches the lungs, and becomes inert. In total, approximately 70% of the bioactive THC is destroyed in this manner while smoking, with a further 3-20% lost to sidestream smoke. However, like nicotine, once it reaches the lungs it is able to enter the bloodstream and brain near-instantaneously. Historically, ingestion is also known (in herbal infusions, hashish as discussed below, etc), but is even more inefficient. The body’s metabolism and the preparation process denaturate into uselessness up to 95% of the THC absorbed orally, with peak effect of the remnant dose only occurring in average 1 or 2 hours after ingestion.7
Archaeological evidence
While the oldest trace of Cannabis in a human context is a finding of seeds from an 8000 BC Jōmon (hunter-gatherer) site in Japan, and in Europe seed impressions reported on a few Chalcolithic (copper-age) pottery finds from around 4000 BC, only two studies with vague methodology describe burned hemp seeds from remains of the Gumelniţa culture (also 4700-4000 BC). However it must be remembered that, as said previously, the seeds were also edible, and charred seeds of many non-psychotropic plants have been found in similarly ancient contexts (flax, rapeseed, cereals, etc).8 Furthermore as discussed above, THC is concentrated in the female flowers and bracts. While burning these could also lead to seeds contained in some fertilised flowers being burned (and they’d survive the scorching fire far better than green plant matter), there’d be no use in burning the seeds alone in this context as they contain very little THC.
The previous review [8] also mentions reports of the Yamnaya or one of their successor cultures the Catacomb (2800-2200 BC) as possibly being the first to burn hemp, as seeds, theorised hemp cordage impressions on pottery, and censers containing traces of soot and charcoal were found in various burials, notably Gatyn Calais in the North Caucasus. However it does not appear that analyses were carried out on the soot to conclusively detect chemical markers of Cannabis.
The oldest certain datation of Cannabis burning dates from the 800-400 BC period in the Eastern Pamir Mountains, at the crossroads between modern-day China, India, Tajikistan and Kyrgyzstan. Wooden burners filled with scorched stones, still containing traces of THC, were uncovered in tombs of the Jirzankal Cemetery. Other dried Cannabis remains were found from this time period in the Jiayi and Yanghai. The plant wasn’t burned at all tombs, as in the case of Yanghai it was also found crushed inside a bowl that likely served as a mortar. However, evidence of braziers around burial places would continue into the first millennia AD of this region with the Zoroastrian Sogdian people, but Cannabis is not confirmed in this context, only suspected.9 In any case the continued usage of funeral braziers for nearly 1500 years shows a tremendous cultural continuity in the area in this regard.
The most famous report of ancient Cannabis use is doubtless from Herodotus, who in the 5th century BC described how the Scythians and Massageteans of the central Eurasian steppe burned the plant on fires with hot stones inside small enclosed tents, practicing an early form of “hotboxing” (concentrating and retaining the THC-rich smoke in a small room). However, once again, the seeds alone are not rich in the active substance, it’s possible this is a confusion born from them throwing whole inflorescences on the fire, some of which may have been fertilised and bearing seeds. While being situated 3000 km away from the area the great historian mentioned, the contemporaneous Pazyryk kurgans seem to confirm his report, with a wooden tent frame being found containing a brazier filled with burned stones and carbonised wild hemp seeds. Nearby, a leather pouch contained a mixture of Cannabis sativa, Coriandrum sativum and Melilotus altissimus seeds.
The Scythians...take some of this hemp-seed, 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 vapour-bath can exceed; the Scyths, delighted, shout for joy, and this vapour serves them instead of a water-bath...
Herodotus, 4.75.
Strabo, in his Geography (written shortly before or after 0 AD), described Cannabis as growing abundantly in Colchis, now known to be on the Black Sea shore in modern-day Georgia. He further mentions “those who walk in smoke”, dancers of the Get tribe (related to Thracians, who also interacted with the Scythians, and lived on the lower Danube) who burned the plant’s flowers during performances.
Thus, with regards to Cannabis, there is definite proof that the plant was burned (whole or in part) on hot stones in a systematic fashion throughout central Asia and Eastern Europe as early as the 8th century BC. A practice which may even have gone back to the nomadic steppe people prior to 2000 BC. This usage seems to have been largely religious or ritual, relating to funeral practices in the case of the Pamir region. In some cases the presence of Cannabis in specific male graves was associated with shamanistic practices, possibly a display of a shaman’s tools of the trade.10
Making a hash of things
The other main Pre-Columbian method of consuming Cannabis outside of open-fire inhalation is the production of a thick resin known as “hashish” or simply “hash”. Textual proof of its existence only goes back to the 10th century AD in the Islamic world, although South-East Asian countries that had not converted to the faith at the time are known to have a long tradition of preparing this product. Briefly, hash is a concentrated mass of the sticky, aromatic resin secreted by the glandular trichomes (plant hairs) on the inflorescences of Cannabis. Because resin production was preferred over actual effect (as yield would be comparatively poor) it tended to be richer in CBD (which has an antagonistic, sedative effect compared to THC) than dried extracts of the whole plant.
The main traditional method of preparation involved filtering the harvested plant matter through increasingly fine textile screens (such as silk) or threshing them in order to collect the resin-rich glandular trichomes and resulting sticky excretions. Alternative labour-intensive methods also existed in Asia, such as walking through Cannabis fields wearing leather overcoats which, through friction, would accumulate a layer of resin that was scraped off, or manually rubbing female flowerheads with one’s hands to obtain the same product (this was far more time consuming).
The collected resin and trichomes are then pressed into brown “cakes” of hash, which are far more potent than the regular plant matter, as the resin is what contains all the active elements. Today, in the USA, hash still contains between 50 and 500% more THC than marijuana. This substance melts under heat and could be used in cooking, eaten as-is, etc (although as stated previously, oral ingestion is still an inefficient delivery mechanism for THC, and the “high” takes between 1 and 2 hours on average to manifest). It’s also possible to produce hash oil by realising a solvent extraction on the resin to solubilise the THC in an organic matrix. [5] [7] While hash is still consumed today, it is mostly associated with the “hash eaters”, or “hashashin”, a group (described as a sect) of fanatics who carried out politically motivated murders in the Middle-East during medieval times and partook of the substance.
Conclusion
As you’ve now seen, Cannabis is a plant with an exceedingly long and storied relationship with humans, such that three posts like this one could barely adequately cover its extent. Nevertheless, I hope this has allowed you a bit of insight over what our usage of the plant was like, before it became one of the most lucrative and contentious crops on the planet. From the nomads and shamans of central Asia, the fearsome warrior-cultures of the Eastern European steppe all the way to the most famous killers of the Islamic world, Cannabis never failed to leave its mark on the cultures it found itself in.
This post will be followed by a sequel in the near future that’ll deal with tobacco, and if there’s room left, other smokable plant-derived substances. So stay tuned!
Post-Scriptum
I’ve been made aware while writing this post that the video-essayist and youtuber Tom Rowsell made a popular video on this very topic a year or so ago. While he puts the accent on different aspects of the subject than this post, the two bear obvious similarities due to the fact we read the same recent scholarship on the same topic. I just wished to make it clear this is purely coincidental as I hadn’t watched the video before writing this post, and I feel both our approaches are different enough that they don’t detract from each other.
Hoffman & Hoffman, 1997. “The Changing Cigarette”. Journal of Toxicology and Environmental Health, 50:307–364.
Brewer, Roberts, Rowell. 2004. “Short-term distribution of nicotine in the rat lung.” Drug and Alcohol Dependence, Volume 75, Issue 2.
Small. 2015. “Evolution and Classification of Cannabis sativa (Marijuana, Hemp) in Relation to Human Utilization”. Bot. Rev. (2015) 81:189–294.
Park, Pauli, Gostin, Staples, Seifried, Kinney, Vanden Heuvel. 2022. “Effects of short-term environmental stresses on the onset of cannabinoid production in young immature flowers of industrial hemp (Cannabis sativa L.)” Journal of Cannabis Research volume 4, Article number: 1.
Russo. 2007. “History of Cannabis and Its Preparations in Saga, Science, and Sobriquet”. Chemistry & Biodiversity. Vol. 4.
McPartland & Hegman, 2018. “Cannabis utilization and diffusion patterns in prehistoric Europe: a critical analysis of archaeological evidence”. Vegetation History and Archaeobotany. Volume 27, pages 627–634.
Ren, Tang, Wu, Spengler, Jiang, Wang, Boivin. 2019. “The origins of cannabis smoking: Chemical residue evidence from the first millennium BCE in the Pamirs”. Sci. Adv. 5: eaaw1391.
H.-E. Jiang, L. Wang, M. D. Merlin, R. C. Clarke, Y. Pan, Y. Zhang, G. Xiao, X. Ding, 2016. “Ancient Cannabis burial shroud in a central Eurasian cemetery.” Econ. Bot. 70, 213–221.
Thank you, fascinating essay!