Is the Shanidar Cave evidence for Neanderthal plant medicine considered reliable?

It's contested. The original 1975 interpretation proposed that pollen clusters around a Neanderthal burial indicated intentional placement of medicinal flowers. Later analysis raised the possibility the pollen was introduced by a burrowing rodent. The evidence remains interesting but inconclusive — it's not considered settled proof of deliberate medicinal plant use.

Does ancient or traditional use of a plant mean it's been proven to work?

No. Traditional use documents that people found a plant useful and continued using it. It doesn't constitute clinical evidence by modern standards, which require controlled trials, standardised preparations, placebo comparison, and defined outcomes. Traditional use is best understood as a hypothesis generator that can inform (but not replace) formal research.

When was cannabis first recorded as being used medicinally?

Written references appear in Chinese texts possibly as early as 2700 BCE (attributed to the legendary Emperor Shennong), though scholarly dating of those sources varies considerably. Physical archaeobotanical evidence from a burial in western China, published in 2019 in Science Advances, dates high-THC cannabis use to around 500 BCE. Cross-cultural references also appear in Egyptian, Indian, and Mesopotamian sources.

What is the endocannabinoid system and how does it relate to ancient plant use?

The endocannabinoid system is a network of receptors and signalling molecules in the human body, only characterised scientifically in the late 1980s–90s. Some compounds in cannabis and other plants interact with this system. Ancient users had no knowledge of its existence — they worked empirically from observed effects. The system's existence explains some of the pharmacological plausibility behind ancient observations, but the two bodies of knowledge developed entirely independently.

Ancient Use of Medicinal Plants

By Hannah Bui · 5 June 2026 · 7 min read

Two botanical drawings of pink roses on pages.

Sixty thousand years is a long time to be right about something. That's roughly the age of the oldest putative evidence for plant-based medicine , pollen found at the Shanidar Cave site in northern Iraq, associated with a Neanderthal burial. The interpretation has been contested for decades (more on that below), but even the conservative estimates put deliberate human plant use well before writing, before agriculture, before most things we associate with "civilisation".

I find that genuinely startling. And I also think it's one of the most frequently misrepresented facts in the plant medicine space, so let's go through what the record actually shows.

What the archaeological evidence says , and doesn't say

The Shanidar pollen story is a good case study in how evidence gets inflated. Ralph Solecki's 1975 paper in Science proposed that the high concentration of flower pollen around Neanderthal remains indicated intentional burial with medicinal or symbolic plants. For years this was cited breathlessly as proof that even Neanderthals were herbalists.

Later analysis suggested the pollen may have been introduced by a burrowing rodent. The Persian jird, specifically, is known to cache flower heads underground. That doesn't definitively disprove ritual plant use, but it substantially weakens the claim. The short version is: the evidence is interesting and inconclusive, not settled.

More solid ground comes from written records. The Ebers Papyrus, dated to around 1550 BCE but likely transcribed from much older sources, is one of the earliest surviving medical texts and lists hundreds of plant-based preparations. Mesopotamian clay tablets from roughly the same era describe willow bark and liquorice in specific preparations. Traditional Chinese medicine texts, including the Shennong Bencao Jing (compiled perhaps around the 1st century CE but encoding far older oral traditions), catalogue over 350 plant substances.

Cannabis appears across several of these records. Chinese texts reference it medicinally. Indian Ayurvedic tradition includes it. Egyptian papyri mention it. This cross-cultural convergence is one reason historians take its ancient use seriously, even where the specific claims attached to it require careful reading.

Ancient use and the phytomedicine tradition

The word phytomedicine refers formally to plant-derived preparations used in a medical context, typically involving some degree of processing and standardisation. Ancient herbalism was its precursor , and a far less standardised one.

This distinction matters. When we talk about ancient plant use, we're talking about whole plants, crude extracts, teas, poultices, and resins prepared without any capacity to quantify active constituents. Modern researchers examining these traditions have to work backwards: identifying which plants were used, hypothesising which chemical compounds those plants contain, then asking whether the chemistry could plausibly account for the traditional use.

Sometimes the fit is striking. Willow bark contains salicin, a precursor to salicylic acid (the basis of aspirin), and it appears in records from ancient Egypt, Greece, Mesopotamia, and Indigenous North American traditions. Opium poppy preparations appear in Sumerian tablets from around 3400 BCE. These aren't coincidences. They're empirical discoveries made by people with no formal chemistry but a great deal of observational patience.

Other traditional uses are harder to map onto modern pharmacology, and some are almost certainly based on ritual, symbolic, or placebo-adjacent mechanisms that don't translate neatly into a bioactive compound. I'd argue we do a disservice to both ancient knowledge systems and to modern science when we flatten everything into "they knew all along." They knew some things, empirically. Other things reflected cosmology, not chemistry.

Cannabis and the ancient record specifically

Cannabis gets particular attention in this space, partly because of current regulatory interest and partly because it genuinely does appear across multiple ancient traditions on different continents.

Archaeobotanical evidence from a burial site in western China, published in 2019 in the journal Science Advances, identified high-THC cannabis plants placed with a probable shaman, dated to around 500 BCE. This is direct physical evidence, not just a textual reference; and it suggests the psychoactive properties were known and possibly valued specifically.

The plant contains dozens of compounds now studied in detail: tetrahydrocannabinol (THC), cannabidiol (CBD), cannabigerol (CBG), and many others, alongside terpenes like myrcene, linalool, and beta-caryophyllene that appear across numerous medicinal plant species. Ancient users had no framework for any of these individual molecules, but they were working with the whole plant, what we'd now call a full-spectrum preparation, even if that term wasn't in use.

The endocannabinoid system, the receptor network these compounds interact with, wasn't characterised until the late 1980s and early 1990s. The fact that the system exists, and that plants produce compounds structurally similar to the body's own endogenous signalling molecules; is one of the more remarkable convergences in recent pharmacology. But that's a story about modern biochemistry, not ancient knowledge. Ancient users didn't know about anandamide or the CB1 receptor. They knew what they observed.

Other traditions worth understanding

Cannabis is overrepresented in modern discussion partly because of commercial interest. The broader ancient record is richer and weirder than a single plant.

Ayurvedic medicine, formalised in texts like the Charaka Samhita (roughly 400–200 BCE, though debated), draws on hundreds of plant substances and represents one of the oldest continuously practised systems of plant-based medicine on earth. Traditional Chinese medicine has a parallel continuity. Unani medicine, derived partly from Greco-Arabic traditions, is still formally practised across parts of South Asia and the Middle East today.

Indigenous Australian plant knowledge is one of the most extensive and least studied bodies of botanical medicine anywhere. Country varies enormously across this continent, and Aboriginal and Torres Strait Islander peoples developed detailed, place-specific knowledge over at least 65,000 years of continuous occupation. That knowledge has been systematically marginalised, and in many cases lost, through colonisation. The few ethnobotanical studies conducted, often in partnership with community knowledge holders, document uses of plants including Acacia species, Eremophila, and various coastal plants, but this represents a fraction of what existed and what continues to be held in community.

I went down a rabbit hole reading some of the earlier 20th-century ethnobotanical records on this last winter, between cold-water swims at Sandy Bay. The depth of specificity in those records, particular preparation methods, seasonal timing, plant part selection; looks nothing like the vague "used medicinally by ancient peoples" language that shows up in supplement marketing. The knowledge was and is technical, place-bound, and precise.

What ancient use does and doesn't tell us about modern applications

Here's where I want to be direct, because this is where historical claims get misused.

Ancient or traditional use of a plant is evidence that people found it useful and kept using it. It is not, by itself, evidence that the plant does what any particular modern claim says it does. Traditional use precedes clinical trial methodology, placebo controls, blinding, standardised extracts, and quantified dosing. A preparation used for centuries might reflect a real pharmacological effect, a strong placebo response, a confounded observation, or cultural transmission of a practice that outlived any original rationale.

Modern research has confirmed the biological plausibility of some traditional uses. It has also quietly buried others. The appropriate response to ancient use is to treat it as a hypothesis generator, not a conclusion. When an ethnobotanist documents a plant used across multiple unconnected cultures for similar purposes, that's a reasonable signal that something pharmacologically interesting may be happening. When that observation feeds into a properly designed research programme, with pharmacokinetics studied, bioavailability characterised, mechanisms hypothesised and tested, then we're getting somewhere.

The history is real, the depth of observation behind traditional systems is real, and some of the chemistry turns out to be very interesting indeed. But "used for thousands of years" is a start, not a finish.

Sources

, Hannah Bui, Evidence & research-literacy writer

]]>

Common questions

Is the Shanidar Cave evidence for Neanderthal plant medicine considered reliable?: It's contested. The original 1975 interpretation proposed that pollen clusters around a Neanderthal burial indicated intentional placement of medicinal flowers. Later analysis raised the possibility the pollen was introduced by a burrowing rodent. The evidence remains interesting but inconclusive — it's not considered settled proof of deliberate medicinal plant use.
Does ancient or traditional use of a plant mean it's been proven to work?: No. Traditional use documents that people found a plant useful and continued using it. It doesn't constitute clinical evidence by modern standards, which require controlled trials, standardised preparations, placebo comparison, and defined outcomes. Traditional use is best understood as a hypothesis generator that can inform (but not replace) formal research.
When was cannabis first recorded as being used medicinally?: Written references appear in Chinese texts possibly as early as 2700 BCE (attributed to the legendary Emperor Shennong), though scholarly dating of those sources varies considerably. Physical archaeobotanical evidence from a burial in western China, published in 2019 in Science Advances, dates high-THC cannabis use to around 500 BCE. Cross-cultural references also appear in Egyptian, Indian, and Mesopotamian sources.
What is the endocannabinoid system and how does it relate to ancient plant use?: The endocannabinoid system is a network of receptors and signalling molecules in the human body, only characterised scientifically in the late 1980s–90s. Some compounds in cannabis and other plants interact with this system. Ancient users had no knowledge of its existence — they worked empirically from observed effects. The system's existence explains some of the pharmacological plausibility behind ancient observations, but the two bodies of knowledge developed entirely independently.
Is Indigenous Australian plant knowledge documented in formal research?: Some of it is, through ethnobotanical studies conducted — ideally in partnership with community knowledge holders. But the documented record represents a fraction of what existed and what continues to be held in community. Much knowledge was lost or disrupted through colonisation, and what remains is community-held cultural property, not a freely available research database.