When herbalism became pharmacology

What it is

The process by which a plant in traditional use becomes an identified molecule with a known dose — and the handful of cases where that has actually happened.

It is the strongest argument herbal medicine has, and it is worth stating precisely rather than loosely, because the loose version — "most drugs come from plants" — is not quite true and the precise version is more interesting.

Where it came from

William Withering is the founding case and the best one.

In 1775, as a physician in Birmingham, he heard of an old woman in Shropshire whose family recipe was succeeding with dropsy — the massive fluid retention of heart failure — where doctors were not. The recipe contained around twenty herbs. Withering, in his own account, worked out that the active one was foxglove.

What he did next is why he matters. He did not simply adopt the remedy. He spent ten years giving it to patients, recording who improved and who did not, varying the preparation and the dose, and documenting the toxicity in detail — the nausea, the disturbed colour vision, the slowed and irregular pulse that told him he had gone too far. An Account of the Foxglove (1785) reports 163 cases including the failures.

He established that the plant part, the drying method and the harvest time all mattered; that the useful dose sat very close to the harmful one; and that it worked on the heart rather than on the kidneys, which was not what anyone expected. Digoxin, isolated from foxglove, is still in the pharmacopoeia.

How it is used

As the reply to blanket dismissal, and — properly understood — as the reply to blanket credulity as well.

What we can and cannot say

We can say the list is real and short.

• Aspirin. Willow bark for pain and fever is in Sumerian, Egyptian and Greek sources. Salicin was isolated in the 1820s; Felix Hoffmann at Bayer produced acetylsalicylic acid in 1897, modifying the natural compound to reduce the stomach damage that the plant version caused.
• Digoxin, from foxglove, as above.
• Morphine, isolated from opium by Friedrich Sertürner around 1804 — the first alkaloid ever isolated from a plant, and the beginning of the whole enterprise.
• Quinine, from cinchona bark, used against malaria in the Andes and brought to Europe in the seventeenth century.
• Artemisinin, and this is the modern showpiece. Tu Youyou, working under a Chinese military programme during the Vietnam War, screened traditional texts for antimalarials and found a fourth-century reference by Ge Hong to soaking qinghao in cold water. The cold-water detail mattered: the standard hot decoction destroyed the compound. She took the first dose herself. She received the Nobel Prize in Medicine in 2015, and artemisinin combination therapy is now the first-line malaria treatment worldwide.
• Paclitaxel, from Pacific yew bark, found by mass screening rather than by tradition.
• Vincristine and vinblastine, from Madagascar periwinkle, which was being investigated for traditional diabetes use when its effect on white cell counts was noticed instead.

We can say what the pattern actually shows, and it cuts both ways.

In favour of tradition: these are not coincidences. Traditional use functioned as an enormous uncontrolled screening programme run over millennia, and it identified genuinely active plants. Artemisinin in particular was found because someone took a fourth-century text seriously enough to read it carefully — including a preparation detail that turned out to be the whole thing.

Against the herbal preparation: in every single case, the tradition was improved on, and substantially. The isolated compound has a known dose where the plant does not. Aspirin was chemically modified because salicin damaged stomachs. Digoxin is dosed to the milligram because foxglove leaf is not consistent enough to be safe. Tu Youyou's contribution was an extraction method, not the plant.

The traditions found the plants. Pharmacology found the doses, and the doses are what make them usable.

We can say the "most drugs come from plants" claim needs a number. Depending on how you count, roughly a quarter to a third of small-molecule drugs approved in recent decades are natural products or derived from them — including microbial and marine sources, not just plants. That is a substantial minority and it is not a majority, and the honest version is more persuasive than the inflated one.

And we can say the failures matter as much as the successes. Bioprospecting programmes have screened enormous numbers of traditionally used plants and found very little. The National Cancer Institute screened over 100,000 extracts across decades to produce a handful of agents. Most traditionally used plants, tested properly, do not do what they are said to do — which is exactly what makes the ones that do interesting.

We can also say who was not paid. Artemisinin, vincristine, quinine and many others were built on knowledge held by communities that received nothing. The Nagoya Protocol on access and benefit-sharing exists because of this pattern, it is recent, it is weakly enforced, and the territory it covers overlaps directly with cultural appropriation as this site treats it elsewhere.

We cannot say tradition is a reliable guide to what works. What we can say is that it was a reliable guide to where to look — and that every time something was found there, the next step was to stop using the plant and start using the molecule.

Further reading

• William Withering, An Account of the Foxglove and Some of its Medical Uses (1785) — readable, and a model of honest reporting two centuries early.
• Tu Youyou's Nobel lecture (2015).
• David Newman and Gordon Cragg's periodic surveys of natural products among approved drugs, for the actual numbers.