The National Institutes of Health keeps a plain-language list of 60 herbs and supplements on its MedlinePlus consumer site. A historian at San Diego State University once checked that list against twelve European medical texts running from the fifth century BC to a Spanish pharmacopoeia of 1865. Only 33 of the 60 turned up anywhere in that historical record. Exactly one turned up in all twelve sources.
That one is pennyroyal. The modern medical literature on pennyroyal is a poisoning literature.
Hold those two facts together and you have the real story of European herbal medicine. Europe did have a written drug tradition, and it stayed remarkably stable for roughly 25 centuries. It also broke, recently enough to put a date on, and most of what gets sold today under labels like “traditional European medicine” is not that tradition.
What follows is what the tradition contained, when it ended, what European law regulates in its place, and how to read the single most over-claimed citation in American herbal marketing.
The tradition that existed
The study is Paula De Vos’s 2010 paper in the Journal of Ethnopharmacology, and its method is simple enough to check. She took twelve medical works that list their ingredients in an index or table of contents, from the Hippocratic Corpus onward, and counted the simples, the old term for a single medicinal substance drawn from a plant, an animal or a mineral.
The twelve texts named 985 different simples between them. Of those, 439 showed up in at least four of the twelve. Plants made up 341 of that core group, with 45 animal substances and 49 minerals. Many did double duty outside medicine, turning up in cosmetics, pigments, varnish, candy and embalming fluid.
The surprise is whose book held the whole thing together. Across all the later works, correspondence with Dioscorides, the Greek army physician who compiled De Materia Medica in the first century AD, averaged 72 percent. Correspondence with the Hippocratic Corpus, four centuries older and far more famous, averaged 43.5 percent. Ancient works shared 75 to 85 percent of their ingredients with Dioscorides. Even the later periods settled in the 65 to 69 percent range.
Authors writing in the same era agreed with each other less than they agreed with one first-century Greek. Consistency ran to only 34 percent among the ancient writers and 56 percent among the medieval ones. Put the two ends of the chain side by side and the point lands harder still: the 1865 Spanish pharmacopoeia shares 45 percent of its ingredients with the Hippocratic Corpus, a higher figure than any medieval or early modern work in the set managed.
One detail shows how settled the material had become. Beginning with Saladino’s work of 1488, not one surveyed text printed the virtues, the therapeutic qualities attributed to each ingredient. Later pharmacopoeias gave the plant part and the compound recipes and assumed the reader already knew what each simple did. If you needed to check, you went back to Dioscorides.
The humoral frame, reported as the tradition’s own
In the Galenic system that organized all of this, every substance and every person carried qualities: hot or cold, moist or dry. Sickness was an imbalance of the humours, and a remedy’s job was to pull the balance back. That is the tradition’s own explanatory language, and this article reports it as the tradition’s language rather than as a description of how bodies work.
It did organize the shelf. Among the fourteen simples that appear in every one of the twelve texts, thirteen were classed as hot and dry, described as diuretic and expectorant, and used mainly to drive thick humours out of the chest. Eight were used in women’s health as emmenagogues, meaning drugs meant to bring on a delayed period, and, in the phrase the Journal of Ethnopharmacology study takes from the sources themselves, to “expel the fetus.” The texts do not say whether that meant ending a pregnancy or starting labor.
That same humoral framework is still carried as a living institutional system by Unani medicine, which has its own article in this library. Europe is the branch that dropped it.
Where the break shows up
Here is the measurement that settles the question. Of the 60 herbs and supplements on the NIH consumer list, 33 appear somewhere among the 985 historical simples. Only pennyroyal appears in all twelve texts. Only pennyroyal, almond and horehound rank among the 26 most common historical simples. Seventeen make the historical top 150, and 25 make the top 439.
Two millennia of near-agreement, then almost nothing in common across the last 160 years. De Vos puts the break between 1865 and the present and attributes it to the rise of organic and analytical chemistry, germ theory, synthetic drugs and the pharmaceutical industry.
The discontinuity is the finding. Anyone selling you an unbroken European lineage is selling something the numbers do not support.
What gets sold as “traditional European medicine” now
The clearest evidence about the modern label comes from the people promoting it. The Association for Natural Medicine in Europe, a Brussels-facing advocacy group, publishes a page with the institute it works with describing traditional European medicine as a collective term for natural diagnostic and treatment methods developed and rediscovered in Europe, used today in updated form.
It claims a spectrum of more than eighty natural methods. The list runs from light and air therapy, the Kneipp water cure and hydrotherapy through humoral therapy, iris diagnosis, complex homeopathy, neural therapy, ear acupuncture, bioresonance, kinesiology, meditation and art therapies.
Ear acupuncture is not a European humoral method. Bioresonance is not a monastic herbal. A list that holds all of these at once describes a coalition of modern practices sharing a continent, not a single inherited system.
The same page supplies its own date, saying a final stage of traditional European medicine arrived at the turn of the millennium into the twenty-first century. It also notes that Ayurveda and traditional Chinese medicine are recognized by the World Health Organization, and makes no equivalent claim for itself.
The US federal picture matches. Checked in October 2026, the health topics index at the National Center for Complementary and Integrative Health carries entries for Ayurvedic medicine, traditional Chinese medicine and naturopathy, and no entry for traditional European medicine, European herbalism or anthroposophic medicine.
What Europe regulates is bottles, not practitioners
The real European institution here is a product law. Directive 2004/24/EC, which amended the EU’s main medicines directive, created two routes onto the market for a herbal product, and the European Medicines Agency describes both.
The first is a well-established use marketing authorisation. It needs scientific literature showing the substance has been in well-established medicinal use inside the EU for at least ten years, with recognised efficacy and acceptable safety, assessed mostly from published data rather than new trials.
The second is a traditional use registration, and this is the one people mean when they say Europe approves herbal remedies. The agency states that no clinical tests and trials on safety and efficacy are required. In exchange the product must show 30 years of documented use, at least 15 of them inside the EU, must be meant for use without a doctor supervising it, and must not be injected.
Here is the part most coverage skips. A full quality dossier is required either way. The traditional route relaxes the evidence standard for whether the product works. It does not relax the standard for how the product is made.
The agency’s November 2023 questions-and-answers document adds the limits that matter to a shopper. Indications involving cancer, psychiatric disorders, infectious diseases such as hepatitis or influenza, cardiovascular disease such as heart failure, and metabolic disease such as diabetes are not acceptable for these products. Where an applicant relies on a comparable product to prove the tradition, the plant part, the preparation type and, for extracts, the main solvent all have to match. And the label and the patient leaflet must carry a statement that the stated use rests exclusively on long-standing use.
Volume: 1,719 traditional use registrations were granted between 2004 and the end of 2016, split 1,089 single-ingredient and 630 combination products. The United Kingdom led with 348, followed by Germany at 285, Poland at 215, Austria at 209 and Spain at 100. Spread across 28 countries and thirteen years, that is a modest trade, not a mass market.
Two warnings travel with those figures. They are nine years old, and the biggest single registrant has since left the EU. No more recent agency count of the scheme’s uptake could be located for this article.
Finally, the boundary. Asked directly whether this law reaches the practitioner, the agency answers that it regulates the placing of products on the market and not the practice of traditional medicine. Most individual herbal products are licensed by national authorities rather than centrally. The scheme is also open to systems that grew up elsewhere. That same questions-and-answers document names nine, Kampo and Unani among them, reporting them from a 2008 European Commission report, and adds traditional African medicine as a further example. Europe built a filing cabinet, not a profession.
What the numbers printed on an extract label do and do not verify is the subject of a separate article in this library.
Commission E, and the claim on the bottle
If you read American herbal marketing you have seen the line “approved by Germany’s Commission E.” It is the most over-claimed citation in the field, and the best English-language account of it is a March 1998 briefing by Edith Palmer, a senior legal specialist at the Law Library of Congress.
Commission E was created in 1977 as an advisory body helping German federal health authorities decide on marketing approval for herbal medications. Its statutory footing is section 25 of the German medicines act. It never licensed anything itself.
It published about 350 herbal monographs from its creation until the mid-1990s. The briefing says 1994 in its body and 1995 in its summary, so the honest version is mid-decade. The monographs existed to justify approving preparations already on the German market before 1978 and carried over under the 1976 law. That is a grandfathering exercise, not a trials programme.
After the 1994 reform the burden shifted. The briefing states that the producer now carries the burden of proof for evaluating a herbal drug. The panel itself ran to roughly 20 experts, including licensed physicians, state-licensed non-physician healing practitioners known as Heilpraktiker, and specialists in pharmacology, toxicology and biometrics.
The briefing also says on its own face that it has not been updated. It is a 1998 document and cannot answer a question about today.
So translate the marketing claim. “Commission E approved it” means an advisory panel wrote a favourable monograph, probably before 1995, inside a grandfathering review, under a legal regime that has since put the burden of proving efficacy on the company selling the bottle. That is a historical document, not a live government endorsement.
The one European system with unbroken institutions
There is a named European system with continuous institutions, manufacturers, clinics and professional bodies, and it is anthroposophic medicine, founded in the 1920s. Its legal position in Europe was settled by a court.
In Case C-84/06, decided on 20 September 2007, the First Chamber of the Court of Justice heard the Netherlands against a patients’ association, a doctors’ association and two manufacturers. The Court held that anthroposophic medicinal products may be marketed only if they have been authorised under one of the procedures in the EU medicines directive. It rejected the argument that because harmonisation is still developing, member states may keep separate national procedures for these products.
There is no bespoke European route for them. They go through ordinary marketing authorisation, homeopathic registration or traditional use registration like any other product. So the frequent claim that anthroposophic medicine is recognized in Europe is wrong as a statement about product law. The ruling came as a preliminary reference from the Dutch supreme court, in a dispute the manufacturers and patient groups had started against the Netherlands State, and the State’s reading prevailed.
At the same time, the European Coalition on Homeopathic and Anthroposophic Medicinal Products, which is the industry group for these very products, reported in June 2017 that from 1 August 2017 Switzerland brought anthroposophic medicine, classical homeopathy, traditional Chinese medicine and herbal medicine into mandatory basic health insurance, limited to conventionally licensed physicians holding an extra qualification recognised by the Swiss Medical Association. That is the industry’s account of a government decision. No Swiss federal source could be reached to confirm it, and this article reports it as the coalition’s claim.
The harm comes from Europe’s own plants
Start with the herb that proves the continuity, because it is also the one that proves the danger.
A 1996 paper in Annals of Internal Medicine from the San Francisco Bay Area Regional Poison Control Center reports four pennyroyal ingestions. One patient died. In that case a blood sample drawn 72 hours after the ingestion still held 18 nanograms of pulegone and 1 nanogram of menthofuran per millilitre, the compounds the liver converts into the toxic products. Reviewing 18 earlier published cases, the authors found moderate to severe toxicity in people exposed to at least 10 millilitres of pennyroyal oil, and described the herb as long used to end pregnancy despite potentially lethal liver effects.
Pennyroyal is the single ingredient with unbroken presence across 25 centuries of the European record, and it is also the one on the modern NIH list most clearly flagged as a poison. Longevity of use is not a safety finding, and this is the cleanest illustration of that anywhere in the material.
Comfrey is the second case. A 2000 review in Public Health Nutrition describes comfrey leaf as a substantial hazard whose main injury is hepatic veno-occlusive disease, a non-clotting blockage of the small veins draining the liver that can progress to cirrhosis and liver failure. The agents are pyrrolizidine alkaloids, including lasiocarpine and symphytine. The authors noted that distribution was banned in Germany and Canada while comfrey stayed freely available in the United States.
European regulators then set a number. The herbal committee of the European Medicines Agency adopted a public statement on 31 May 2016 putting the long-term ceiling at 0.35 micrograms of pyrrolizidine alkaloids per day, based on a 50 kilogram body weight and lifelong exposure, with a transitional allowance of 1.0 microgram a day for no more than three years while manufacturers clean up their supply. The limit is built on the framework used for genotoxic carcinogens, with liver hemangiosarcoma in rodents as the driving effect and sinusoidal obstruction syndrome as the acute concern.
The twist is which products the statement names as most affected. They are the European kitchen cabinet: St John’s wort, passionflower, chamomile flower, lady’s mantle, licorice root, lemon balm, peppermint leaf, sage leaf, dandelion and thyme. Read the attribution, though. The committee does not identify those ten itself. It reports them from a database compiled by BAH, the German medicines manufacturers’ association, and calls that preliminary evidence. The alkaloids are not in those herbs. They come from weeds growing in the same field and harvested with the crop.
Signs worth taking to a clinician quickly, with the product in hand: new yellowing of the eyes or skin, a swelling abdomen, or pain under the right ribs that starts within weeks of beginning a herbal preparation. Anyone pregnant or trying to conceive has reason to raise any product sold as bringing on a late period with a clinician first, because that is precisely the historical use category these plants occupied.
Four questions to ask of a “traditional European” claim
These are worth running through before the claim does any work in your decision.
Which tradition, and dated how? Ask whether the seller means a documented pre-1865 materia medica or a modern assemblage. If the answer includes ear acupuncture or bioresonance, it is the second kind.
What does the approval word cover? A German monograph from the grandfathering era, a national registration in one EU state, and a current marketing authorisation are three different things, and only the last involved anyone assessing efficacy to a modern standard.
What is the product allowed to claim? If a registered European herbal product cannot carry a diabetes or cardiac indication under EU rules, a US version of the same plant making that claim has not cleared a higher bar. It has cleared no bar.
Who answers for the contents? The EU route that waives trials still demands a full quality file and a named national authority. A supplement sold here carries neither.
The one thing to take away
European herbal medicine is a real historical subject with hard numbers attached, and almost none of those numbers support the way the phrase is used commercially. The tradition was anchored on one first-century text, held for about 25 centuries, and broke within living memory of the industrial era. What replaced it in Europe is a product-registration system that explicitly declines to regulate practitioners, plus one named twentieth-century system whose legal status a court settled in 2007.
For a reader in the United States, the practical content is thin and worth knowing. There is no European herbal practitioner credential recognized here, the registrations that do exist belong to products sold in Europe under European labels, and the single most repeated endorsement in the category expired as a live regulatory act around the time of the first web browser. Treat “traditional” as a claim about dates, and ask for the dates.
Sources
- De Vos P.. “European materia medica in historical texts: longevity of a tradition and implications for future use. Journal of Ethnopharmacology, 2010;132(1):28-47. PubMed. DOI: 10.1016/j.jep.2010.05.035 (full text read via PubMed Central: 12 texts, 985 simples, 439 core, 72% Dioscorides correspondence, 33 of 60 NIH-listed herbs, pennyroyal in all 12).” pubmed.ncbi.nlm.nih.gov
- Anderson IB, Mullen WH, Meeker JE, et al.. “Pennyroyal toxicity: measurement of toxic metabolite levels in two cases and review of the literature. Annals of Internal Medicine, 1996;124(8):726-34. PubMed. DOI: 10.7326/0003-4819-124-8-199604150-00004.” pubmed.ncbi.nlm.nih.gov
- Stickel F, Seitz HK.. “The efficacy and safety of comfrey. Public Health Nutrition, 2000;3(4A):501-8. PubMed. DOI: 10.1017/s1368980000000586.” pubmed.ncbi.nlm.nih.gov
- European Medicines Agency. “Herbal medicinal products (regulatory overview): the two legal routes, the 30-year and 15-year requirements, the statement that no clinical tests and trials on safety and efficacy are required, and the full quality dossier required on both routes.” ema.europa.eu
- European Medicines Agency, Committee on Herbal Medicinal Products. “Questions and Answers on the European Union framework for (traditional) herbal medicinal products, including those from a non-European tradition, EMA/HMPC/402684/2013 Rev. 1, 22 November 2023 (mandatory labelling statement; excluded indications; the legislation does not regulate the practice of traditional medicine).” ema.europa.eu
- European Medicines Agency, Committee on Herbal Medicinal Products. “Uptake of the traditional use registration scheme and implementation of the provisions of Directive 2004/24/EC in EU Member States, EMA/HMPC/322570/2011 Rev. 7, published 18 April 2017, status 31 December 2016 (1,719 traditional use registrations granted; Member State table).” ema.europa.eu
- European Medicines Agency, Committee on Herbal Medicinal Products. “Public statement on contamination of herbal medicinal products / traditional herbal medicinal products with pyrrolizidine alkaloids, EMA/HMPC/328782/2016, adopted 31 May 2016 (0.35 micrograms per day long-term limit; 1.0 microgram transitional limit for three years; genotoxic carcinogen framework).” ema.europa.eu
- Law Library of Congress, Global Legal Research Directorate (Edith Palmer). “Commission E in the German Drug Approval Process: Briefing Notes, March 1998, LL File No. 1998-1910 (created 1977; advisory; about 350 monographs to the mid-1990s; producer now bears the burden of proof; the report states it has not been updated).” tile.loc.gov
- Court of Justice of the European Union (judgment text hosted by IPPT). “Case C-84/06, Staat der Nederlanden v Antroposana and Others, First Chamber, judgment of 20 September 2007 (anthroposophic medicinal products may be marketed only if authorised under a procedure in the EU medicines directive).”
- National Center for Complementary and Integrative Health, National Institutes of Health. “Health Topics A-Z (checked October 2026: entries for Ayurvedic Medicine, Traditional Chinese Medicine and Naturopathy; no entry for traditional European medicine, European herbalism or anthroposophic medicine).” nccih.nih.gov
- Association for Natural Medicine in Europe (ANME e.V.) with InstiTEM. “The Healing System of Traditional European Medicine (TEM): the proponents' own definition, the claim of over eighty natural methods, and the statement that a final stage of TEM was reached at the turn of the millennium.” anme-ngo.eu
- European Coalition on Homeopathic & Anthroposophic Medicinal Products (ECHAMP). “Complementary Medicine in Switzerland Now a Mandatory Health Insurance Service, 19 June 2017 (industry coalition's account of the 1 August 2017 change; cited as the coalition's claim, not a Swiss government source).” echamp.eu