Star Anise: History and Traditional Use
Star anise is the eight-pointed woody fruit of Illicium verum, an evergreen tree of the hills where south-west China meets northern Vietnam. It is the sweet, warm note in five-spice powder and in a bowl of phở, and a spice Europe met through Chinese tea and turned into liqueurs. It also has two histories no other kitchen spice shares: a poisoning story, in which a near-identical Japanese relative, Illicium anisatum, was found mixed into star anise teas that put infants in hospital across Europe and America between 2000 and 2004; and a pharmaceutical story, in which most of the world's harvest briefly went into the antiviral drug oseltamivir because the fruit is unusually rich in shikimic acid. This article follows kitchen, clinic and chemistry, and says plainly where the record is firm and where it is only tradition.
Table of Contents
- Origins, Botany and a Fruit With Three Names
- Star Anise in Chinese Medicine and the Chinese Kitchen
- Vietnam: Lạng Sơn, Phở and the French Distilleries
- The Journey West: Badian, Tea and Liqueurs
- The Chemists Arrive: Anethole and Shikimic Acid
- The Poison in the Look-Alike: Anisatin
- Adulterated Teas and Infant Seizures, 2000–2004
- From Spice Rack to Pharmacy: Shikimic Acid and Oseltamivir
- Tradition, Evidence and Star Anise Today
- Key Research Papers
- Connections
- Featured Videos
Origins, Botany and a Fruit With Three Names
Star anise is not a seed, and it is not related to anise. It is the dried fruit of Illicium verum, an evergreen tree of 8 to 15 metres with glossy leaves and small waxy pink-to-red flowers. The fruit is a ring of usually eight boat-shaped segments (follicles) around a central axis, each holding one shiny seed; it is picked unripe and sun-dried, which is when it turns rust-brown and develops its scent. That scent is the molecule anethole, the same one that flavours true anise (Pimpinella anisum, a Mediterranean herb of the carrot family) and fennel — a coincidence of chemistry between unrelated plants, and the reason a Chinese tree fruit carries the name of a European seed.
The tree belongs to one of the oldest lineages of flowering plants; modern classifications place Illicium in the Schisandraceae beside the five-flavour berry Schisandra, where older books gave it its own family, Illiciaceae. Its native range is small — the limestone hills of Guangxi and neighbouring provinces of south-western China and the mountains of north-eastern Vietnam — and almost the entire world supply still comes from that one region.
The fruit carries three names, each a small piece of history. In Chinese it is bā jiǎo, "eight horns" or "eight corners," and in medicine bā jiǎo huí xiāng, "eight-cornered fennel." The old European trade name badian (French badiane, Russian badyan) is usually traced through Persian bādiyān, a word for fennel, and possibly back to the Chinese "ba jiao" itself — the dictionaries call the ultimate origin uncertain. The scientific name came last. Illicium is from the Latin illicere, "to allure," for the fragrance; verum means "true." Joseph Dalton Hooker, then director of Kew, published the name Illicium verum in Curtis's Botanical Magazine in 1888 (plate 7005), which is why it reads Illicium verum Hook.f. The "true" was pointed: Linnaeus had founded the genus in 1759 on the Japanese species, Illicium anisatum, and for generations European writers had muddled the edible Chinese fruit and its toxic Japanese cousin. Separating the two was not a botanical nicety; as later sections show, it was a matter of life and death.
Star Anise in Chinese Medicine and the Chinese Kitchen
Star anise's earliest documented home is southern Chinese medicine and cooking. How early is harder to pin down than the internet suggests: several modern reviews repeat that the fruit was first recorded in a Tang-dynasty (618–907) medical text, but we could not confirm that in a primary source and do not assert it. What is secure is that by the late Ming it had a settled place in the materia medica. Li Shizhen's Bencao Gangmu (Compendium of Materia Medica, completed in 1578 and printed in 1596) treats the "large fennel" alongside the ordinary fennel seed, describes it as hot in nature where fennel is mild, and warns that eating too much of it injures the eyes and provokes sores — a caution that reads differently once you know about the toxic look-alike.
In the language of traditional Chinese medicine — a framework for reasoning, not a modern pharmacology — ba jiao hui xiang is pungent and warm, enters the liver, kidney, spleen and stomach channels, "dispels cold and warms the yang" and "regulates the flow of qi to relieve pain"; the classic indications are cold-type abdominal colic, vomiting and cold pain in the stomach, and low-back pain attributed to kidney deficiency. Stripped of the theory, those are the complaints for which every food culture has used a warming, carminative (wind-expelling) spice, and the modern Chinese Pharmacopoeia still lists the dried fruit as Anisi Stellati Fructus with much the same indications. None of this is evidence that the spice treats those conditions; it is evidence of how long people have reached for it.
The kitchen record is at least as important as the medical one. Star anise is one of the five spices in wǔ xiāng fěn, five-spice powder, with cassia, clove, fennel seed and Sichuan pepper; it is the signature aromatic of the red-braised (hóng shāo) dishes of eastern and southern China, where whole stars simmer for hours in soy, sugar and rice wine. A single star in a large pot is the traditional dose, and that pattern — a whole, recognisable fruit, in small quantity, in a shared dish — is exactly what kept people safe for centuries; it was the concentrated infusions of a later era that caused harm.
Vietnam: Lạng Sơn, Phở and the French Distilleries
The Vietnamese half of star anise's homeland is Lạng Sơn province on the Chinese border, with smaller plantings in Quảng Ninh and Cao Bằng. In Vietnamese the spice is hồi or đại hồi, "great fennel," and the tree is a fixture of the upland economy of the Tày, Nùng and Dao peoples, harvested by climbing and shaking the fruit down. Trees begin to bear around their fifth year, improve for decades and are reported to live a century or more, so an orchard is an inheritance rather than a crop. Star anise also gives phở its unmistakable perfume: the beef noodle soup that emerged around Hanoi in the early twentieth century, its broth simmered with charred onion, ginger, star anise, cassia, clove and black cardamom, is the most famous use of the spice outside China.
The French colonial administration saw the tree as an industrial crop: farmers in Lạng Sơn credit the French with encouraging plantations in the early twentieth century and building distilleries there to steam-distil star anise oil for Europe's liqueur and confectionery trades. A second expansion came under Vietnam's reforestation programmes of the late 1990s and early 2000s, when degraded hill land was allocated to households for replanting and star anise was a recommended species. Vietnam is today the world's second producer after China, and Lạng Sơn supplies the large majority of its crop.
The Journey West: Badian, Tea and Liqueurs
The story most often told about star anise's arrival in Europe is that the English navigator Thomas Cavendish brought it back to London from the Philippines at the end of his circumnavigation (1586–1588), and that Queen Elizabeth's apothecary, Hugh Morgan, took it up and sent samples to his correspondents on the Continent. It is repeated in dozens of spice books, with the date given variously as 1578, 1588 or 1601, but we could not trace it to any sixteenth- or seventeenth-century document, and Morgan's biographers credit him with introducing vanilla, not star anise. Treat it as tradition: plausible, since Cavendish did call at the Philippines in January 1588, but unproven.
The route that is documented is humbler. Star anise came west with Chinese tea. The Russian caravan trade across Siberia and Mongolia, opened in the seventeenth century, carried the fruit with the tea leaf, and Russians and Germans learned to drop a star into the pot — hence the old name "Siberian anise," though it never grew anywhere near Siberia. By the eighteenth century the habit had reached the Dutch, Europe's tea merchants, and the first edition of the Encyclopædia Britannica (1771) needed one sentence for badiane: "the seed of a tree which grows in China, and smells like anise-seed. The Chinese, and the Dutch in imitation of them, sometimes use the badiane to give their tea an aromatic taste." That was the whole European understanding of the spice: a Chinese tea flavouring that smelled of anise.
Its European future lay in the bottle. Anise-flavoured spirits are an old Mediterranean family — anisette, ouzo, arak, rakı, sambuca — and distillers found that star anise oil, arriving in bulk from Indochina by the late nineteenth century, delivered the same anethole as true anise at a fraction of the price. After France banned absinthe in 1915 its distillers rebuilt their recipes around anise, and the pastis launched in Marseille in 1932 combined liquorice root, green anise, fennel and star anise. Star anise also went into mulled wine, spiced preserves and confectionery across northern Europe, where whole stars still float in Glühwein at Christmas.
The Chemists Arrive: Anethole and Shikimic Acid
The dried fruit yields about 5 to 8 per cent essential oil, and 85 to 90 per cent of that oil is one molecule, trans-anethole — a sweet-smelling phenylpropanoid (a small aromatic compound on a nine-carbon skeleton) some thirteen times sweeter than sugar. Nineteenth-century chemists reached it through European anise rather than the Chinese fruit: in 1832 Jean-Baptiste Dumas showed that the crystals separating from anise oil and fennel oil in the cold were the same compound and found its formula, and Charles Gerhardt is credited with naming it anethole in 1845, from the Latin for anise and oil. Star anise oil later proved a richer source than either European plant, and the modern trade was born of that finding: it is cheaper to distil anethole from a Vietnamese tree than from a Mediterranean seed.
The second molecule in star anise's history was found in the wrong star anise. In 1885 the Dutch chemist Johan Frederik Eijkman, working in Japan, published a study of the constituents of Illicium religiosum — the Japanese temple tree, shikimi, now called Illicium anisatum — and isolated a new crystalline acid from its fruit, which he named shikimic acid after the plant's Japanese name. Its structure took another half-century: the Swiss-based chemists Hermann Fischer and Gerda Dangschat settled the constitution of shikimic acid in 1934, as the fifth paper in a series on quinic acid and its relatives. Nobody then imagined the compound mattered. That changed in the 1950s, when biochemists found shikimic acid to be a universal intermediate in the pathway by which plants and bacteria build aromatic amino acids — the "shikimate pathway" — and again in the 1990s, when it became the starting point for an influenza drug. Chinese star anise happens to accumulate it in extraordinary quantity, several per cent of the fruit's dry weight; that accident of plant metabolism is the subject of a later section.
The Poison in the Look-Alike: Anisatin
Japanese star anise, Illicium anisatum, has never been a food. In Japan it is shikimi, a sacred tree of Buddhist temples and graves whose leaves and bark are burned as incense, and its fruit was known for centuries to be poisonous. That fruit looks very much like the Chinese spice: a star of woody segments, usually more numerous (ten to thirteen rather than eight), thinner, with sharper upturned tips and a weak resinous rather than sweet smell. Whole and side by side an expert can tell them apart; broken, ground or steeped in a tea bag no one can.
The toxic principle was isolated and named in 1952 by John F. Lane and colleagues in the United States, who extracted a convulsant substance from Illicium anisatum fruit and called it anisatin. Its unusual structure — a sesquiterpene bearing two lactone rings, one of them a strained four-membered beta-lactone — was solved in 1965 by Kiyoyuki Yamada and Yoshimasa Hirata's group at Nagoya. In 1981 Yoshihisa Kudo, Jun-Ichiro Oka and Yamada showed what it does: anisatin is a potent antagonist of GABA, the brain's main inhibitory neurotransmitter, and work from Toshio Narahashi's laboratory in 1999 characterised it as a non-competitive blocker of the GABA-A receptor channel, much like the convulsant picrotoxin. GABA is the brain's brake; anisatin jams it, and the result is tremor, vomiting, agitation and seizures. Related compounds — the veranisatins — occur in trace amounts even in genuine Chinese star anise, which is one reason concentrated infusions are a different proposition from a star in a stockpot.
Chinese pharmacology had, in its own idiom, recorded the danger centuries earlier: the Bencao Gangmu's warning that too much "large fennel" injures the eyes and raises sores is what one would expect of a market in which a toxic relative sometimes passed for the real thing. The twentieth century added the molecule, the mechanism and, at its end, the poisonings that made the distinction a matter of public health.
Adulterated Teas and Infant Seizures, 2000–2004
Across the Hispanic world and much of Asia, a weak infusion of star anise is a traditional remedy for infant colic. From the spring of 2000 Spanish paediatricians began to see young infants, most under three months, arriving with irritability, tremor, vertical nystagmus (jerking eye movements), myoclonic jerks, vomiting and sometimes seizures after being given star anise tea. A national working group traced the cases, found that the odds of illness rose steeply with the amount given, and — the crucial finding — had laboratories show that commercial Illicium verum on the Spanish market was contaminated with Illicium anisatum. The Ministry of Health withdrew every star-anise product from sale in October 2001 and did not permit sales again until May 2002. One published case from the outbreak is a one-month-old girl in convulsive status who needed three doses of diazepam.
The Netherlands followed within months. At the end of September 2001 the Dutch health inspectorate and poisons centre were alerted to illness after a herbal tea; 63 people reported malaise, nausea and vomiting two to four hours after drinking it, 22 were admitted to hospital and 16 had generalised tonic-clonic seizures. NMR analysis of the tea confirmed anisatin, the product was withdrawn, and everyone recovered. In the United States, physicians at a Miami children's hospital reported seven infants with neurological reactions after home-brewed star anise tea and found Chinese star anise on the American market contaminated with the Japanese species; their 2004 paper in Pediatrics called for stricter import regulation and concluded that star anise tea should no longer be given to infants at all. French paediatricians published a series of infant poisonings in 2003, and a Swiss group later described a baby whose poisoning presented as an apparent life-threatening event.
The regulators acted. On 10 September 2003 the US Food and Drug Administration warned consumers against drinking teas brewed from star anise, citing some 40 cases of illness over two years, 15 of them in infants, and noting that the Japanese and Chinese species could not be distinguished by eye in dried or processed form. European authorities tightened import controls and testing on star anise consignments, and by 2006 German food chemists had published chromatographic and mass-spectrometric methods that detect anisatin in star anise at trace levels, now routine in quality control. None of these actions banned the culinary spice, and none needed to. The lesson of 2000–2004 was narrower and sharper: unlabelled or ground material of uncertain origin, brewed strong and given to a baby, is the one way star anise has demonstrably hurt people.
From Spice Rack to Pharmacy: Shikimic Acid and Oseltamivir
The pharmaceutical chapter opens in 1997, when a team led by Choung U. Kim at a California biotechnology company published a new family of influenza neuraminidase inhibitors — molecules shaped to block the enzyme the virus uses to cut itself free from an infected cell. The best of them, GS-4104, became oseltamivir. Its core is a six-membered ring bearing three neighbouring groups in a precise three-dimensional arrangement, expensive to build from scratch; nature already makes it, because shikimic acid has the right ring and the right handedness. In 1998 process chemists led by John Rohloff published a practical route from quinic or shikimic acid to oseltamivir in about a dozen steps, and shikimic acid won because it is two steps shorter. Its commercial source was Chinese star anise; roughly 30 kilograms of fruit yielded a kilogram of the acid.
Then, in 2005, avian influenza H5N1 spread through poultry across Asia into Europe, governments began stockpiling oseltamivir, and the manufacturer announced that it could not keep up. The bottleneck, it said, was shikimic acid: the tree grows in a few provinces of China, the fruit is harvested between March and May, and the company was already buying most of the crop — figures of up to 90 per cent of the world harvest were widely reported. Prices spiked in Guangxi and Lạng Sơn, farmers planted new orchards, and for a season a kitchen spice was front-page news as a strategic commodity.
The way out had been prepared before the crisis. Because the shikimate pathway exists in bacteria too, metabolic engineers had spent the 1990s persuading Escherichia coli to over-produce shikimic acid from glucose, and a 2003 paper from an industrial group in Germany and the Netherlands described strains that made the process viable at scale. Fermentation was scaled up from 2005; by 2006 the manufacturer reported that most of its shikimic acid came from bacteria, by around 2012 the switch was essentially complete, and star anise prices fell back. The episode left one durable piece of folklore that the science does not support: that star anise tea treats influenza. Shikimic acid has no antiviral activity. It is a starting material, a convenient source of one carbon ring, and the part of oseltamivir that blocks the viral enzyme is built in the factory. The spice stands to the drug as sand stands to a silicon chip.
Tradition, Evidence and Star Anise Today
What does modern evidence say about the traditional claims? Less than one might expect. Reviews of Illicium verum in 2011 and 2020 found a long list of laboratory findings — antimicrobial, antioxidant, insecticidal, analgesic and sedative effects of extracts and of trans-anethole in cells and animals — and almost no controlled trials in people. Anethole's carminative and antispasmodic action, studied mainly through anise and fennel, is the strongest thread from the tradition (colic, wind, cold pain in the belly) to a plausible mechanism: the digestive use is consistent with the chemistry, and no one has properly tested it. The fruit has a quality monograph in the European Pharmacopoeia and is an official drug of the Chinese Pharmacopoeia — statements about identity and purity, not efficacy.
Where the record is unambiguous is safety, and it points two ways at once. The spice as cooked — a whole star or two in a large dish — has an unblemished history. The spice as an infusion for infants has a documented record of harm, and every regulator that has looked, from the Spanish ministry in 2001 to the FDA in 2003, has said the same thing: do not give star anise tea to babies. This site repeats it.
Star anise today is a large and living crop. China is reported to produce on the order of 90,000 tonnes a year, chiefly in Guangxi and Yunnan; Vietnam roughly 20,000 tonnes, overwhelmingly from Lạng Sơn, where organic certification has become the route to premium export markets now that the pharmaceutical boom is over. Most of the harvest is still distilled for anethole or sold whole for cooking. The eight-pointed fruit that flavoured Ming braises and Dutch teapots, that poisoned infants when its cousin was passed off in its place, and that briefly fed a global drug stockpile, is once again what it began as: the most beautiful spice in the kitchen, and one to use whole.
Key Research Papers
Reviews of Illicium verum, the chemistry landmarks for shikimic acid and anisatin, the clinical reports of the 2000–2004 poisonings and the papers behind the oseltamivir connection. Historical primary texts are named in the article, not cited here.
- Wang GW, Hu WT, Huang BK, Qin LP (2011). Illicium verum: a review on its botany, traditional use, chemistry and pharmacology. Journal of Ethnopharmacology. — PubMed PMID: 21549817
- Patra JK, Das G, Bose S, Banerjee S, Vishnuprasad CN, del Pilar Rodriguez-Torres M, Shin HS (2020). Star anise (Illicium verum): Chemical compounds, antiviral properties, and clinical relevance. Phytotherapy Research. — PubMed PMID: 31997473
- Eijkman JF (1885). Sur les principes constituants de l'Illicium religiosum (Sieb.) (Shikimi-no-ki en japonais). Recueil des Travaux Chimiques des Pays-Bas. — doi:10.1002/recl.18850040202
- Lane JF, Koch WT, Leeds NS, Gorin G (1952). On the Toxin of Illicium Anisatum. I. The Isolation and Characterization of a Convulsant Principle: Anisatin. Journal of the American Chemical Society. — doi:10.1021/ja01133a002
- Kudo Y, Oka J, Yamada K (1981). Anisatin, a potent GABA antagonist, isolated from Illicium anisatum. Neuroscience Letters. — doi:10.1016/0304-3940(81)90105-1
- Garzo Fernández C, Gómez Pintado P, Barrasa Blanco A, Martínez Arrieta R, Ramírez Fernández R, Ramón Rosa F; Grupo de Trabajo del Anis Estrellado (2002). Cases of neurological symptoms associated with star anise consumption used as a carminative. Anales Españoles de Pediatría. — PubMed PMID: 12392661
- Johanns ES, van der Kolk LE, van Gemert HM, Sijben AE, Peters PW, de Vries I (2002). An epidemic of epileptic seizures after consumption of herbal tea. Nederlands Tijdschrift voor Geneeskunde. — PubMed PMID: 12014242
- Ize-Ludlow D, Ragone S, Bruck IS, Bernstein JN, Duchowny M, Peña BM (2004). Neurotoxicities in infants seen with the consumption of star anise tea. Pediatrics. — PubMed PMID: 15492355
- Rohloff JC, Kent KM, Postich MJ, Becker MW, Chapman HH, Kelly DE (1998). Practical Total Synthesis of the Anti-Influenza Drug GS-4104. The Journal of Organic Chemistry. — doi:10.1021/jo980330q
- Krämer M, Bongaerts J, Bovenberg R, Kremer S, Müller U, Orf S, Wubbolts M, Raeven L (2003). Metabolic engineering for microbial production of shikimic acid. Metabolic Engineering. — PubMed PMID: 14642355
PubMed Topic Searches
- PubMed: Star anise ethnobotany and traditional use
- PubMed: Japanese star anise and anisatin poisoning
- PubMed: Shikimic acid, star anise and oseltamivir