The tradition
典籍
Chinese medicine is a written tradition. Its working ideas were set down in books that survive, with dates, authors, editions and commentaries, and each later book argues with the earlier ones by name. Four of those books do most of the work in a reading, and this section explains what each of them actually says.
Not folklore
A great deal of what is written in English about Chinese medicine begins with the word ancient and stops there, as though age were the whole of the argument. It is not. What matters about this tradition is duller and more useful than antiquity: it wrote things down, it kept what it wrote, it printed corrected editions under state supervision, and it argued with itself in public, in books, for two thousand years.
That is an ordinary description of a scholarly field. It is also exactly what folklore is not. Folklore has no bibliography. Nobody can tell you which edition of a folk remedy is the good one, or who revised it in 1057, or which twelfth-century physician thought it was wrong and said so in print. In Chinese medicine you can answer all three questions, because the paperwork survives.
The scale of that paperwork is easy to underestimate. The standard modern collection of pre-modern Chinese medical texts runs to more than a thousand surviving titles. Most were written by working physicians for other working physicians, which is why so many of them are structured as case notes, corrections and complaints about colleagues rather than as inspirational prose.
The finest physicians attend to what is not yet illness; they do not wait until disorder has taken shape and then act on it.
How it travelled
The earliest medical writing we can hold was written on bamboo slips and silk. Fifteen medical manuscripts came out of tomb three at Mawangdui, sealed in 168 BCE and opened in 1973 — recipe collections, a chart of vessels, exercises drawn in colour. They are older than the Neijing in its finished form, and they show the same ideas mid-argument, before anyone had settled them.
From the Han onwards the texts were copied by hand, which means they drifted. The fix, when it came, was bureaucratic. In 1057 the Song court set up an office whose entire purpose was to collate the medical classics against one another and issue corrected printed editions. Woodblock printing did the rest. A physician in Fujian and a physician in Shanxi could, for the first time, argue about the same sentence.
Almost everything read today descends from those Song editions. It is why a clause number in the Shanghan Lun means something specific, why commentaries can be stacked on a single line for nine hundred years, and why the four books on this page are still assigned as coursework rather than kept behind glass.
The bibliography
Nineteen entries. Each one is a text you can find in a university library, an office of state, or a printed edition, and each one changed what the physicians after it were taught. Nothing here is an oral tradition recovered later.
c. 200 BCE
Two parts — Suwen, Plain Questions, and Lingshu, the Divine Pivot — of eighty-one chapters each, written as a dialogue between a ruler and his physicians. Qi, yin and yang, the organ networks and the seasons are set down here as a working system.
c. 200 CE
The oldest surviving Chinese herbal: 365 entries, one for each day of the year, sorted into three grades by how freely they could be used.
c. 200 CE
Composed after an epidemic took two thirds of his clan. Later divided in two; the Shanghan Lun half carries 398 clauses and 112 formulas, and invents the method of matching a named pattern to a named response.
259
Huangfu Mi collates three older texts into the first ordered point manual: 349 points, each named, located and cross-referenced.
c. 340
A pocket book of what to do when no physician is available. Sixteen centuries later a single line in it — steep 青蒿 in cold water and wring out the juice — gave Tu Youyou the extraction method behind her 2015 Nobel Prize.
610
Chao Yuanfang’s Sui-court compilation runs to 1,739 entries and offers not one formula. It is a book purely about describing what is in front of you — the ancestor of everything a reading does.
652 · 682
Roughly 5,300 formulas across the pair, opened by an essay on how a physician should conduct himself. It is the first sustained piece of medical ethics written in Chinese.
1027
Wang Weiyi has hollow bronze men made, pierced at 354 points and sealed under wax, so that a candidate’s accuracy could be examined against an object rather than an opinion.
1057
Scholars collate the old texts against one another and print the results. This is why the editions read today largely agree — the tradition has been under continuous editorial revision since the eleventh century.
1249
One of four schools contending in print during the Jin and Yuan. They disagreed with each other publicly and in writing, which is the behaviour of a scholarly field rather than a folk practice.
1596
Li Shizhen’s survey appears in Nanjing three years after his death: 1,892 entries, 11,096 formulas, 1,109 illustrations, sixteen divisions, sixty categories, twenty-seven years of work.
1601
The Great Compendium pulls a millennium of scattered point texts into one book, and it is still the backbone of how the points are taught.
1642
He proposes that epidemic illness enters through something carried in the air rather than through cold or wind — a break with the Shanghan model, made from watching an outbreak.
1798
Wu Jutong describes four levels through which a warm illness moves inward. It is the last great structural argument added to the classical literature.
1868
In The Variation of Animals and Plants under Domestication he draws on a Chinese work for the breeding of goldfish and fowl. Scholars identify it as the Bencao Gangmu.
1956
Beijing, Shanghai, Guangzhou and Chengdu. The five-year curriculum they set — classics, then modern anatomy and physiology, then supervised practice — is the one our family trained under.
1987
One consulting room, one wall of drawers, in a city where a family had practised for three generations already. Everything on this site is written from the notes taken in it.
2011
The Huangdi Neijing and the Bencao Gangmu are inscribed on the Memory of the World Register, which exists for documentary heritage.
2019
ICD-11 includes a supplementary chapter of traditional medicine pattern terms. It grades none of them as effective; it simply agrees that the vocabulary is stable enough to be counted.
Why it matters here
Three things, and they are practical rather than romantic.
The first is vocabulary. When your report says that the picture you describe reads, in Chinese medicine, as dampness, that word is not a metaphor somebody chose this morning. It has a defined sense that goes back to the Neijing, a set of signs attached to it, and a literature arguing about its edges. You can look it up. So can anyone you show the report to.
The second is method. The Shanghan Lun is built out of 398 short clauses, each of which names a set of signs and then names what follows from them. That structure — this constellation, therefore this description — is the whole procedure of 辨证, bianzheng, and a reading is a small instance of it. We are not inventing a way of organising your answers; we are using an old and rather rigid one.
The third is restraint. The tradition is unusually clear about what it cannot see. Its own texts insist that the four examinations must be taken together, and we can only take three of them at this distance. Reading the classics closely makes you more careful about the gaps, not less. That is why the list of what we will not do is longer than the list of what we will.
Tradition
Three canonical texts and one organising scheme. Each page is written to be read on its own, and each says plainly what it has to do with the report you would receive.
Questions
The oldest text still in continuous use, the Huangdi Neijing, was compiled around 200 BCE from material that was already old. Excavated manuscripts from Mawangdui, sealed in 168 BCE, are older still. So the written record runs a little over two thousand years, and the practice behind it is older than its books.
It is written down, and it has been edited in public for a thousand years. The Song court opened a bureau in 1057 whose whole job was collating and printing corrected editions of the medical classics. That is why physicians in Chengdu and Hangzhou can quote the same clause and mean the same thing.
Yes. The Neijing, the Shanghan Lun, the Jingui Yaolue and the Wenbing texts are the four required classics in a five-year Chinese medicine degree, read in the original. The physicians who write our reports sat those examinations and can cite the clause a reading rests on.
Mostly two. The Neijing supplies the vocabulary — qi, yin and yang, the organ networks, the seasons. The Shanghan Lun supplies the method of naming a pattern from a set of signs. The Bencao Gangmu supplies the food notes. The five phases sit underneath all three as the filing system.
No, and we do not argue that it does. Age tells you an idea has been used, argued over and revised by a lot of careful people, which is a reason to take it seriously and not a proof of anything. Our reports are educational, and they say so on every page.
The reading
A reading applies the vocabulary on these four pages to thirty answers and one photograph, and returns it to you in writing within 48 hours.