Why the Classics Record Where a Herb Came From

Why the Classics Record Where a Herb Came From

Origin is not a footnote

Open almost any entry in the Bencao Gangmu and you will find, early on, a sentence about where the thing grows. It is easy to read past — a bit of geography before the interesting parts. That reading is wrong. In the classical materia medica the place of origin is doing real work: it is part of how the material is identified and part of how it is graded. Sometimes the place is written into the drug's own name. Sometimes it narrows all the way down to a single well. And very often the authorities simply disagree about which origin is best, and the record keeps every one of their answers.

This is the idea later called dao di (道地) — roughly, "genuine to its place." This article does not try to define that term or settle it. It does something narrower and more checkable: it transcribes the origin passages from four entries, each attributed to the authority who wrote it, and shows what the texts actually did with them.

A note on scope, because it matters here. What follows is about where a material came from and how it was told apart — identity and provenance only. There is no dosage, no preparation, and no claim about what any of these substances do. (For a case where one name spread across roots of very different grade, see wild and cultivated dang gui — that entry's competing origins are worked through there, so we leave it aside below.)

When the place is the name: e jiao

The clearest case is e jiao (阿胶), a gelatin. You do not have to hunt for the provenance, because the provenance is the word. Tao Hongjing (陶弘景) says so in the 【释名】 section in eight characters:

出東阿，故名阿膠。

"It comes out of Dong'e, and so it is called e jiao." The name is a place. The oldest layer, the Mingyi Bielu, locates it the same way and adds what it was made from:

阿膠出東平郡東阿縣，煮牛皮作之。

E jiao "comes from Dong'e county in Dongping commandery, made by boiling ox hide." Li Shizhen then pins the geography down with an outside authority, noting that the Dong'e well sits in what was, in his day, Yanggu county in Yanzhou prefecture, and citing Li Daoyuan's Commentary on the Water Classic (《水經注》) for a well there "as large as a wheel, six or seven zhang deep," whose water was boiled into gelatin as tribute each year.

Then Su Song (苏颂) does the striking thing. He narrows "origin" from a region to one specific water source:

以阿縣城北井水作煮者為真。

Only that "boiled with the well-water north of the wall of E county" counts as genuine (真). Not the county — the well. And he immediately records what follows from a definition that tight:

其井官禁，真膠極難得，貨者多偽。

"That well is under official prohibition; genuine gelatin is extremely hard to get, and most of what is sold is counterfeit." Read those two lines together and you can see the whole logic of origin-as-identity, along with its consequence: the narrower and more prized the source, the more of the market is fake. The texts are not naive about this. They say it plainly.

When the commentators fight over what the place-name means: dai zhe shi

Dai zhe shi (代赭石), a red mineral, also carries a place in its name — Dai. But here the record does something a tidy database would erase: the commentators cannot agree on what the place-name refers to, and the disagreement is preserved verbatim.

The Bielu gives a region: 「代赭生齊國山谷」 — it "grows in the mountain valleys of Qi." Tao Hongjing offers a much more specific, and much stranger, reading:

是代郡城門下赤土也。

It is "the red earth from beneath the city gate of Dai commandery." Su Gong (苏恭) contradicts him head-on:

此石多從代州來，云山中採得，非城門下土也。

"This stone mostly comes from Daizhou; they say it is gathered in the mountains — it is not earth from under the city gate." That is about as flat a correction as the genre gets, and it survives on the page next to the claim it corrects. Su Gong then ranks a third source above both of the ones already named: material dug from level ground north of the river in Lingzhou, he says, is 「大勝齊、代所出者。」 — "far surpassing what comes out of Qi or Dai." Li Shizhen, centuries later, widens it back out:

赭石處處山中有之，以西北出者為良。

"Hematite is found in mountains everywhere; the material from the northwest is good." So four voices, four different answers, including an argument about whether the name refers to a mountain or a city gate — and no verdict. The record's job here is not to tell you which one was right. It is to make sure you can see that the question was open, and who took which side.

When the best origin moves: huang lian

Huang lian (黄连, coptis) is the entry that states the whole problem out loud. Line up its 集解 chronologically and you watch the preferred origin travel across the map for five hundred years.

The Bielu starts it in the west: 「黃連生巫陽川谷及蜀郡、太山之陽」 — Wuyang's river valleys, Shu commandery, and the south face of Mount Tai. Tao Hongjing then rates the western material down:

今西間者色淺而虛，不及東陽、新安諸縣最勝。

The western sort "is pale in colour and hollow, not equal to the counties of Dongyang and Xin'an, which are best." Han Baosheng (韩保昇) names somewhere else again: 「今秦地及杭州、柳州者佳。」 — "today the material from Qin, Hangzhou and Liuzhou is fine." Su Song ranks a fourth, and grades the runners-up while he is at it:

而以宣城九節堅重相擊有聲者為勝，施、黔者次之，東陽、歙州、處州者又次之。

Best is Xuancheng's "nine-jointed, hard and heavy, ringing when struck together"; Shi and Qian come second; Dongyang, Shezhou and Chuzhou third. Su Gong had earlier described the Shu material as 「粗大」 — coarse and large — against the Jiangdong sort whose root is "jointed like a string of pearls" (節如連珠), and rated Lizhou's above both. Note that the grading criteria here are physical: the joints, the weight, the ring of one root struck against another.

And then Li Shizhen writes the sentence this whole article exists for:

惟取蜀郡黃肥而堅者為善。唐時以澧州者為勝。今雖吳、蜀皆有，惟以雅州、眉州者為良。藥物之興廢不同如此。

At the end of the Han, he says, the Li Dangzhi Bencao took only Shu commandery's "yellow, fat and firm" material as good. In the Tang, Lizhou's was held best. Today, though both Wu and Shu have it, only Yazhou's and Meizhou's is considered fine. Then the closing clause: 「藥物之興廢不同如此。」 — "the rising and falling of medicinal materials differs like this."

That is a sixteenth-century author telling you, in his own voice, that the "authentic origin" of a herb is not a fixed fact but a thing with a history — one that rises and falls, and that he can only report as of his own day. It is exactly the kind of statement that vanishes the moment someone flattens an entry into a single field reading "origin: Sichuan."

When origin is baked into the drug's names: shu jiao

Shu jiao (蜀椒) makes the point almost typographically. Its 【释名】 section is a list of synonyms that are mostly just place names with "pepper" attached: 巴椒 (Ba pepper, from the Bielu), 汉椒 (Han pepper), 川椒 (Sichuan pepper, the Gangmu's own). The material is named, over and over, for where it came from.

The Bielu locates it: 「蜀椒生武都山谷及巴郡。」 — the mountain valleys of Wudu, and Ba commandery. Tao Hongjing describes the northern-Shu cultivated material by its physical characters:

蜀郡北部人家種之。皮肉厚，腹裏白，氣味濃。

Households in the northern part of Shu commandery grow it; "the skin and flesh are thick, the inside white, the aroma and flavour rich." He contrasts this with material from Jiangyang, Jinkang and Jianping, which is 「細赤，辛而不香」 — "fine and red, pungent but not fragrant." Su Gong then breaks ranks and prefers somewhere else entirely: 「今出金州西域者最佳。」 — "today what comes from Jinzhou and the western regions is best." Su Song circles back toward Shu, noting that material from the Jiang-Huai region and the north is much the same in stem and leaf 「但不及蜀中者良而皮厚、里白、味烈也。」

Same pattern once more: a base location, a set of physical tells, and authorities ranking regions against each other without ever converging.

What the four cases have in common

Put side by side, the four entries show origin doing four distinguishable jobs:

1. Origin as name — e jiao, dai zhe shi and shu jiao all carry their provenance in the word itself, so that using the name at all is a provenance claim;
2. Origin as the definition of genuine — Su Song's e jiao well, where 真 ("genuine") is defined by a single water source, with counterfeiting recorded as the direct consequence;
3. Origin as a contested reading — dai zhe shi, where the commentators disagree about what the place-name even denotes, and the correction sits beside the claim;
4. Origin as a moving target — huang lian, where the preferred source demonstrably changes from Han to Tang to Ming, and the author says so.

None of that survives compression. Fold these entries into one clean line each — "e jiao: from Shandong," "huang lian: from Sichuan" — and you keep a fact while destroying the record: the well, the prohibition, the counterfeits, the argument about the city gate, the five centuries of drift, and Li Shizhen's own warning that it drifts.

How MateriaTrace handles it

Our approach is the unglamorous one. Each origin line is kept verbatim, attributed to the authority who wrote it (《別錄》, 弘景, 保昇, 恭, 頌, 時珍), placed in its section of the Bencao Gangmu, and traced back to the source so you can open the book and confirm it. Where the authorities disagree about which origin is best — and in these four entries they always do — we keep every ranking side by side rather than choosing a winner or silently averaging them into one. Where a text records that the answer changed over time, that sentence is kept too, because it is the most honest thing in the entry.

Worth stating plainly: this is a record of how classical texts described the origins and grades of certain materials, kept for study and traceability. It is not advice on buying, sourcing, identifying for purchase, using, or dosing any herb, and it makes no claim about any substance's effects. MateriaTrace records what sources say and makes no treatment recommendation; any question about a herb belongs with a qualified professional.

Why it matters that the argument survives

A reader who wants to know where huang lian "really" comes from will not find a verdict in the Bencao Gangmu. They will find something more useful: five named people, across five centuries, each saying where the good material came from in their own time, and one of them pointing out that this is a moving thing. That is not a defect in the record. It is the record working — provenance kept as a claim with an author and a date attached, rather than as an anonymous fact that cannot be checked or dated at all.

See the entries themselves: e jiao, dai zhe shi, huang lian, and shu jiao. For the same source-tracing pointed at which plant rather than which place, read verifying real ginseng against the classics and the herbs most often faked. And to see how you would check a claim like these yourself, step by step, read how to fact-check an AI claim about a Chinese herb.

If this "every line clicks back to its own source" approach is useful to you, see what else is in MateriaTrace on the homepage.

MateriaTrace is a source-traceable, bilingual knowledge graph of Chinese medicine. It records how classical and modern sources describe each entry — each field traceable to its primary source, contradictions kept side by side — and makes no treatment or usage recommendations. Every line quoted in this article can be traced to its classical source.