Bencao Gangmu

Bencao Gangmu (known in English as the Compendium of Materia Medica) is an encyclopedic work of medicine and natural history compiled by Li Shizhen during the Ming dynasty in China. Its first draft was completed in 1578, and the book was printed in Nanjing in 1596, after the author's death. It catalogues the materia medica known at the time, including plants, animals and minerals believed to have medicinal properties, and it became a central reference for traditional Chinese medicine.

The work is widely regarded as the most complete medical book in the history of traditional Chinese medicine. It describes roughly 1,892 substances and gives about 11,000 prescriptions, drawing on hundreds of earlier texts, literary and historical sources, and the author's own fieldwork. Its scheme of classification, its standard entry format and its critical correction of older sources are central to its significance, and in 2011 it was inscribed in the UNESCO Memory of the World Register.

The entry moves from the author and the compilation process to the book's structure, entries, materials, prescriptions and sources, and then covers its errors, illustrations, editions, reception, translation and modern assessment. Where sources differ on figures and dates, the entry notes the discrepancy.

Author and medical background

Li Shizhen was born in 1518 in Qizhou Town, Qichun County, Hubei, during the Ming dynasty. He was born into a family of doctors and began studying medicine at an early age under his father's guidance. His medical training therefore came from inside the household rather than from a formal institution. Accounts of his family differ in emphasis. One describes his grandfather as an itinerant physician of modest social standing, and his father as a practitioner and scholar who wrote several influential books and hoped his son would seek a government post. His grandfather had been a doctor who traveled the countryside and was considered relatively low on the social scale of the time. His father was a traditional physician and scholar who had written several influential books. He tried to move up in society and encouraged his son to seek a government position.

Li's early career followed the path of a would-be official before it turned to medicine. He passed an early county-level stage, but failed at higher levels multiple times, and redirected his life into medicine. The dates of his first medical posts vary between sources. One account says that his reputation as a physician soon reached the prince of Chu, who in 1543 entrusted him with medical and administrative responsibilities at his court in Wuchang. Another places the prince episode in the early 1550s, when he is credited with saving the prince's son during a sudden collapse. From 1544 to 1549 Li practiced at the Imperial Academy of Medicine (Taiyi Yuan) in Beijing. A later account instead dates his recommendation to serve as an imperial physician in the capital to the mid-1550s. The sources therefore disagree by roughly a decade on this part of his career, and the chronology should be read as uncertain. After that, he returned to his native village, where he worked as a doctor until his death in 1593.

The bencao tradition that Li inherited was already large and well developed. The Shennong Bencao Jing, traditionally attributed to the Han period, is the earliest surviving canonical bencao text, and the Tang-era Xinxiu Bencao (659) expanded the genre with official state authority. Li's immediate models were Song-period works. He compiled his large book on materia medica according to the pattern of the Song period pharmacopoeia Zhenglei bencao. That earlier text was itself important for the genre's expectations: the Jingshi Zhenglei Beiji Bencao, which unlike many other books, had formulas and recipes for most of the entries. Kou Zongshi's Bencao Yan Yi (1116) belongs to the same line of Song compilations, and its existence shows that by the twelfth century, scholars were already expected to reconcile drug descriptions with clinical formulas and critical commentary. Li's book built on this foundation. He modelled his work on a Song dynasty compilation, especially its use of non-medical texts.

What Li changed was the scale and the method. He kept the expectation that a drug entry should connect to practical prescribing, but he extended the evidence base well beyond earlier pharmacopoeias. Where the inherited tradition drew mainly on medical texts and established authorities, Li also drew on literary and historical material. The details of that departure, and how later readers judged it, belong to the sections on sources and classification, but the basic move was already present in his approach to the older literature.

Popular writing often calls Li the "Prince of Pharmacists." This label is applied in popular accounts to a scholar from Hubei who lived between 1518 and 1593. The title is a later honorific rather than a contemporary designation and is best treated as a sign of Li's reputation. Similarly, the intellectual setting of the late Ming, which some biographies describe as shaped by Neo-Confucian thought, provides background for his classificatory project but does not by itself explain his methods. He lived during the Ming Dynasty and was influenced by the Neo-Confucian beliefs of the time.

Compilation and publication history

Li Shizhen's project took roughly three decades to assemble. One account counts the manuscript as revised three times The compilation process consumed almost three decades, and the manuscript was revised three times. Another dates the revisions specifically: the first draft was finished in 1578, when Li was about sixty, and it was revised again in 1580 and 1587 The resulting Bencao Gangmu was completed in 1578, when Li Shizhen was 60 years old, though he revised it in 1580, and again in 1587. The "three revisions" figure therefore appears to count the 1578 text together with the two later revisions, which is worth keeping in mind when reading the sources.

The 1580 revision also marks the start of the prefatory apparatus. Li asked the scholar Wang Shizhen to write a preface, and Wang agreed As part of the 1580 revision, he approached a scholar, Wang Shizhen, to write the preface; Wang, who agreed. The preface is dated 1590, the year Wang died, and the front matter also includes an author's note of 1590 and a memorial to the throne dated 1596 This English translation includes all Li's original 52 volumes plus a translator's note, three prefaces (one that appeared in the first printing and two that appeared in the reprinting of 1603), a memorial to the throne (1596), and an author's note (1590).

Publication was carried out largely by Li's family. Li's son Li Jianyuan presented a copy to the Ming court, where it was seen but attracted little notice Li Jianyuan presented a copy to the Ming dynasty emperor, who saw it but did not pay much attention. Printing was then arranged in Nanjing, using the workshop of Hu Chenglong Li arranged to have the book printed in Nanjing, relying on the resources of the printer, Hu Chenglong. The text was first printed in 1596 in Jinling, the old name of Nanjing The Bencao gangmu was first printed in 1596 by Hu Chenglong 胡承龍 in Jinling 金陵 (modern Nanjing 南京, Jiangsu). The family-edited Jinling copy is generally regarded as the original edition The Jinling copy is the earliest version, drawn and proofread by the Li family and considered the original edition. The book appeared three years after its author's death, and the carving of the blocks had begun in 1593 He finished a draft of the text in 1578, the printer began to carve the blocks in 1593, but it was not published until 1596, three years after Li died.

The sources disagree on the date of first publication. Most give 1596 for the first printing, but some describe 1593 as the year of publication. A Library of Congress record states that the text was not published until the 21st year of the Wanli reign, which corresponds to 1593 it was not published until the 21st year (1593) of the Wanli reign of the Ming dynasty, after his death, when it was printed by Hu Chenglong in Jinling. A review of the 2003 English translation likewise says the work was published originally in 1593 and reprinted in China numerous times since. These accounts treat 1593 as the year of publication, while the blocks were being carved and the printing was being completed, rather than the year of the finished book. The 1596 date is the one most often used for the first edition.

Early printings were not always high in quality. One account says that Hu's Nanjing work was done rather carelessly, and attained only limited distribution. The first reprint followed in 1603, in Nanchang, Jiangxi, In the 31st year (1603) of the Wanli reign, the work was reprinted in Jiangxi. Later editions were based on it and spread across the country The later editions were all based on the 1603 edition and spread widely throughout the country.

Structure and classification

Library catalogues describe the Ming printed text as 52 juan of text plus two separate juan of illustrations, so some totals for the book's length depend on whether the illustration volumes are counted. Printed during the Wanli period, Ben cao gang mu is a work on an encyclopedic scale, in 52 juan of text with two juan of illustrations, in 25 volumes.

The opening juan work as a general introduction. In one account they set out the seven basic methods of application and the ten preparation methods, along with the concepts used to explain drug action, such as qi, taste, yin and yang, and rising and sinking. They also assign ingredients roles in a formula as lord, minister, and assistant. The introduction also includes an overview over pharmacopoeias of earlier times and mentions 41 pharmaceutical books. One review places the drug material differently, describing juan 3 and 4 as two volumes (3 and 4) describing "drugs for treating a hundred diseases", which suggests that the front matter is not identical across accounts.

The main body, juan 5 to 52, is organized into sixteen large classes. The headings are commonly listed as water, fire, earth, metals and stones, herbs, cereals, vegetables, fruits, woods, fabrics and utensils, insects, scaly animals, shells, fowls, animals, and humans. Some of these classes occupy several juan. Metals, stones and minerals fill juan 8 to 11, and herbs fill juan 12 to 21, so the herbal material alone accounts for a large share of the text (waters (Vol. 5); fires (Vol. 6; earth Vol. 7); metals, stones, and minerals (Vols. 8-11); herbs (Vols. 12-21)).

The counting itself is not uniform. One source gives sixteen large pharmaceutical classes according to their natures and sources, with drugs under the same large class further divided into sixty sub-classes according to their similarities. Another reports sixty-two sub-classes. The total for the top level also depends on how headings are tallied: the list above reaches sixteen only if cereals, vegetables and fruits are counted separately and fabrics and utensils are treated as a single class. The title itself offers a further terminological point. The word gang denotes the main heading and mu a subordinate one, though one account reads the division of species into groups called gang and smaller categories called mu, while another glosses mu as the synonyms of each name.

The logic of the arrangement runs in several directions at once. Broadly, it moves from the most basic natural elements toward living things and humans. Li's own preface describes placing waters and fires first, then soils, then worms, scaly animals and crustaceans, fowl and quadrupeds, with humans last. Within each class, the order is also graded: inside the sub-classes the materia medica are arranged from the smaller to the larger, the less valuable to the expensive drugs, and the simple to the complex. The human-substance chapter closes the book; the human-drug chapter contains 37 entries.

Western-language accounts commonly describe this scheme as more systematic than its predecessors. Li is said to have set out to compile correct information with a logical system of organization. The grouping was not purely by natural origin, however. One account notes that while he categorised various medicinal substances according to the disease they allegedly cured, his taxonomy of species used the gang and mu division. The terms also later passed into biological classification, where the same characters were used for 'class' and 'order', respectively.

Entry format

The Bencao Gangmu handles each substance under a repeated set of headings, and this regularity is a large part of why the book works as a reference rather than a loose anthology. Accounts describe each drug's description as running through eight segments, beginning with the names by which it is known.The description is systematically divided into eight parts, beginning with an analysis of the different names of drugs The first heading, shiming (釋名), lists the synonyms a substance has accumulated and explains where those names come from. The second, jijie (集解), gathers statements about origin, habitat, general appearance, cultivation and harvesting.集解，集录诸家对该药产地、形态、栽培、采集等的论述 The remaining segments cover processing, nature and flavor, indications, Li's own elaborations, and corrections of older errors, followed by an appendix of prescriptions (fufang, 附方).修治，介绍该药的炮制法和保存法；气味，介绍该药的药性；主治，列举该药所能治的主要病症

The segments are not counted the same way everywhere. English-language summaries list doubtful points (bianyi, 辨疑) as a separate heading, while a Chinese listing of the entry headings omits it and names only the corrections heading.discussions of doubts (bianyi 辨疑) and corrections of errors occurring in older texts (zhengwu 正誤) 每種藥物分列釋名（確定名稱）、集解（敘產地）、正誤（更正過去文獻的錯誤）、修治（炮製方法）、氣味、主治、發明（前三項指分析藥物的功能）、附方（收集民間流傳的藥方）等項 The difference is largely one of how sections are grouped, not of content, but the count given in any description depends on which grouping is used.

Most of these segments draw heavily on earlier writers, and Li's own commentary is layered on top of that material.All these parts quote extensively from older sources and are enriched by Li Shizhen's own remarks. The faming (發明) segment is where his interpretive comments usually appear, and the format therefore separates inherited statements from his analysis, even though the two sit side by side on the same page.

The format also records Li's critical judgments. He reviewed the provenance of most of the materials and corrected earlier identifications, so the headings on doubts and corrections are not decorative.Li Shi-Zhen comprehensively checked and corrected the source of most of the medicinal materials recorded in the Compendium of Materia Medica Specific corrections of this kind are discussed in the errors section of the entry.

Prescriptions are handled in the same systematic way. Accounts disagree on the number. One source gives 8,161 recipes and treatment methods within the entries' appendices.The entry of each drug is enriched by an appendix on treatment (fufang 附方), in which Li Shizhen explains 8,161 different recipes and treatment methods. Other sources give a larger total for the whole book: identified 1,892 medicinal drugs and listed 11,096 prescriptions. The difference probably reflects how prescriptions are counted, and the entry should not treat either figure as settled without noting the method.

Finally, the consistency of the layout across roughly 1,892 entries is a central reason for the book's practical value. Each entry follows a predictable path from name to origin to use, which allowed readers to locate information quickly.Under each entry, the name explanation, commentary, argument, processing, taste and property, indications, development and formula have been documented. The format's durability is also reflected in later use: the book is still used today for consultation and reference.

Materials catalogued

Reference works usually put the total at 1,892 items, and they treat 374 of these as additions beyond earlier pharmacopoeias. The Bencao gangmu describes 1,892 different pharmaceutical objects, which is 374 more than in earlier pharmacopoeias. The wording of the second figure varies. One account says Li himself added the 374 entries, while another says they had not appeared in other works. The two phrasings describe the same number from different angles, and the difference matters for whether these entries reflect Li's own fieldwork or his reading.

The category totals are less settled. Encyclopedia summaries commonly assign about 1,094 entries to herbs, 444 to animal-derived substances and 275 to minerals. One source gives 276 drugs made of minerals, and 444 out of parts of animals, so the mineral figure shifts by one between sources. The three commonly cited groups sum to 1,813, which leaves roughly 79 of the 1,892 entries outside them; the sources do not itemize that remainder. A percentage breakdown gives a different picture: roughly 58% of the materials are plant based, 23% animal based, and 15% mineral based, with the other 4% consisting of outliers such as fire and water. Applied to 1,892, those percentages yield about 1,097, 435 and 284 items, which do not match the absolute counts. The discrepancy suggests that the figures come from different tallies, possibly with different treatments of multi-source or borderline items, and readers should treat the exact totals as approximate.

The book's scope reaches well beyond plants and minerals. Li's sixteen sections include Water, Fire, Earth, Metals and Minerals, Herbs, Grains, Vegetables, Fruits, Trees, Utensils and Garments, Insects, Scaled Creatures, Shellfish, Birds, Beasts, and Humans. The order runs from the elemental toward the complex: Li's own prefatory remarks move from the most elemental (water) to the most complex (humans). The water section is itself subdivided, with entries for waters from heaven and waters from earth. Human-derived materials form a distinct group; one overview notes that the text covers drugs made from bones, fingernail, hairs, dandruff, earwax, impurities on the teeth, feces, urine, sweat, and organs, and that most of them are no longer in use. The inclusiveness has intellectual roots. Li's approach drew on the principle of Chen Cangqi's Bencao shiyi "Collected addenda to the pharmacopoeias" not to leave out any single information, and on the Neo-Confucian pursuit of investigating things. The additions were also geographically distinct: they mainly originate in regions that were formerly not part of the Chinese empire, especially the northwest and the southwest.

The range of substances runs from familiar medicinals to unusual ones. Ginseng is among the best-known entries, and it receives unusually extended treatment: one comparison notes that the ginseng entry in the Shennong Bencao Jing comprises only 47 words, while the entry in the Compendium of materia medica (Li, 1578) contains nearly 10,000 words. Marine materials form another substantial strand. One review states that the classics, including this one, have detailed more than 110 marine herbs and thousands of marine herbal formulas. At the unusual end, the human-derived and elemental entries show how far the category of medicine extended in Li's scheme. Placing these items in one framework, rather than in separate lists, is part of what distinguishes the book's organization from that of earlier collections.

Prescriptions and therapeutic methods

The Compendium is as much a formulary as a pharmacopoeia. Reference summaries put its prescription total at roughly 11,000, along with directions for preparing them. Completed in 1578, this influential materia medica details 1,892 drugs and provides instructions for over 11,000 prescriptions. Counts differ on how these are distributed. One encyclopedia entry gives 11,096 side prescriptions, of which 8,160 are compiled in the text. The remainder are not explained in that source, so the exact figure should be read as approximate.

The link between prescriptions and drug entries is structural. Each drug entry ends with an appended section on applications. The description is systematically divided into eight parts, beginning with an analysis of the different names of drugs, collected explanations about the places of origin, general appearance and collecting methods, and discussions of doubts... An appendix chapter for each drug is commented regarding concrete treatment methods. In practice, this means a substance is identified, assessed and then linked to specific treatments in the same place. The prescriptions are typically brief. For example, the mugwort chapter includes formulas for wind-cold, sore throat, toothache, worm infestations and several kinds of bleeding, with the bleeding formulas standing out in particular.

Several techniques are credited to Li in general reference works. Those works state that Li was one of the first to recognize gallstones, use ice to bring down a fever, and to use steam and fumigants to prevent the spread of infection. These are historical characterizations of the text and of Li's practice. Whether they were novel in their own period, and how they should be assessed clinically, are questions for historians and medical scholars rather than settled facts. Their presence does show that the book's scope reached well beyond drug identification into diagnosis, symptom management and infection control.

Li also gave weight to prevention. He emphasized preventative care and listed over 500 treatments for maintaining good health and strengthening the body, of which he is said to have devised 50 himself. This material belongs to the broader tradition of yangsheng, or health cultivation, and it sits alongside the pharmacological entries rather than forming a separate volume. The attribution of the 50 original practices to Li comes from a secondary account and is not confirmed by the primary text in the sources reviewed here.

Sources and fieldwork

Li Shizhen's sourcing combined a very large body of printed texts with direct inquiry outside the library. Most accounts put the number of works he consulted at more than 800, but they describe that pool differently. Some call them simply books, while others call them medical texts or medical books. A breakdown given in one source lists 41 kinds of ancient herbal monographs and 271 kinds of medical books, together with historical and philosophical writings such as the Zuo's Biography of the Spring and Autumn, Guo Yu, Shi Ji, Han Shu, Hou Han Shu and a comprehensive materia medica compendium. The 800 figure is therefore a total, and its composition is not uniform across sources.

Li's entries drew on more than the bencao tradition. He relied on hundreds of earlier works in the bencao medical tradition, but from literary and historical texts, and his model, the Song-era Jingshi Zhenglei Beiji Bencao by Tang Shenwei, was itself valued for its non-medical material. Later accounts stress that Li drew a lot of information from non-medical writings, he references poetry, and religious and historical texts. This breadth is the feature most often cited as departing from the narrower textual habits of earlier compilers.

Li's fieldwork is documented in biographies and modern studies, though the details come mainly from later secondary accounts. He travelled to places including Hunan, Jiangsu, Jianshi, and Anhui, where he could observe local herb markets and collect specimens. He also sought out people who worked with materials every day. Accounts describe interviews with farmers, fishermen, hunters, and local healers, along with the collection of specimens and recipes. Some commentators argue that this approach treated local knowledge as a legitimate input and then tried to fit it into a repeatable entry format. Whether that was Li's explicit intention is a matter of interpretation rather than something his own text states directly.

Li's use of poems and tales of the strange raises a separate question. Li reasoned that a poem might have better value than a medical work and that a tale of the strange could illustrate a drug's effects. Such material was not evidence in the sense that Ming physicians or modern pharmacologists would accept. Read on its own terms, however, the choice shows that Li treated narrative and verse as sources of information about how substances behaved and were perceived, not only as decoration. Tales of the strange (zhiguai) and poetry thus functioned as a channel for observed or reported effects, alongside the medical literature.

It is useful to separate Li's stated aims from later judgments of his method. His stated aim, as reflected in his criticism of earlier texts, was to correct errors in existing works; he corrected some mistakes in previous Bencaos. Readers judged the method differently over time. The Ming court paid little attention to the book, and it was ignored for some time after publication. Its later reputation rested partly on the breadth of its sources, but the same breadth made it harder to assess which claims rested on checked observation and which on inherited or literary report. Modern assessments therefore tend to treat the incorporation of literary material as a methodological feature that explains both the book's range and the uneven reliability of some entries, rather than as a simple virtue or defect.

Scholars also note that travel gave Li a second source of data. When written sources disagreed, fieldwork offered an independent check, particularly on names and practices that had not yet been fixed in writing. Li's handling of drifting names and their identification with observed materials is discussed further in the sections on errors and classification.

Errors and limitations

The Bencao Gangmu contains statements that were later shown to be false. Commentators usually trace these to the limits of scientific and technical knowledge in Li's time. The best-known example is the claim that Li's work contained errors and mistakes due to his limited scientific knowledge at the time. For example, Li claimed that all otters were male. Other entries have similar problems: the text records that quicksilver (mercury) and lead were not toxic, and it gives an unreliable account of the Moupin langur, which it describes with backwards feet and says could be caught by drawing its upper lip over its eyes. These attributions to Li's era are the view of later commentators; the book itself does not state them.

Corrections and retained beliefs

Li did correct many older errors, and his entries show this critical method. He recorded that misinformation in other works, such as the assertion that consumption of mercury could lead to immortality, was false. He also resolved identification problems: the herb penglei was shown to be five different types of rubus, and Polygonatum odoratum (weirui) and Clematis apiifolia (nüwei) are two different plants.

The corrections were not complete, however. The same book that rejects mercury immortality still records the claim that mercury and lead were non-toxic, so inherited assertions and critical judgments sit side by side. A reader therefore has to check each entry's claims separately rather than assume that the author's method guaranteed accuracy throughout.

Uneven reliability of sources

Accuracy also varies with the kind of evidence Li used. He drew on a poem and on a tale of the strange to illustrate drug effects. Where a description rests on literary or folkloric sources rather than direct observation, its reliability depends on that source. The otter and langur entries are examples of how such material could produce confident but incorrect statements.

Transmission introduced further errors. A study of the Jinling (Nanjing) edition counted 376 misprints appearing 615 times, mostly from missing or wrong character strokes. These mistakes came from carving and proofreading rather than from Li, so they should be distinguished from errors in the original content.

Identifying old drug names

Identifying an older drug name with a specific modern species or substance was a recurring difficulty, not a set of isolated mistakes. Li himself wrote that drug names were mixed, which made it hard to determine a drug's shape and growth. Earlier authorities disagreed sharply: one account of a plant is contradicted by a Song-era critic who blamed Tao Hongjing for not knowing Yuanzhi at all. For a drug called "dog ridge," some say it looks like a Chinese medicine, some say it looks like pulling Smilax, and some say it looks like Guanzhong. Because the same names and descriptions recur across centuries of texts, the problem of matching names to organisms in the book is structural. It affects the classification discussed elsewhere in this entry, and it is one reason modern researchers examine the book's identifications so closely.

Illustrations and editions

Sources do not agree on the number of illustrations. The figure most often given is 1,109, which appears in several general descriptions of the text, including one that places the plates at the front of most editions. The book is enriched with 1,109 illustrations of minerals, plants and other pharmaceutical objects. These illustrations are in most editions of the Bencao gangmu positioned at the beginning of the book. Other accounts give 1,160. One ties that number to the front matter, noting that the opening table of contents lists entries, including 1,160 hand drawn diagrams and illustrations. A dictionary entry uses the looser phrase more than 1100 drawings. The sources do not explain the gap, so the entry reports both figures rather than choosing one.

Attribution is also uneven. The Wellcome Collection credits the first-edition plates to Li Shizhen's son, describing them as credited to his son Li Jianyuan. Another description separates the roles: the illustrations were compiled by Li Shizhen's son Li Jianzhong and painted by Li Jianyuan and Li Jianmu. A third account simply names Li Jianyuan as the draughtsman, noting that he worked with his father during the long project. The sources do not reconcile these roles, and the entry does not assign specific work to any one person.

The history of the illustrations is contested. A popular article states that the first edition of the Bencao gangmu contains no illustrations, which conflicts with the Wellcome record for the first edition. The same article describes the early plates as not high quality, but sufficed to identify plants. This assessment of early woodcuts is a modest one, and it can be contrasted with the later revisions of the plates in other countries.

Later printers reworked the plates. A Japanese woodblock edition of 1653 kept the text of the Jiangxi-line edition but the illustrations are replaced with those from the Hangzhou edition. More broadly, the Japanese woodblock editions of Ben Cao Gang Mu from the mid-to-late Edo period prioritized reproducing the Hangzhou edition's illustrations. This shows that the plates were treated as separable from the text and could be replaced in new editions, a practice that also shaped the later Chinese printing tradition. Further editions were then published in 1603, 1606, 1640, and then in many editions, with increasing numbers of illustrations, down to the 21st century.

The Chinese editions form several lineages. The Jiangxi-line edition of 1603 carried a Zhang Dingsi, 1543-1603 reprint preface and reused the original 1590 preface by Wang Shizhen. Among the Qing-period descendants, the edition printed by Zhang Shaotang at the Weigu Studio in Nanjing in the 11th year of the Guangxu era of the Qing dynasty (1885) is the main lineage of the Zhang family. Overall, approximately seventy-two editions are extant within China, and these are grouped into the original edition and its derived lineages. The illustrations in these later printings are what most modern readers encounter, which is why the discrepancies in counts and attributions matter for any description of the book's visual record.

Reception, influence and translation

The Compendium's early standing in China was modest. Accounts agree that the Ming court showed little regard for the book after it appeared. A copy was presented to the emperor through Li's son, Li Jianyuan, but the emperor looked at it without showing much interest. The work was printed in 1596, and in the years that followed it had less effect at home than it later did abroad.

Its reputation grew in the seventeenth century. The book sold quickly enough to go through eight reprints in the seventeenth century alone. Its spread beyond China followed trade routes. It reached Tokugawa Japan via European traders around 1607, and the first Japanese edition appeared in 1637. European scholars also took notice in the mid-seventeenth century, and the French Jesuit Jean-Baptiste du Halde cited it extensively in his 1735 account of the Manchu empire. In the Qing period, the imperial editors included a 1655 edition in the Complete Collection of the Four Treasuries, which placed it within the official canon of medical writing. Further editions followed at intervals, with additional printings in 1603, 1606 and 1640, and then many more editions, with growing numbers of illustrations, down to the twenty-first century.

Its influence in East Asia was broader than medicine. One study describes the book as the most influential source of natural knowledge in East Asia after 1596. Its structure and terminology were used as a model for natural history writing in Japan. Its classifying methodology constituted the unchallenged template of Japanese works on natural history until the nineteenth century. Research on the Japanese reception qualifies this picture. One paper argues that early Japanese use of the work relied mostly on its lexicological and illustrative aspects rather than on the application of its classification and worldview to the local flora and fauna.

The first English translation was published in the early twenty-first century. Luo Xiwen's translation appeared from Foreign Languages Press in 2003, and a library catalogue record describes it as 52 volumes in 6. The publisher's note calls it the "First full English translation of the entire Compendium of materia medica", a claim that the entry attributes to the publisher. Before this edition, a bookseller's description of the work says that with the exception of a complete Japanese translation published in Japan in the 1970s, only parts of the Compendium of Materia Medica, mostly having to do with pharmacological therapeutics, have been translated. The same description adds that more difficult sections of the book involving history, culture and science have never been translated, with only abridged versions available in the West. The volume count also varies between sources: an encyclopedia article gives 53 volumes in total, while the translation's catalogue record gives 52, so the count should be read against the edition in question.

Modern assessment and research

Pharmacological claims in reference works

Reference accounts often credit the book with describing seemingly modern processes such as distillation and the uses of mercury, ephedrine, chaulmoogra oil, iodine, and even smallpox inoculation. Several of these summaries repeat one another's wording, so the list should be read as a secondary tradition rather than as independent verification of the primary text. Its reliability also depends on how each item is read. Ephedra (ma huang) was already a drug in the older materia medica tradition that Li inherited, so its inclusion reflects continuity more than discovery. The mention of inoculation is best understood as a record of a practice circulating in Li's period, not as a description of a method he invented.

These are historical statements about what the text records. They do not establish that the preparations worked, were safe, or matched modern pharmacology. Clinical evidence would require a separate evaluation of each substance and dose, and that is outside the scope of this entry.

Taxonomy and comparison with modern systems

Modern researchers focus less on individual remedies than on the book's classificatory architecture. A 2026 comparative study found that the 1,892 drug species are organized into the inorganic, plant, and animal kingdoms, and it evaluated Li's four-tiered hierarchy against the Linnaean system. Its conclusion is cautious: the parallels are structural and analogical; the Compendium does not articulate a theory of evolutionary transformation, and its ordering follows a cosmological framework rather than a genealogy. Another study of the plant section argues that the classification is functional rather than phylogenetic and should be evaluated on its own terms.

The terminology itself invites comparison. The characters gang and mu were later adopted as the terms for "class" and "order" in biological classification. The main structure of the book also reflects an explicit ordering logic. Li's preface describes moving from waters and fires through soils, worms, scaly animals, crustaceans, fowl and quadrupeds, ending with humans, so that the sequence proceeds from the low to the noble. Researchers therefore ask how far Li's groupings match later botanical, zoological and mineralogical taxonomy, and how far they encode other principles.

The totals also invite checking. The reported figures for herbs (1,094), animal substances (444) and minerals (275) sum to 1,813, which leaves about 79 of the 1,892 drugs in other categories, such as water, fire, tools and human-derived substances. The remaining substances are not assigned to a category in the summaries consulted here, so the exact breakdown should be verified against the primary text. Matching old drug names to modern species remains the main obstacle in this kind of comparison, and it recurs throughout the book's entries.

Recognition

The book received formal recognition in 2011, when it was inscribed on the UNESCO Memory of the World Register. This inscription is an institutional recognition of the book as documentary heritage. It is not an endorsement of its pharmacological claims. Later assessments of the work tend to separate its value as a record of Ming knowledge, its classification, and its illustrations from the question of whether its therapeutic statements are supported today.

References

1. Li Shizhenlink.springer.com
2. Li_Shizhenbionity.com
3. Li Shizhen 1518-1593 Definitive Best Guide To Bencao Gangmufuturafeed.com
4. Bencao gangmu 本草綱目 (www.chinaknowledge.de)chinaknowledge.de
5. Li Shizhenen.wikipedia.org
6. Bencao Gangmuen.wikipedia.org
7. The Ben Cao Gang Muancient-origins.net
8. Bencao Gangmu — Grokipediagrokipedia.com
9. 1 LI SHIZHEN Scholar Worthy of Emulationearthwormexpress.com
10. Review: <i>Bencao Gangmu</i>, Li Shizhen -- china.org.cnchina.org.cn
11. Top of pageloc.gov
12. Book/Printed Material Ben cao gang mu : wu shi er juan, fu tu er juan Juan 38loc.gov
13. Ben cao gang mu : wu shi er juan, fu tu er juanloc.gov
14. The Compendium of Materia Medica - ecph-chinaberkshirepublishing.com
15. Retrospect JournalThe Impacts of Li Shizhen’s Bencao gangmuretrospectjournal.com
16. 本草綱目 Bencaogangmu - Chinese Text Projectctext.org
17. Traditional Chinese medicines derived from the human bodyen.wikipedia.org
18. Selected reading of Li Shizen's medical works : the Chinese materia medica Ben cao gang mu 2021943097, 9789811238116, 9789811238123, 9789811238130, 9811238111 - DOKUMEN.PUBdokumen.pub
19. 《本草纲目》全文及翻译,原文及译文,白话文在线阅读-黄历吉日huanglijiri.com
20. The Original Source of Modern Research on Chinese Medicinal Materials: Bencao Textsheraldopenaccess.us
21. Combining classic and novel tools in the study of Historicalscholarlypublications.universiteitleiden.nl
22. Introduction According to ancient Chinese tradition, the mythological founderheraldopenaccess.us
23. Chinese herbologyen.wikipedia.org
24. Compendium of Materia Medicahealthandfitnesshistory.com
25. Compendium of Materia Medica（A classic text of Traditional Chinese Medicine）_Baiduwikibaike.baidu.com
26. © 2009 by Berkshire Publishing Group LLC 462 Comprehensive index startscarlanappi.com
27. Chinese Marine Materia Medicancbi.nlm.nih.gov
28. Bencao gangmubritannica.com
29. Chinese herbology - New World Encyclopedianewworldencyclopedia.org
30. (PDF) Li Shizhen and The Grand Compendium of Materia Medicaresearchgate.net
31. Li Shizhen - Infogalactic: the planetary knowledge coreinfogalactic.com
32. Li Shizhen & Compendium of Materia Medicaen.hubei.gov.cn
33. Li Shizhen: Scholar Worthy of Emulationitmonline.org
34. Who was Li Shizhen, barefoot doctor and father of traditional Chinese Medicine? - TropicalHainan.comtropicalhainan.com
35. The Compendium of Materia Medica (《本草纲目》)link.springer.com
36. Tang Shenweien.wikipedia.org
37. Compendium of Materia Medicahealthahoy.com
38. [Study on themisprints of the primary version of the Compendium of Materia Medica] - PubMedpubmed.ncbi.nlm.nih.gov
39. Li Shizhen Recompile the Compendium of Materia Medica - ChineseLearning.Comchineselearning.com
40. Bencao Gangmu -- C.16 Chinese materia medica, Precious metalswellcomecollection.org
41. Observing Images to Identify Herbs: A Study of Materia Medica Images from Early to Mid-Edo Japanjournals.lww.com
42. Chemistry:Compendium of Materia Medica - HandWikihandwiki.org
43. Ming Dynasty Compendia and Encyclopedias (Leishu)princeton.edu
44. Bencao gangmu and the World It Created - Chicago Scholarshipchicago.universitypressscholarship.com
45. The Bencao Gangmu in 17th century Japan: A Book of Nature or a Book of Words?nomadit.co.uk
46. Compendium of materia medica : Bencao gangmuiucat.iu.edu
47. Koeltz Botanical Books. Compendium of Materia Medica (Bencao Gangmu). Translated and annotated by Luo Xiwen. 6 volumes. 2003. XXXII, 4397 p. gr8vo. Hardcover. - English.koeltz.com
48. Bencao Gangmuen.wikipedia.org
49. Li Shizhenbritannica.com
50. TYPE Review PUBLISHED 21 July 2026 DOI 10.3389/fphar.2026.1867126frontiersin.org
51. Frontiersfrontiersin.org