Skip to content
COSMOS ORDER｜天人之學

Materia Medica

Research Domain

Materia Medica

Section titled “Materia Medica”

Reconstructing drug-selection logic from medicinal ontology, classical causality, single-drug validation, and functional layers within formulas.

Cognitive Curve
Data Overload → Medicinal Ontology → Causal Extraction → Single-Drug Validation → Global Distribution → Clinical Formula Construction

Domain Guide

Section titled “Domain Guide”
The most common way to learn materia medica today is to attach a long list of functions to each drug: supplement qi, strengthen the Spleen, dry dampness, promote urination, stop sweating, calm the fetus, and so on. Such summaries are convenient, but they create a serious problem. When every herb has a dozen functions and every symptom can be treated by dozens of herbs, clinical selection ultimately falls back on memory, habit, or accumulated experience.
Read the full domain guide

From “What Does This Drug Treat?” Back to “Why Can This Drug Do It?”

Section titled “From “What Does This Drug Treat?” Back to “Why Can This Drug Do It?””

The most common way to learn materia medica today is to attach a long list of functions to each drug: supplement qi, strengthen the Spleen, dry dampness, promote urination, stop sweating, calm the fetus, and so on. Such summaries are convenient, but they create a serious problem. When every herb has a dozen functions and every symptom can be treated by dozens of herbs, clinical selection ultimately falls back on memory, habit, or accumulated experience.

The Materia Medica domain tries to recover a deeper question: why can a medicinal substance do what it does? What relatively stable properties does the substance itself possess? And how do those properties generate different clinical effects when they enter different bodies, pathomechanisms, doses, and combinations?

The first stage therefore does not teach individual herbs. It explains why modern materia medica became so crowded. The Herbal-Medicine Fishing Net and Returning to the Source describes the “fishing-net” effect created as historical materia-medica literature accumulated ever more indications. Three-Dimensional Symbolic Convergence in Bencao Qiuzhen begins searching for dimensions that can compress that material again. The Rise of Zang-Fu Pattern Differentiation and the Retreat of the Three Yin and Three Yang looks back at how the rise of zang-fu pattern differentiation partially obscured earlier qi-transformation and Three-Yin/Three-Yang contexts. The Six Excesses, Phlegm, Accumulation, Qi and Blood, and the Priority of Root and Branch distinguishes causes, intermediate products, and branch manifestations among wind, cold, dampness, heat, phlegm, accumulation, qi, and blood. Multi-Label Medicines and the Dilemma of Formula Selection carries the resulting multi-label problem into formula selection, while The Shortcut and Limits of Symptom- and Disease-Specific Pattern Differentiation shows both the convenience and the limits of selecting “commonly used herbs” from a symptom or disease name alone.

Only after defining that problem do we enter the medicinal substance itself. General Framework for Decoding the Shennong Bencao Jing proposes the general method: do not treat the Shennong Bencao Jing as hundreds of isolated efficacy cards; search instead for the stable dimensions the text uses to describe medicinal ontology. The Image within Drug Names and Reasoning from Nomenclature examines whether names preserve clues about morphology, habitat, function, or qi transformation. The Five Qi, Five Flavors, and Locating Medicinal Nature and Flavor returns to the familiar system of qi and flavor, while Bodily Form and Treating Form through Form brings root, stem, branch, bark, seed, fruit, and bodily form back into the model. Ascent, Descent, Floating, Sinking, and the Four Seasons examines ascent, descent, floating, sinking, and seasonal direction. Clarity, Turbidity, Density, and Speed of Onset considers clarity, turbidity, density, depth, texture, and speed of onset. Harvest Season and the Endowment of Heaven and Earth finally reminds us that when a substance is harvested and which part is used may themselves be part of its medicinal nature.

By this stage a different view of drugs should begin to form: a drug is not a list of functions but a multidimensional vector. Flavor, thermal nature, form, part used, direction, clarity/turbidity, season, and naming clues constitute a relatively stable ontology. “Treat cough,” “stop diarrhea,” or “remove vexation” are often downstream clinical results when that ontology interacts with a concrete pathomechanism.

How can we know these interpretations are more than elegant retrospective stories? The third stage therefore turns to causality and cross-validation. Xu Lingtai’s Method of Reading Causal Arrows trains the reader to reconstruct causal arrows in Xu Lingtai’s explanations: because a substance has a certain property, it changes a particular pathological process, and a clinical effect follows. An Inner Canon Symptom Rule Base and Explaining One Classic through Another provides another classical evidence chain by connecting materia-medica descriptions with disease location, disease nature, and symptom rules in the Inner Canon. Only then does AI-Assisted Extraction of Causal Reasoning from Ancient Texts become useful: AI is not asked to invent drug properties, but to extract already-defined causal nodes from large bodies of classical text. Validating a Classification Framework for the Principles of Materia Medica asks whether the same dimensions reappear in another classificatory framework, and Zhang Zhicong’s Bencao Chongyuan and Extracting Innate Endowment uses Zhang Zhicong’s interpretation to test whether many surface-level functions can converge again onto fewer and more stable medicinal endowments.

Abstract methods must then survive pressure from real drugs. Atractylodes: Supplementing Spleen-Earth and Layering Root and Branch asks whether “supplementing the Spleen” is an intrinsic property or a downstream result. Licorice: The Five Zang and Six Fu and Its Relatively Unbiased Nature tests whether a drug with an apparently very broad range can still possess a coherent core. Cinnamon Twig: Branch Morphology as a Channel for Communication of Qi Above and Below examines whether branch morphology and qi movement correspond. Scutellaria: Yangming, Taiyin, and the Scope of Heat Clearing asks what precise scope lies behind the overly broad label “clearing heat.” Aconite and Asarum: Root-Branch Relations of Shaoyin and Taiyang studies one medicinal role across different disease positions and root-branch layers, while Kudzu and Eucommia: Derivation through Morphology and Channels reconnects morphology, channels, and concrete symptoms. These are not six unrelated drug stories; they are stress tests of the same ontology.

Once single-drug ontology becomes more reliable, The Spacetime Distribution of the 360 Medicines of the Shennong Bencao Jing can place the 360 medicines into a common spacetime distribution. They are no longer 360 disconnected points. We can ask which drugs cluster, which share surface indications but differ in underlying direction, which regions of the model are densely populated, and which remain sparse. A global distribution turns single-drug analysis into a system-level materia medica.

The genuinely clinical problem belongs near the end. The Inner Canon Spacetime Model as the Master Coordinate System for Materia Medica places human pathomechanism and drug vectors into a shared coordinate system. Zhu Jinzhong’s Thirty-Three Clinical Perspectives on Drug Use reminds us that drug choice is influenced not only by disease name but by position, nature, time, constitution, accompanying patterns, and other conditions. Dose-Based Pattern Differentiation and Dosage Design in Classical Formulas then makes dosage itself a control parameter: choosing the correct drug does not mean that the formula has already been determined. Proportional Pattern Differentiation and Functional Layers within a Formula raises the problem from individual dosage to the relative weights and functional layers among drugs in a formula.

Finally, Choosing among Physicians and the Criterion of Classical Self-Consistency asks what to do when physicians disagree. The answer cannot simply be to choose the master one prefers. Claims should be weighed against classical evidence, cross-text consistency, pathomechanistic coherence, and the ability to explain actual differences among formulas.

After all thirty articles, the reader should gradually leave behind the one-layer chain “symptom → find a drug” and acquire a fuller derivation:

What endowment does the drug itself possess → which pathomechanistic nodes can that endowment act upon → which node is actually abnormal in this patient → why choose this drug among several candidates → why this dose → why this relative proportion with the other drugs?

At that point materia medica stops being an enormous pharmacological memory task and becomes a discipline that can be reasoned through: from medicinal ontology to pathomechanism, from pathomechanism to prescription, and then from the prescription back to a test of whether our model of the drug was correct.

What You Will Be Able to Do

Section titled “What You Will Be Able to Do”

After completing this domain, readers should be able to move from isolated terms toward traceable models, rules, and derivations, while distinguishing direct textual support from structural hypotheses still under investigation.

Problem Reconstruction

Section titled “Problem Reconstruction”

Medicinal Ontology

Section titled “Medicinal Ontology”

Causal Extraction

Section titled “Causal Extraction”

Single-Drug Validation

Section titled “Single-Drug Validation”

Global Distribution

Section titled “Global Distribution”

Formula Construction

Section titled “Formula Construction”

Deeper Research

Section titled “Deeper Research”

This section is reserved for datasets, validation reports, experiments, counterexamples, and cross-domain studies.