The Material World of Medicine as a Voynich Comparator

One of the easiest mistakes in medieval medical history begins with a modern picture.

We see a herbal.

We see plants.

We conclude that the material world of medicine was botanical.

It was not.

Medieval pharmacology worked with materials from several domains of the natural world. Plants were crucial, but so were animal products, minerals, resins, gums, aromatics and substances transported over long commercial routes.

A medical book therefore does not merely classify “plants”.

It participates in a larger problem:

How do you keep the identity, properties and uses of heterogeneous materials stable while they move across ecologies, languages, merchants, containers and recipes?

That question is highly relevant to Voynich because the manuscript’s visual world is also heterogeneous.

Plants.

Fragments.

Containers.

Human figures.

Stars.

The right historical lesson is not that those things must all be medicine.

It is that a real medieval medical material system was already far broader than one illustrated herb.

Quick Read

Modern scholarship on medieval Mediterranean pharmacology treats materia medica as a cross-cultural field spanning Latin, Byzantine, Islamicate and Jewish traditions.
Medicinal materials included plants, minerals and animal products, not plants alone.
Some substances were local; others travelled from Asia, the Indian Ocean and wider Mediterranean commercial networks.
Myrobalans are a classic example: Asian plant materials became common Mediterranean therapeutic ingredients through long-distance circulation.
Musk and ambergris demonstrate animal-derived materia medica whose high value reflected long transport and rarity.
Animal pharmacology could use blood, bile, fat, liver, milk, urine and other substances; some were common ingredients in topical preparations.
Mineral substances also entered medical practice, including materials catalogued in manuscripts such as BnF Latin 6823.
Multilingual drug names show that material identity had to survive translation, transliteration and commercial movement.
Substitution texts such as Quid pro quo demonstrate that medieval practice recognised functional alternatives when the preferred ingredient was unavailable.
This means a practical pharmacology is organised not only by what a thing is, but sometimes by what job it can perform.
That distinction creates a strong Voynich test: visual similarity should not be assumed to equal material identity, and material difference does not rule out functional equivalence.

Comparator Firewall

This article reconstructs the material ontology of historical medicine. It does not identify any Voynich plant, vessel, colour or fragment as a specific drug.

The comparator is valuable because it widens the historically real classes while narrowing the inferences we are allowed to make from pictures.

Materia Medica Is a Classification of Usable Matter

The Latin phrase materia medica is useful because it points our attention toward material rather than modern disciplinary category.

The key question is not:

Is this botany?

It is:

Can this material be recognised, obtained, prepared and used within a therapeutic system?

That definition immediately admits several material families.

Plants: The Largest Visual Class, Not the Entire System

Plants dominate many surviving medical manuscripts because they are visually rich and pharmacologically important.

But even within the plant world, one organism can generate several distinct medical objects.

root;
bark;
leaf;
flower;
fruit;
seed;
resin;
gum;
juice;
oil.

The living plant and the stored drug are therefore not always the same visual object.

A whole plant in a manuscript may support recognition.

An apothecary may encounter only the dried root.

A recipe may require the seed.

A trade record may use a market name.

A synonym list may preserve a learned Latin term.

One biological source has become several information states.

organism ≠ harvested part ≠ traded commodity ≠ prepared ingredient

This is one reason Voynich whole-plant ↔ fragment hypotheses are worth testing structurally rather than by species name alone.

Animal Products: Medicine Does Not Stop at the Botanical Boundary

Medieval pharmacology also used animal-derived materials.

Historical work on animal materia medica documents the use of eggs, milk, blood, flesh, bile, fats and other products in recipes. Fats from pigs, goats and sheep, for example, could serve as ingredients in unguents.

Some animal substances were locally obtainable.

Others belonged to long-distance trade.

Musk and ambergris could travel from far outside the immediate Mediterranean and become highly valued ingredients.

This matters because an illustrated medical source field can legitimately contain animals without becoming a zoology book in the modern sense.

BnF Latin 6823 demonstrates the same category permeability materially: the Manfredus plant text is followed by animal material, and Masson 116 later re-hosts some animal and figurative imagery alongside plants.

Minerals: The Medical World Also Comes Out of the Ground

Minerals complicate the visual story further.

A plant can be drawn as an organism.

A mineral often does not offer the same obvious diagnostic silhouette.

Its useful identity may depend more heavily on name, colour, texture, provenance, preparation or trade convention.

Biblissima’s indexed descriptions of BnF Latin 6823 include mineral entries such as ceruse, cinnabar and vitriol, alongside other natural and medicinal substances.

Medieval pharmacological literature more broadly contains minerals and mineral-like substances within the same therapeutic world as plants and animal products.

This tells us something important about visual absence.

A material system can contain important classes that are not represented by large naturalistic pictures.

Trade Changes the Meaning of “Local”

A manuscript can be locally made and materially global.

This is one of the strongest lessons from recent work on medieval Mediterranean pharmacology.

Myrobalans originated in Asia and were naturalised into Mediterranean pharmacological practice. Other imported substances moved through Islamicate, Byzantine and Latin networks. Names crossed languages. Commodities crossed seas.

The medical system therefore contains at least three different geographies:

ecological origin — where the organism or material naturally came from;
commercial route — where merchants carried it;
medical use location — where a practitioner or patient used it.

And then there is a fourth:

manuscript production location — where a particular book was made.

Those four places can all differ.

A foreign drug in an Italian manuscript is not evidence that the manuscript was made where the drug grew.

This firewall matters enormously whenever a Voynich plant identification is used as a shortcut to geography.

Names Travel Differently From Materials

A commodity can move while its name changes.

A name can move even when the commodity does not.

Medieval pharmacological scholarship documents Greek, Arabic, Syriac, Persian and Latin names coexisting in drug traditions. Some foreign names were transliterated. Others were paired with local equivalents. Some exotic animal names remained difficult to translate.

This means identity is partly linguistic infrastructure.

The material has physical properties.

The market has names.

The learned book has names.

The local practitioner may have another name.

A robust system needs a crosswalk.

Substitution Reveals a Deeper Ontology: Function Can Matter More Than Identity

This is where medieval pharmacology becomes especially interesting.

Suppose a recipe requires ingredient A.

Ingredient A is unavailable.

A substitution tradition may permit ingredient B.

A and B are not the same substance.

They become equivalent within a bounded practical relation.

identity difference + functional equivalence under conditions

This is a much more sophisticated knowledge structure than a dictionary of ingredient names.

It is a graph.

One substance can point to another through an allowed replacement relation.

Early and medieval substitute-drug traditions preserve hundreds of such relations, and later Quid pro quo lists became a practical part of pharmacy.

Egerton MS 747 and BnF Latin 6823 physically place substitution material beside other pharmaceutical infrastructure.

Why Substitution Matters to Voynich Image Matching

A naïve cross-modal theory may assume:

same label family = same ingredient.

But a practical system could encode a broader relation.

Same preparation class.

Same medical property.

Same substitute class.

Same part type.

Same storage class.

This does not give us permission to reinterpret every mismatch as substitution.

It gives us a new variable that must be tested independently.

A substitution hypothesis should predict explicit structural asymmetry: two different material identities participating in similar operational contexts.

Physical Properties Can Become Classification Variables

Once materials are transformed into medicines, their physical behaviour begins to matter.

Dry.

Powdered.

Oily.

Sticky.

Pourable.

Spreadable.

Recent work on early pharmacy has emphasised that material form and making process could help organise pharmaceutical categories.

This is why the Grabadin preparation classes and apothecary-container evidence matter beside the material world.

The source substance may begin as plant, animal or mineral.

After transformation, the practical category may become powder, syrup, oil, pill or ointment.

source ontology → preparation ontology

A working pharmacy may therefore classify the same material differently at different stages.

The Same Substance Can Have Several Identities at Once

Consider what one medicinal material might be simultaneously.

a biological species;
a particular harvested part;
a commodity;
a learned Latin name;
a vernacular name;
an imported drug;
a member of a property class;
a substitute for another drug;
an ingredient in a compound;
a stored object in a particular container.

Those identities overlap.

They are not interchangeable.

This gives us a powerful rule for interpreting technical labels:

A label can be correct about an object without naming its biological species.

It could identify the part, property, trade class, preparation class or operational role.

This is one reason “image + label = species name” remains too narrow as a default Voynich model.

The Material World Makes the Provenance Problem Harder

Suppose a Voynich plant is identified convincingly as a species from a distant region.

What follows?

Perhaps the manuscript was made there.

Perhaps the plant was imported.

Perhaps the image was copied from a travelling exemplar.

Perhaps the tradition preserved a foreign drug name long after the material had become normal in Mediterranean pharmacy.

Perhaps the image is wrong.

Botanical geography narrows only when the trade and manuscript-transmission alternatives are modelled too.

What the Material World Can Legitimately Teach Voynich Research

Medieval materia medica was materially heterogeneous.
Plants were central but not exclusive.
Animal and mineral substances belonged to the same pharmacological environment.
Long-distance trade separated ecological origin from medical-use geography.
Drug names could cross and change among languages.
One biological source could generate several usable parts and commodity forms.
Substitution created functional equivalence between materially different drugs.
Preparation could reclassify raw materia medica by physical form.
A technical label need not encode biological species identity.
Medical provenance claims must model trade and intellectual transmission, not only natural distribution.

What It Does Not Prove

that Voynich’s plants are medicinal;
that its fragments are harvested drug parts;
that any unusual shape represents an animal or mineral drug;
that a repeated label family marks a substitution class;
that an exotic-looking plant localises the manuscript to its ecological origin;
that all jars encode physical-state categories;
that missing numerals can be ignored in a pharmaceutical reading;
that every mismatch between two proposed plant identities can be rescued through substitution.

A Material-Ontology Test for Voynich

Separate whole-organism images from part images without naming species.
Classify fragments by observable morphology: root-like, leaf-like, seed-like, resin-like, uncertain.
Classify vessels independently by form.
Record label families without semantic assignment.
Test whether label families predict biological source, part type, preparation-like class or vessel class.
Allow substitution as a candidate relation only if two materially different classes share a repeated operational context.
Compare results with known materia-medica and pharmacy controls.
Include non-medical natural-history manuscripts as negative controls.
Test on held-out pages.
Reject the medical-material ontology if the proposed classes do not generalise better than simpler visual or scribal explanations.

Student Lens: One Object Can Belong to Several Classification Systems

Primary: classify a lemon as a fruit, food, plant product, yellow object and source of juice. Notice that one thing can have several correct categories.

Lower Secondary: compare the biological identity of a substance with its commercial and practical identity.

Upper Secondary: model substitution as a relation: two non-identical materials can become equivalent for one bounded purpose.

JC and adult readers: treat materia medica as a typed knowledge graph linking source organism/material, names, geography, usable part, property, trade route, preparation and substitution relations.

Research Sources

Continue the Source-Field Route

The Final Idea

The medieval medical world was not organised around the modern question, “What species is this?”

It had to answer several questions at once.

What is it?

What is it called here?

Which part is useful?

Where can it be obtained?

What can replace it?

What does it become when prepared?

If Voynich encodes practical materia medica, its visual and textual categories should eventually resolve some of those same operational distinctions. Until they do, the material world is a control—not a translation.

Voynich Comparator | The Material World of Medicine: Plants, Animals, Minerals, Trade and Substitution