Some books are hard to classify because they are badly organised.
Others are hard to classify because the organisation is larger than our categories.
The Liber Floridus belongs to the second group.
Ghent University Library MS 92 is the autograph manuscript of Lambert of Saint-Omer’s vast illustrated encyclopedia, completed around 1121. Albert Derolez describes it as the earliest illustrated encyclopedic compilation of the Latin West and one of the most complicated manuscripts ever made.
Hundreds of texts and pictures move through astronomy, geography, natural history, history, religion and other domains. Maps sit beside cosmological diagrams. Genealogies meet world history. Circular structures recur. The manuscript does not behave like a modern textbook with clean disciplinary chapters.
It behaves like a knowledge world.
Liber Floridus is a vital Voynich comparator because it proves that a medieval manuscript can be coherent while looking associative, heterogeneous, diagram-heavy and difficult to reduce to one linear subject.
Quick Read
- Ghent University Library MS 92 is the autograph manuscript of Lambert of Saint-Omer.
- It was completed around 1121 at Saint-Omer.
- Ghent research describes it as the earliest illustrated encyclopedic compilation of the Latin West.
- It contains hundreds of texts and pictures.
- Its subjects include astronomy, geography, natural history, history, religion and many other fields.
- Its maps are central to the history of medieval cartography.
- Its cosmological diagrams use circles, spheres, zodiacal structures and layered world models.
- Recent scholarship argues that the manuscript’s structure is not merely linear but has circular and concentric organisational logic.
- The manuscript is therefore a powerful control for the Voynich Rosettes, cosmological pages and apparently mixed section architecture.
- Its twelfth-century northern European origin makes it a deep structural control, not a provenance candidate.
Comparator Firewall
Circular diagrams + maps + encyclopedic mixture ≠ Voynich source. Liber Floridus shows what medieval knowledge architecture can do; it does not identify the Voynich manuscript’s lineage.
The Manuscript Is an Encyclopedia Without Modern Shelves
A modern encyclopedia separates astronomy from theology, geography from history and natural history from cosmology.
Lambert’s manuscript does not obey those boundaries.
The world is one created order.
Heaven, Earth, history, peoples, creatures and salvation can be connected because they belong to one intellectual cosmos.
This matters for the Voynich “Six Sections” problem.
Herbal.
Zodiac.
Cosmological.
Biological.
Pharmaceutical.
Recipes.
Those are our filing labels.
They may not correspond to the manuscript’s conceptual architecture.
Association Can Be an Organising Principle
Derolez’s long study of the autograph emphasises how unusually complex its construction is.
The manuscript gathers excerpts, diagrams and historical materials through a process of compilation rather than writing one continuous treatise from beginning to end.
That creates a crucial comparator for Voynich.
A manuscript can possess coherence at the level of relations even when neighbouring pages change subject.
The correct test is therefore not simply:
Does page B continue the topic of page A?
It may be:
What conceptual relation made the compiler place these information units inside the same architecture?
Circular Structure Can Exist Above the Page Level
Recent Ghent scholarship argues that the Liber Floridus should not be read as purely linear.
Its structure can be understood through concentric layers organised around central theological ideas.
This is an extraordinarily useful conceptual control for the Voynich Rosettes.
A circular diagram can be one page.
But a whole book can also have a circular conceptual architecture.
That means we should ask whether repeated Voynich page systems may be projections of a larger relational structure rather than isolated illustrations.
Maps Are Moral, Historical and Geographic at the Same Time
The Liber Floridus is famous in the history of cartography.
Ghent’s project materials emphasise that medieval mappae mundi were not simply modern road maps drawn badly.
They could combine geography, biblical history, moral structure and inherited cosmology.
This is a powerful warning for the Voynich Rosettes.
If the foldout is map-like, “map” does not necessarily mean:
- accurate scale;
- north at the top;
- one location per symbol;
- modern geographic realism;
- routes intended for physical travel.
A medieval map can compress cosmology, time, sacred history and geography into one surface.
Therefore a Rosettes-map hypothesis must specify what kind of map it means.
The T-O Map Shows How Much Geometry Can Compress
One of the traditional medieval world-map forms reduces the inhabited world to a circle divided by a T-like arrangement.
The representation is extremely compressed.
Yet it can still encode continents, seas and a theological-historical world order.
This teaches a general principle:
low geometric detail does not imply low semantic density.
The Voynich’s circles may likewise compress many relations into a small number of shapes.
The challenge is recovering the rule system rather than naming the picture from appearance.
Zonal Maps Turn the Earth Into Bands
The Liber Floridus also transmits zonal-map traditions linked to older authorities such as Martianus Capella.
Climatic or cosmological zones can be represented as bands.
The zodiac and ecliptic can be projected onto a world model.
This is a strong comparator for Voynich radial and ring diagrams because it demonstrates how geography and celestial structure can be joined geometrically.
Again, the motif is generic.
The discriminating evidence is how the sectors, labels and relationships behave.
The Manuscript Is an Autograph—That Gives Us a Rare Production Control
MS 92 is unusually valuable because Lambert identifies himself and the surviving manuscript is his autograph.
That means scholars can study an exceptionally complex medieval compilation without guessing whether every irregularity comes from later copying.
The autograph shows how one compiler’s project can grow complicated.
Material can be gathered.
Reorganised.
Placed into unusual relationships.
Illustrated with several visual grammars.
This matters for the Voynich because complexity does not automatically require many unrelated authors or accidental rebinding.
A single ambitious compiler can produce a genuinely difficult object.
But Autograph Does Not Mean Simple
This is the inverse lesson.
Voynich scholars sometimes look for production irregularity as evidence of multiple authors.
The Liber Floridus warns that a single-author compilation can contain:
- different source traditions;
- different page architectures;
- different visual modes;
- associative ordering;
- material revisions;
- large-scale conceptual layering.
Production plurality and information plurality must therefore be measured separately.
Natural History Is One Node, Not the Whole Book
Ghent describes the manuscript as containing natural history among many other fields.
This gives us another control for Voynich plants and animals.
A plant image can belong to an encyclopedic world model rather than a medicinal herbal.
An animal can belong to geography, cosmology, moral knowledge or natural history.
Therefore the presence of plant-like or animal-like images does not establish medical purpose automatically.
The functional relation to neighbouring text and diagrams must decide.
This Is a Strong Control Against Paduan Overfitting
Nothing about the Liber Floridus is Paduan.
It was made at Saint-Omer more than two centuries before the Voynich parchment.
Yet it already contains:
- complex cosmological diagrams;
- zodiacal structures;
- maps;
- natural history;
- multiple visual grammars;
- associative organisation;
- circular conceptual structure.
These features therefore have little or no localisation value for Padua by themselves.
They belong to a much wider medieval diagram culture.
What Liber Floridus Can Legitimately Teach Voynich Research
- A medieval manuscript can be coherent without following modern disciplinary boundaries.
- Association can organise an encyclopedia.
- Circular and concentric structure can exist at both page and whole-book levels.
- Maps can combine geography, cosmology, sacred history and moral meaning.
- Several visual grammars can project different dimensions of one world model.
- An autograph compilation can still be materially and conceptually very complicated.
- Natural history can be one module inside a much wider encyclopedic project.
- Many Voynich-looking cosmological features were already widespread centuries before 1400.
What It Does Not Prove
- that Voynich is an encyclopedia;
- that the Rosettes foldout is a mappa mundi;
- that Voynich circles represent climatic zones;
- that its mixed sections have theological unity;
- that one compiler made the whole Voynich manuscript;
- that natural-history imagery is non-medical;
- that broad medieval cosmology localises the Voynich anywhere.
The eduKate Associative-Architecture Test
- Represent Liber Floridus as a graph of concepts, texts, images and page positions.
- Measure whether related topics are adjacent, recurrent or connected through shared diagrams.
- Encode its circular and concentric structures independently of content labels.
- Measure transitions between visual grammars.
- Compare with other medieval encyclopedic compilations.
- Build a neutral graph for Voynich page systems without assuming six semantic sections.
- Test whether recurrent relations predict page transitions better than modern section labels.
- Ask whether the Rosettes and circular diagrams act as hubs in that graph.
- Reject encyclopedic models that cannot explain local textual regularities.
- Keep function, source lineage and provenance separate.
Student Lens: A Book Can Be a Knowledge Map
Primary: make a poster linking animals, places, weather and history instead of putting each subject on a separate page.
Lower Secondary: compare a linear table of contents with a concept map.
Upper Secondary: distinguish thematic adjacency from network-level coherence.
JC and adult readers: model the codex as a knowledge graph projected onto a linear physical sequence.
Research Sources
- Ghent University — Albert Derolez, The Making and Meaning of the Liber Floridus
- Biblissima — Ghent University Library MS 92
- Ghent University — Liber Floridus: Early Twelfth-Century Cartography
- Ghent University — The Book and the City: Circular Structure in Liber Floridus
- eduKateSG Comparator — La Sfera
- eduKateSG — The Rosettes Foldout
The Final Idea
The Liber Floridus is not useful because it gives us a Voynich twin.
It does something better.
It demonstrates that medieval knowledge could be organised in ways that look strange when forced through a modern table of contents.
Circles can be cosmology.
Maps can be history.
Natural history can sit beside theology.
Association can be structure.
The Voynich may remain difficult not because it lacks organisation, but because we have not yet found the level at which its organisation becomes visible.