Sloane MS 4016 as a Voynich Comparator

If we want to know what a real manuscript relationship looks like, Sloane MS 4016 is one of the best controls in the entire Voynich comparison library.

Not because it resembles Voynich.

Because its own relationship is unusually explicit.

The British Library catalogues Sloane MS 4016 as an Italian herbal of the North Italian group, produced in Lombardy around 1440, and—following Felix Baumann’s study—as a copy of Paris, École des Beaux-Arts, Masson 116.

Sloane MS 4016 gives us something Voynich research desperately needs: a control case in which we can study what copying, transformation and visual continuity look like when descent is supported by a manuscript family rather than guessed from one resemblance.

That makes this article less about “finding the Voynich source” and more about building a proper evidential ruler.

Quick Read

British Library shelfmark: Sloane MS 4016.
The British Library dates it to about 1440, with a catalogue range of 1435–1445.
Origin: Lombardy, Italy.
Material: parchment codex.
Dimensions: approximately 360 × 255 mm.
Foliation: 108 folios plus preliminary/end material recorded by the Library.
Script: Gothic.
Decoration: full-page coloured miniatures on each folio, mostly plants in greens and browns, with captions and often animals or people.
The British Library classifies it, following Baumann, as a copy of Masson 116.
Its known relationship lets us measure how images can be copied, resized, regrouped, relabelled and stylistically updated.
For Voynich, it is a control for descent, not evidence that Voynich belongs to the same family.

Comparator Firewall

Sloane 4016 is not being proposed here as a Voynich exemplar, source, cipher key or provenance anchor. Its value is methodological: it shows what real copying leaves behind.

This distinction is non-negotiable.

The Voynich Manuscript and Sloane 4016 are both late-medieval illustrated books. That shared historical breadth is enough to justify comparison. It is not enough to justify lineage.

What the British Library Record Gives Us

The British Library’s catalogue is unusually useful because it separates description from interpretation.

It records Sloane 4016 as one volume, written in Latin script on parchment, with Gothic script and full-page miniatures. The Library places its origin in Lombardy and dates it to approximately 1440. Its physical size is about 360 × 255 mm, and the foliation is recorded as 108 folios plus additional unfoliated material.

The key sentence for our purposes is the catalogue’s classification of the manuscript as part of the North Italian group and as a copy of Masson 116.

That statement is not based on one plant looking similar.

It belongs to a larger art-historical and manuscript-tradition argument developed through systematic comparison.

The Copy Is Not a Facsimile

This is the first major lesson.

When modern readers hear “copy”, they often imagine a photocopy.

Medieval copying was not like that.

A copyist or workshop could preserve the identity of an image while altering its:

scale;
position;
colour;
surrounding figures;
page grouping;
clothing and hairstyle of human figures;
caption form;
decorative emphasis;
relationship to blank space.

Sloane 4016 is especially useful because the British Library describes full-page compositions with plants, captions and additional animals or people. The inherited visual material has been re-hosted inside a new fifteenth-century interface.

The identity can persist while the presentation changes.

Masson 116 to Sloane 4016: A Real Transformation Edge

Masson 116, probably made in northern Italy in the later fourteenth century, survives as an extraordinary visual corpus with very little continuous explanatory prose. Its pages appear to have been prepared for more text than was ultimately supplied.

Yet the incomplete state did not make its images useless.

Sloane 4016 demonstrates the opposite.

Images could be extracted from an earlier visual corpus and given a new life.

The transformation can be written schematically:

earlier image tradition → Masson visual corpus → Sloane copying and regrouping → new fifteenth-century book

That edge is valuable because it is not merely conceptual. The British Library explicitly records the relationship in its catalogue.

What Actually Survives Across a Copying Event?

A proper comparison between Masson and Sloane asks which features remain stable enough to preserve object identity.

overall plant silhouette;
branching logic;
root form;
flower or fruit placement;
associated animal or human motifs;
the relative order of entries;
name or synonym;
rare iconographic oddities;
shared mistakes.

Then we ask which features are free to move.

page scale;
colour intensity;
costume;
spacing;
composition;
decorative details;
number of objects placed together.

This separation between stable and plastic features is exactly what Voynich comparisons need.

A true relationship should preserve more than mood.

Why Full-Page Composition Matters

Sloane’s full-page format is an information-design decision.

It changes how the reader encounters the material.

Instead of placing a small image inside dense explanatory prose, the page can foreground visual recognition. Several objects may share the space. Captions provide a compact naming layer. People or animals can become part of the identification or narrative environment.

The lesson is not that Sloane was “just a picture book”.

The lesson is that manuscript function can be redistributed across modalities.

Less prose does not mean less structure. A visual interface can carry ordering, identity, memory and retrieval functions that prose would otherwise carry.

Captions Are a Small but Powerful Layer

The British Library specifically notes captions.

That makes Sloane useful for the Voynich label problem.

A caption does not need to be a sentence. It may function as:

object name;
synonym;
cross-language equivalence;
material name;
category marker;
memory cue.

Therefore the existence of short label-like strings in Voynich cannot by itself tell us whether the surrounding system is prose, catalogue, recipe, index or something else.

But a known captioned image book lets us ask better questions.

Are labels placed consistently?

Do repeated image types attract repeated label forms?

Are captions syntactically different from running text?

Do label lengths cluster differently?

These are testable.

A Copy Can Modernise the World Around the Object

Art historians have used costume and stylistic details in Sloane 4016 to help date and localise it. This illustrates a subtle point about copying.

A copied medicinal image can keep an inherited core while its social wrapper updates.

A collector becomes a fifteenth-century figure.

A garment changes.

A page becomes more courtly.

A decorative preference enters.

If we compare Voynich with any known manuscript, we therefore have to distinguish:

the inherited informational skeleton;
the copyist’s transformation;
the receiving culture’s style.

Without that separation, dating from imagery becomes circular.

Known Descent Versus Visual Similarity

Suppose one Voynich plant resembles one Sloane plant.

That is weak evidence.

Suppose ten do.

Better, but still vulnerable to selection bias.

Suppose the same ten also occur in the same relative order, preserve the same unusual root distortions, share the same rare associated objects, and generate comparable label structures.

Now the hypothesis becomes much more interesting.

Suppose the predicted eleventh image is then found where the model says it should be.

That is stronger again.

Sloane teaches us to move from resemblance toward prediction.

The Negative Evidence Matters

When Masson and Sloane are compared systematically, mismatches can be explained within a known copying process.

When a proposed Voynich relationship produces only the matches and ignores the non-matches, the analysis is incomplete.

A serious source hypothesis must count:

expected correspondences that are absent;
order breaks;
images with no plausible counterpart;
label behaviour that does not transfer;
features common to unrelated controls;
chronological or geographic contradictions.

Failure is data.

Why Sloane Is More Valuable Than a “Look-Alike Gallery”

Voynich discussions often accumulate images that look similar.

That can be useful for generating hypotheses.

It is not enough for testing them.

Sloane gives us a calibration set in which the family relationship exists independently of our desire to explain Voynich.

That lets us measure:

how similar true copies actually are;
how much true copies can diverge;
which features remain stable under copying;
which visual features are too plastic to carry much evidential weight;
how page redesign affects apparent similarity.

That is a much harder and much more useful benchmark.

Provenance: Another Lesson in Layering

The British Library records a later ownership chain including Nicolas-Joseph Foucault, then Sir Hans Sloane, before the manuscript entered the British Museum with the Sloane collection in 1753.

Those later owners matter to the object’s history.

They do not tell us who originally commissioned it.

This is another discipline Voynich research needs constantly.

Custody history is not production history. A later owner is not an author, and a later annotation is not original text.

What Sloane 4016 Can Legitimately Teach Us About Voynich

Real manuscript descent produces families of correspondences.
Copies can substantially redesign page composition without losing inherited image identity.
Captions can carry a compact naming layer independently from prose.
Visual materials can move between manuscript environments.
Human and animal details can be updated while plant structure remains recognisable.
Image order, rare motifs and shared errors can be more diagnostic than general resemblance.
Later provenance layers must be separated from original production.
A copy can preserve information while changing the interface through which the information is read.

What Sloane 4016 Does Not Prove

that Voynich belongs to the North Italian Tractatus de herbis group;
that Voynich was produced in Lombardy;
that Voynich plant figures are disguised copies of Masson or Sloane;
that Voynich labels name plants in Latin;
that a text-light visual medical book has the same function as Voynich;
that any isolated Sloane–Voynich resemblance is historically meaningful;
that shared medieval date range establishes contact.

A Blind Copy-Detection Test for Voynich

Select a fixed Voynich subset before consulting candidate comparators.
Encode image features neutrally: branching, leaf attachment, root form, flower position, auxiliary figures, label position.
Hide manuscript identities from scorers.
Mix known Masson–Sloane pairs with unrelated medieval image pairs and Voynich–comparator pairs.
Ask scorers to rank structural correspondence.
Measure whether known copies separate cleanly from unrelated controls.
Only then ask where Voynich pairs fall in that calibrated distribution.
Require any proposed lineage model to predict additional unseen correspondences.

This is more defensible than browsing until something “looks right”.

Student Lens: Copying Is Not Duplication

Primary: copy a simple plant drawing, then allow the second student to change colour and size while preserving identity. Which details matter?

Lower Secondary: compare several known Masson/Sloane-style correspondences and separate stable features from decorative ones.

Upper Secondary: design a blind image-matching experiment and calculate false-positive rates.

JC and adult readers: treat manuscript copying as a noisy transmission channel: which features have high fidelity, which are lossy, and which are replaced by receiver-specific representation?

Research and Primary Sources

Continue Through the Voynich Comparator Library

The Final Idea

Sloane MS 4016 matters because it shows that copying can be both faithful and transformative.

A visual tradition can survive while the page changes.

A plant can remain recognisable while its scale changes.

A manuscript can preserve inherited identity while adopting the style of a new place and generation.

That is what a real copying process looks like.

And because Sloane’s relationship to Masson can be studied outside the Voynich problem, it gives us a proper control.

Before claiming that Voynich descends from anything, make the proposed relationship survive the standard set by manuscripts whose descent we can actually see.

Voynich Comparator | Sloane MS 4016: What a Known Medieval Copy Teaches Us About Images, Labels and Descent