A simple question came up recently in a discussion on the Voynich.ninja forum: are there trees in the Voynich manuscript? Although to be fair, nothing is simple ’round these parts. I decided to get to the roots of the matter in this blog post.

Herbal Manuscripts

When studying the Voynich for a while, a bit over a year in my case, there are some things one runs into sooner or later. For example, we often discuss illustrated herbal manuscripts, because just like large sections of the VM they contain images of plants.

A page of a 13th century copy of al-Ghafiqi’s herbal.

Standard herbals generally focus on the healing properties of plants, which is why they depict mostly herbs. Since scientific medicine hadn’t been invented yet, the early herbal traditions included plants which would not be expected by the modern reader, like the poisonous Aconite (wolf’s bane). In case you’re wondering, it apparently makes you pee and sweat, among other things. Like, you know, dead. The point is, a herbal in its most usual forms describes herbs and their medicinal uses.

A second thing to keep in mind is that herbals have a tendency to belong to a certain tradition. There were a number of early authors like Theophrastus and Dioscorides whose work formed the basis for later herbals. These works were illustrated, copied, translated, adapted and combined many times in the middle ages, which means that a relatively large amount of medieval herbal manuscripts remains today.


For example, someone with a bit of experience studying early herbals will be able to tell at a glance that the image above is a page from an early Dioscorides manuscript, while the one below shows plants from the Pseudo-Apuleius corpus.


Of course many of these manuscripts were somewhat tailor-made, with drawings added or replaced to suit the needs of the patron, but generally we expect a medieval herbal to belong to a certain tradition, a family of manuscripts. And even between these families there are clear correspondences and partial overlaps. A herbal manuscript usually sits snugly within a small group of related manuscripts, or more likely a large network that spans various groups.

A herbal without a tradition?

As far as we know, the Voynich plants do not belong to any surviving tradition. This is one of several reasons why the majority of past and current researchers are inclined to see the Voynich as the work of an eccentric individual author. Since we cannot link it to a herbal tradition, the implicit reasoning goes, it must be an original creation.

This is why the Voynich plants are problematic. It’s not because they are fictional, badly drawn, hybrids or American, as one often reads in uninformed publications. No, it’s because we lack a decent foothold, a frame of reference. The Voynich confronts those who want to understand its plants with two unique problems:

  1. We cannot read the text. Herbal manuscripts without explanatory text exist, but even those label the plants, at least providing us with a name. Even if the accompanying plant drawing is hard to recognize, we know what they intended.
  2. We cannot link the plants to a corpus of a known tradition.

Now to get to the point of this post. Despite this “in the dark” starting position, many researchers are convinced that  – ignoring one or two exceptions – there are no trees in the Voynich. In other words, that its botanical section is a herbal in the absolute purest sense, with images of relatively small herbs that can be collected for medicinal purposes.

Better don’t get sick…

This is no unusual assumption since indeed most illustrated botanical manuscript are herbals with a medical focus. Even Georg Baresch, the first known owner of the manuscript and certainly not the last to fall under its spell, was convinced that it contained medicinal knowledge. He did, however, also suspect that the plants were exotics since they could not be recognized by contemporary botanists.

Today’s mainstream Voynich research, as far as such a thing exists, dismisses the “exotic plants” part of Baresch’ comments, but still embraces the “medicinal herbs” part. Any readers accustomed to sensitivities within the field of Voynich research will understand then, that asking the question “are there more trees in the manuscript than we think?” is not trivial and even somewhat ideologically charged. But here goes…

The problem with trees

When artists needed to render a tree in a manuscript, they were often confronted with a problem – let me show you. Here’s a typical oak tree:


Now imagine that on a small piece of parchment, using a quill, you have to draw this tree, showing the following elements:

  • the overall structure of the tree: trunk, large branches, smaller branches…
  • the way the leaves are arranged on the twigs
  • the shape of the leaves, their edges and maybe veins
  • the shape of the fruits and/or blossoms…

Modern images will address some of these issues by making a composition like this one by the Tree Council of Ireland. They even provided a little man for scale, something which would have been quite handy in the VM.


Manuscript Solutions

Since it is not easy to properly draw a tree in a manuscript and give the reader a good view of the leaves and their arrangement, one wonders how trees were drawn in other manuscripts. Let’s have a look at some examples.

Some purely herbal manuscripts still contain tree-like plants, like tree spurge (Euphorbia dendroides) in the Juliana Anicia Codex. But E. dendroides is more of a shrub that can get pretentious after a while and look like a small tree. Hence, it is depicted relatively naturalistically without much trouble.

Tree Spurge

A bush of similar size is Juniperus phoenicea, a type of juniper, which means that the berries need to be shown.


In the image above, we can see a strategy in development. The oldest herbal, the Juliana Anicia on the left, just reduces the volume of the leaves somewhat in order to reveal the berries. In the Codex Neapolitanus (right) the size of the berries has been dramatically enhanced to increase their visibility, at the cost of realism.

The early Dioscorides manuscripts are clearly not the best place to look for trees, so let’s see what other herbals have to offer. My search for trees brought me first to a manuscript in the Historia Plantarum tradition, which was created at the Visconti court in Milan for King Wenceslas IV at the very end of the 14th century. It was based on the Taqwim al-Sihhah (the maintenance of health) by the 11th-century Baghdad physician Ibn Buṭlān, which in turn was based on Dioscorides’ corpus, among other sources. [1]

The manuscript contains several clear depictions of full trees. In most, of not all cases, it uses the strategy of enlarged leaves and fruits to provide a proper look. In the date palm for example, individual dates are blown up to the size of footballs.


And in the plum tree, the fruits’ surface has been enlarged over ten times:

One can only dream of such plums…

The “enlarged leaf and fruit” strategy is in fact used in many herbals, as is also the case in the famed Sloane 4016 and its manuscript family. Just have a look at these lads picking cherries the size of fists next to leaves large enough to serve as umbrellas.


All of this leads to the first important point I wish to get across in this post. If the Voynich manuscript uses a similar strategy, then it contains many more trees than one would suspect at first sight. Have a look at this plant from Egerton 747, the mother of a group of North Italian herbals often linked to the VM. What is it? A herb? A bush, like laurel? A shrub? Small tree? Oak-size tree? Do take into account the size of the leaf bottom right, which spans over half the length of the stalk.


It is in fact a cinnamon tree, which does have a slender stem when young but can grow quite large and either way looks much more tree-like than this. The point is, going by a drawing without legible text, it is hard to tell the scale of things.

So to recap, my first point is the following: since the Voynich botanical section does not provide us with any legible text or labels, nor any indication of scale, we cannot assume that its subject matter is limited to healing herbs. Yet this is exactly what many researchers assume, for no good reason whatsoever.

The Voynich oak

Luckily there is one plant in the manuscript about which most, if not all researchers agree it must be a tree, the one on f35v. It must be a tree since it is host to a vine that twines around it.


Additionally, the pair of plants on f35v is known to Voynich researchers as one of the few strong parallels between a VM plant and other herbal manuscripts. More specifically, a similar drawing is found in a group of Italian herbals known as the Tractatus de Herbis tradition [2]. More on this in the next post, for now the focus is on the VM.


Now, when I learned about this image I had the question of trees in the back of my mind, and I thought: “Well, here’s a tree in the Voynich, and everybody agrees it’s a tree. Hence, we can learn how the Voynich draws trees.”

One thing I noted right away is that, if this is indeed an oak, it’s a pretty bad one. Oaks have more than eleven leaves and those are much smaller. Also, oaks have large branches with multiple leaves on them – the arrangement we see in these plants is more appropriate for palms.

And then it dawned on me: it’s not a tree, it’s a single twig! Oak leaves are arranged alternatingly and form terminal clusters. This is exactly the behavior of the leaves in the VM plant. In the following image I removed all traces of the vine for clarity, leaving just the host plant, generally believed to be an oak. Next to it is a single oak twig: note the alternating buds along the twig and the cluster of buds on top.


So this is rather surprising: both the Voynich and the Tractatis manuscripts use a new strategy here, different from the “inflated leaf and fruit” approach. Instead of blowing out of proportion the parts they want to show, they just select one twig of the tree and depict it faithfully.

But unlike the Tractatis, the Voynich does what it does best: it adds massive, peculiarly shaped roots to the thing. So what we are seeing on f35v is in fact a decent twig with added roots. Diane O’Donovan already noted the lack of roots in the Tractatis manuscripts, yet she describes this as an (understandable) omission. I would like to turn this around and say that it is in fact the Voynich which stubbornly adds roots to everything that grows – even single twigs.

The twigs-with-roots hypothesis

This is the second point I want to make in this post. If it is indeed true that the Voynich depicts trees by adding roots to a twig, then we must take this into account in our identifications. Something which looks like a herb may in fact be a tree twig with an added root. 

Just some examples. F25r would make a fine twig:


The plants on f42r are little more than single leaves with roots:


And finally, here’s f6v, which looks like a herb with insignificant roots (a rarity in the manuscript). Next to it, a picture of a chestnut tree.


The position of the nuts at the end of the branch is spot on. The leaves do have the typical hand shape, yet in chestnuts they are shaped differently. Hence, this is probably not a correct identification, but my aim is not to identify this plant. Rather my aim is for the reader to understand how many of these “herbs” may well be twigs with roots.

It is only by specifically examining and describing these phenomena that we might arm ourselves with much needed tools to crack the Voynich’ toughest nuts.


[1] See https://www.wdl.org/en/item/11560/ for background and images on the Historia Plantarum.

[2] This parallel has been noted by a number of people independently (Rene Zandbergen, JK Petersen) and since then many have written about it, including renowned herbal manuscript expert Alain Touwaide. Some links :

O’Donovan: https://voynichimagery.wordpress.com/2016/07/04/poscript-the-juliana-anicia-codex-concluded/
Petersen: http://voynichportal.com/2013/07/21/large-plants-folio-35v/
Zandbergen: http://www.voynich.nu/illustr.html#herbal