The Voynich Ninja

I am not sure why it is also referred to as "Tacuinum Sanitatis"; that seems wrong.



You  are right, of course, the text of the Casinatense ms. comes from the Circa Instans tradition, while there is some overlapping in the illustrations

Adding to what Rene wrote, the two main traditions of illustrated herbals in medieval Europe were:

• Pseudo-Apuleius: You are not allowed to view links. Register or Login to view. dating to late antiquity (IV Century) that was copied during all the Middle Ages
  
• Tractatus de Herbis: i.e. the illustrated version of Circa Instans. The text was created in Southern Italy in the XII Century. The earlies illustrated version is BL Egerton 747, also from Southern Italy (1300 ca). Manfredus de Monte Imperiali (BNF Lat.6823), still in Southern Italy (1340 ca), extended the text and improved the illustrations: this work was the source for Casanatense ms 459. You are not allowed to view links. Register or Login to view. lists some of the manuscripts in this tradition.
  

Dioscorides mostly remained confined to Greece, but for the passages that were incorporated into Pseudo-Apuleius and De Herbis.

The graph attached to You are not allowed to view links. Register or Login to view. gives the overall picture according to Sergio Toresella.

---

In my opinion, the main problem with the approach discussed in this thread is that it heavily relies on wishful thinking. In particular, it assumes that the text written in a unique script and accompanied by unique illustrations is a more or less word-to-word mapping of texts that, in the XV Century, already were hundreds of years old and had been copied numberless times. I believe that the originality of the illustrations suggests a corresponding originality of the text. As to confirm the originality of the VMS, the few cases in which we can pin something specific do not fit with the know traditions, e.g.:

• Viola bicolor (if this is the subject of f9v) does not appear in any medieval illustrated herbal
  
• f35v, that so closely parallels the oak-and-ivy illustration by Manfredus (BNF Lat. 6283) cannot correspond to ivy, since the VMS wine has no leaves.
  

Another argument that I find even stronger is that medieval herbals tended to have a regular textual structure. Most plants were presented in a format that was uniform across each work (though typically different from work to work). The most basic trait is that each plant description either starts with the name of the plant or with an expression (e.g. "Nomen Herbae") that introduces the name of the plant. In medieval herbals, plant names tend to be hapax legomena or at least infrequent words in the whole work. In the VMS, the first word of herbal pages typically is a "Grove word"; these words indeed are hapax legomena, but they also are morphologically different from words that appear elsewhere: typically they begin with EVA:p/f. In several cases, if one removes the initial p/f, a fairly common word is produced, posing some doubts about the idea that most Grove words really are hapax legomena after all.

Plant names aside, other structural elements can be easily detected in medieval herbal. For instance, Egerton 747 has a first part providing:

• names of the plant in different languages;
  
• its galenic properties (degrees of cold/hot wet/dry);
  
• a description of the plant.
  

(Sometimes not all these elements are present)

This introduction is followed by a list of recipes, each of them typically starting with "ad" or "contra".
If one highlights the words that are common to Canapa and the following plant (Cammeleunta Alba) in You are not allowed to view links. Register or Login to view., it is easy to see that the two parts of the two short "chapters" can be connected. Also, "canapa" occurs 3 times in the paragraph against a total of 5 occurrences in the whole manuscript.

[attachment=5374]

For the Voynich manuscript, we have not been able to point out any comparable systematic structure yet. The fact that Herbal pages include both Currier A and B makes this section both particularly interesting and particularly hard to analyse (though similar attempts on more homogeneous sections have also failed). While I began my Voynich research following the ideas of Stephen Bax and hoping to find some correspondence with external sources, I later switched to Emma May Smith's approach. Since then, I prefer to focus on searching patterns internal to the VMS. If and when we find such structure, we can hope to find it reflected elsewhere, though it may be the case the Voynich herbal has no structure or has a structure that is totally unique to it and unparalleled in other herbals.

(06-03-2021, 03:01 PM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.

• f35v, that so closely parallels the oak-and-ivy illustration by Manfredus (BNF Lat. 6283) cannot correspond to ivy, since the VMS wine has no leaves.
  

Just a small point... my favorite ID for the twining plant is cuscuta (which does not have leaves because it is a parasitic plant that sucks off its host).

But, it is possible that it is ivy. My second favorite ID is Smilax. One of the distinctive things about this ivy is that it keeps its berries after the leaves are gone:

You are not allowed to view links. Register or Login to view.

There are a couple of species of Smilax in medieval manuscripts, although they are usually shown with the leaves.

MarcoP wrote

Dioscorides mostly remained confined to Greece, but for the passages that were incorporated into Pseudo-Apuleius and De Herbis.

That is simply wrong, the alphabetical D. in the Vienna D. tradition was translated (6th c., D. longobardus), reworked very likely by Constantinus Africanus and commented on by Petrus de Abano and was very present in the Late Middle Ages, only it was not illustrated.

I think it is a great  mistake to ignore the non illustrated plant description and recipe tradition, especially Galen, Avicenna, the Encyclopedists and so on, of course the VMs is NOT a traditional illustrated herbal, whose text was copied one to one somewhere.

[attachment=5375]
Example from a book.

Re F35v. In my opinion, it can only be Vitis vinifera sylvestris ( wild grape).
It cannot be ivy, which already occurs in the VM.
Its clusters and the way it twines around trees give the VM drawing a sense.
Strongly represented in medicine. ( Classical)

You are not allowed to view links. Register or Login to view.

The basic point of determination is whether or not any sort of text and image correlation can be found through the process of plant identification in the VMs. If there is a proposed positive determination beyond hypothetical, then there needs to be at least a single example. The best potential candidates are going to be those where correspondence and evidence are strongest. 

If we are seeking correspondence in text, then there must be a comparison with historical texts, and the primary characteristic of the VMs 'text' on Cannabis is the short section formatting structure with three sections. Correspondence in the VMs is built on structure far more than on appearance. So, are there relevant, historical historical texts on Cannabis that match the three-part VMs format? This still seems to be one of the stronger candidates. so if the comparison can be made, it should be.

Does illustration to text correspondence exist? If the members of the ninja forum can't find it, who will?

R. Sale, maybe there's a plant that is better than Cannabis. Even though it's a probable ID, the problem with Cannabis is that it had many different names and many different spellings of those names.

So, are there other options? Yes, there may be.

Alternate Plant IDs

There are two plants you might want to consider. One is Centaurea (f2r), the other is Calendula (f18r).

These IDs are also disputed (partly by me), but they are less controversial than most of them.

• I think it is possible to ID the VMS plant that looks like Centaurea down to the binomial because the elliptical shape of the leaves has been emphasized and there are only a few species of Centaurea that have the elliptical leaves together with the distinctive spots on the scaly part below the petals. It's my opinion that f2r might be Centaurea jacea. Even if it is not C. jacea, it is probably Centaurea.
  
• The other one is Calendula. There is a good drawing of what might be Calendula on You are not allowed to view links. Register or Login to view. . The illustrator has captured the proportions, the petals, and the budding to the sides very well. Almost everyone who knows plants says Calendula for this drawing, and it's provisionally at the top of my list. The only reason I am not 100% sure it is Calendula is because there is a mountain aster that is shaped almost exactly like Calendula, but which has blue flowers instead of yellow. I don't think the flower colors in the VMS are literal. Blue has been chosen for most of them including plants that don't usually have blue flowers, so there's no way to know if it is Calendula or a blue aster without some other clue (e.g., in the text).
  

There are three advantages in selecting these plant drawings:

• Both are reasonably naturalistic and reasonably well drawn.
  
• Both Centaurea and Calendula are commonly included in medieval plant books (both illustrated and unillustrated). They were considered medicinal plants. 
  
• Both Centaurea and Calendula have relatively stable names (most of the time this is what they are labeled, and the medieval spelling for these plants did not vary as much as other plants), which might be easier to deal with than Cannabis, which had many different names/spellings.
  

Caveats

I am strongly of the opinion that the VMS text is not a one-to-one substitution code. But, I am perfectly happy to point to resources if they are useful to anyone for looking for relationships between the drawings and text (whatever the text may be).

For what it's worth, when I was attempting this with a possible verbose cipher (and was spectacularly unsuccessful), I settled on:
You are not allowed to view links. Register or Login to view. Centaurea You are not allowed to view links. Register or Login to view. Lilium You are not allowed to view links. Register or Login to view. Viola You are not allowed to view links. Register or Login to view. Malva

I KNEW I was making huge assumptions with each and every one of these -- but I had to start somewhere.  I was careful to document each of the assumptions I made at each step, so I would not forget all the possible places for a misstep my decisions had made.

On the upside, going through the exercise really made it abundantly clear the issues involved.  Even when using combinations of cipher types that take into account issues with the text (e.g. cipher approaches that could accommodate the low entropy, the word length peculiarities, etc).

I'm not saying I'm entirely convinced of these IDs (or these particular names, even if the ID is correct) -- but these were better than average, in my opinion.

I am not in favor of any particular species per se. I think we need to examine the strongest candidates because they present the most probable opportunities for a positive comparison.  Obviously the correct VMs plant identification is necessary before text comparison can start, and all we have to go on there is the VMs illustration. The text comparison needs to be based on plants of the same identification, apples to apples, not petunias to pistachios. It is also important to have good historical text sources *for* comparison.

The second comparison is totally based on the nature of the text, as it is found in the VMs. Is it all one section, long or short, or is it two sections, or is it three? What other clues can be derived from the VMs text?

If this doesn't work for the 'pick of the litter' candidates, it will be less likely to work for other choices. If the doesn't work for the best possibilities, then finding linguistic connections in the botanical section is probably not feasible, and investigators of comparative texts will need to look elsewhere.