The Voynich Ninja

Here is a link to access Gerard Cheshire's paper about VMS plants:

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

Once I was able to read an entire PDF document.
So I give my criticism right away.
Basically I don't say anything against the work on the plant itself. But ...
Castello Aragonese, Ischia. There is no indication that it comes from the island. Since it is an island, the salt concentration is higher. He will have trouble finding other plants from the VM on the island.
For Mary, of Castile, Queen of the Crown of Aragon.
I have seen books where kings were written for. As commissions or as gifts.
The quality of the VM alone does not allow it, except for the naked ladies.
He'd probably be executed for insult, and then it would be in the history books.
But I can't judge the whole thing by itself. I don't see why not.

This is what Cheshire claims in his abstract (so you know the basic premise):

"Abstract
The plants individually described in Manuscript MS408 have all been identified as species from the environs of the Mediterranean Basin, in accordance with the location of origin for the manuscript. ...'''

My reaction to this:
The majority of medieval plant books are based on exemplars that trace back to this region and thus include a high proportion of Mediterranean plants, so if there are Mediterranean plants in the VMS, it would not be surprising or unusual for the time.

However, Cheshire's claim has nothing to do with research into the history of plant manuscripts, he has chosen this region for the origin of the plants because it is consistent with his theory that the VMS was created by Dominican nuns on the island of Ischia and that it remained there until 1912 "when the Italian government sold [the contents of the library] into private hands".


Purpose of the Plant Drawings


"Within the manuscript there is a series of illustrations of medicinal herbal plants..."

There's no proof yet that the purpose of the plants is medicinal. Some of them do appear to be medicinal (many plants were considered medicinal), but there are also some that were considered to be of low medicinal value in the Middle Ages, so it's possible the VMS plant section serves some purpose other than documenting medicinal plants.

----------------------------------
Structure of the Text

"The algorithmic method, of priority array queuing, was used to translate and identify the words in the text, as described in the following paper: You are not allowed to view links. Register or Login to view. The method takes Latin as the principal source, with Old and Modern Romance as the secondary and tertiary sources."

There's nothing in Cheshire's vitae or posted credentials that indicates he's a programmer (his PhD is in belief systems), and his methodology doesn't appear to have anything to do with "array queuing" as far as I can tell from reading the paper. It's a simple substitution cipher, so I'm not sure he knows what "priority array queuing" means in relation to deciphering text.

----------------------------------
Linguistics

Cheshire is not using the term "proto-Romance" any more, he's calling it "prototype Romance", but his change in vocabulary is still very questionable both in terminology and in terms of time frame. Now he is saying:

"We can see that the language is placed somewhere between Latin and Romance in linguistic evolutionary terms: i.e. it is a vestigial form of prototype Romance."

Latin was already evolving into Romance languages more than 10 centuries before the VMS was created, before the fall of the Roman Empire. I've never seen any historian or linguist say that "vestigial" forms of "prototype Romance" languages still existed in the Middle Ages, and Cheshire has not provided any evidence in his previous papers or in this paper that a "vestigial prototype Romance language" existed in the 15th century.


----------------------------------
"It is apparent that the illustrations essentially function in substitution for the plant names, simply because scientific names were not yet conceived, and common names would have varied regionally."

This is an oversimplification and is, in many cases, untrue. Linnaeus didn't invent the binomial system—he standardized and enlarged upon a system that had been evolving for several centuries. There was actually a surprising amount of consistency in plant names for many medieval plants. Even some of the vernacular names are similar to the Latin or Greek names. For example, many plants with medicinal value had binomial Latin names ending in "officinalis". Those that were good for dying were given binomials ending in "tinctoris". Linnaeus kept many of the old Latin names and some of the more common vernacular names when he developed his system, names like Yris (Iris), Calendula, Arthemisia, etc.

The Plants

1) Cheshire has identified a plant drawing with ten leaves as Paris quadrifolia ( f5r ). I know this is a common ID for the VMS plant, but P. quadrifolia rarely has more than four leaves and it's difficult to find one with five in Europe. Asian species of Paris have more leaves, but they are not called Paris quadrifolia and most of those with multiple leaves are east Asian. But even in east Asia, it's hard to find Paris with 10 leaves.

Medieval manuscripts with drawings inferior to those of the VMS were able to convey a recognizable P. quadrifolia, and always drew it with four leaves, so I don't think there's enough evidence yet to say You are not allowed to view links. Register or Login to view. is P. quadrifolia. There are other plants with whorled leaves and a round seed or fruit at the end of the stalk.


Linguistic basis for the plant ID

Cheshire's argument for the Paris quadrifolia ID is to link the tickmarks on the round "berrylike" fruit at the top of the plant to eyelashes. Then he relates the eyelashes to linguistic terms related to "eye". He says the word "eye" comes from Greek ιλλος and the old Latin word for "eye" is aeo. In manuscripts, I frequently see Latin "oculus" for eye. When I see aeo, they usually mean "aeon" (a time period or era) rather than "eye".

Cheshire further claims that Latin aeo is a phonetic variant of Greek "illos". It's difficult to follow his argument about how you get from Illos to aeo but he lists a multitude of words for eye in numerous languages some of which resemble the Greek or Latin words. I don't think this strengthens his argument for the plant. He is saying that the first phrase on the folio translates as "l'aeo'na" (the eye here).


In terms of methodology, what Cheshire has done is to convert Voynichese into whatever he thinks it might be in his "prototype Romance" language, and then he hunts for a word in a Romance language for each part of the token (often a different language for each part) and claims similarity with that language in order to explain his translation of each portion of the phrase.

He also relates parts of the plant to religious imagery (e.g., "Lazerus of the Four Days").


After this, he mixes up a lot of information about other plant species as though they were simply different names for Paris quadrifolia, but in medieval manuscripts, these are definitely treated as different plants, not different names for the same plant. Thora and Paris are often in the same manuscript, with different drawings and different plant labels for each one. They were not considered the same plant.

After this are pages and pages of information on plant medicine that might be irrelevant if the plant turns out to be something other than Paris quadrifolia or, even if the plant is Paris, it may be included for other reasons.


Confusion between Paris and Aconitum

He refers to Paris as "Leopard's Bane". Paris was not called Leopard's bane. This name was used for a plant called Doronica that looks like a big yellow daisy, or sometimes for one of the species referred to as Aconitum (this name refers to a couple of different kinds of plants, not just the tall monk's hood).

I think I know why Cheshire got it wrong (and why he confused the two species). A picture of Aconitum is sometimes next to a picture of Paris in some of the later plant books. He probably got the labels confused. Or maybe he got it wrong because there's a stock photo on Alamy that's labeled wrong. The drawing is Paris quadrifolia, but it's labeled Aconitum Pardalianches. Contributors to the stock photo sites often know nothing about the images they are uploading. I see wrong labels every day.

How do you tell them apart when archaic drawings are not very accurate? Aconitum Pardalianches drawings have the leaves growing from the base of the stalk (and sometimes include a beadlike root) and the leaves have stems. Paris has the leaves growing a few inches up the stalk (several inches from the ground), the leaves are longer and attached more closely to the stem.

Note how this drawing of Aconitum pardalianches has the leaves growing from the base of the stalk. This is not a drawing of Paris. They are completely different plants but Cheshire thinks they are the same:

   
Left: Aconitum with leaves growing from the base and petioles (stems).    Right: Paris with leaves growing several inches up the stem and attached directly to the stem.


The plant drawings Cheshire used for comparison range from c. 1450 to 1780.


.
The substitution chart for the VMS glyphs is at the end.

• The letter u/v is extremely common in Romance languages, but Cheshire has assigned this letter to one of the VMS rare glyphs.
  
• The letter "q" is also very common in Romance languages and it has been assigned to EVA-f.
  
• The letter q in Voynichese is almost always at the beginnings of tokens, but in Cheshire's chart, it represents the letter "d", which is at odds with where "d" appears in words in Romance languages.
  

In other words, his glyph assignments are very different in both frequency and position from what you would expect in a Romance language.

Prototype Romance   
It's not finished yet, just a prototype.

You are certainly right about the selection of plants to match his preferred area. The confirmation bias is strong in this one...

It can't be the Paris alone.
It is also native to our region, and I live in the Northern Prealps.
So I'll wait and see.

Personally, I don't like pdf files, because I don't know how to translate them quickly, that's why I only read the abstract and references first. The surprise with the article you cite, JKP, is that there is no precedent for identifying the plant drawn in the manuscript, it's the one and only author. This is the point that bothers me, the rest is less important, in my opinion. The works cited in the text do not appear in the references either. I hope that the situation will evolve and that its articles can become more respectful of the rules to be cited.

(29-04-2020, 10:27 AM)Ruby Novacna Wrote: You are not allowed to view links. Register or Login to view.
... The surprise with the article you cite, JKP, is that there is no precedent for identifying the plant drawn in the manuscript, it's the one and only author. This is the point that bothers me, the rest is less important, in my opinion...

I'm not sure what you mean, Ruby, about the plant identification.

Do you mean he's the only one who has suggested this identification? Or do you mean he didn't mention all the other people who suggested this identification (there are quite a few)?


I think the biggest problems with his paper are the following:

• He chose the origin of the plants based on where he thinks the manuscript originated. As Koen pointed out, this is confirmation bias.
  
• He confused two plants as being the same. He apparently can't tell the difference between a plant with petioles and one without, and a plant with a central stalk with a stem-based whorl and one without (these are very basic identification features). Half the paper is about a plant that is different from his plant ID of Paris because he got confused about a plant label online and mistakenly believes that Paris and Aconitum are the same plant. In other words, he included a large amount of irrelevant information about the wrong plant to promote his idea that the plant is Paris.
  
• He makes no effort to learn the correct terms for basic linguistics concepts or even for basic manuscript concepts (terms like "folio", "recto" and "verso"). He's in a hurry. He wants quick results, not correct results. This kind of haste almost always leads to sloppy research.
  
• His "solution" and translations are based on a simple substitution cipher combined with a significant dose of subjective interpretation. Unfortunately, his substitution chart is completely at odds with letter frequencies in Romance languages and he has provided no rationale for this.
  
• If he can't turn his translation into a word, he chooses a different word from a dozen different languages to try to force it to work. If the VMS is drawn from a dozen different languages, as he suggests, then it's a one-way cipher based on subjective selection and would not be translatable by Cheshire or by anyone else. Every person trying to use his system would get different results.
  

There are other problems, historical inaccuracies, logic problems, etc., but the points above need to be dealt with first before you can even get to the deeper problems.

What I don't understand is the focus on this specific paper. Isn't this paper something like 1 of 8 plant papers that he has written, in addition to his other papers, but this one seems to have attracted here significantly more interest than those other papers?

Think of it as a case study.

(29-04-2020, 11:37 PM)Mark Knowles Wrote: You are not allowed to view links. Register or Login to view.
What I don't understand is the focus on this specific paper. Isn't this paper something like 1 of 8 plant papers that he has written, in addition to his other papers, but this one seems to have attracted here significantly more interest than those other papers?

There aren't enough hours in the day to read them all at once. As for the previous nonplant papers, I have posted critiques both here and on my blogs.