The Voynich Ninja

Moving on, I’ll try to explain it as clearly as possible. Using two dictionaries and Google, and always aiming for perfect spelling wherever possible.
The aim is for everyone to be able to understand and follow what’s happening and how it’s structured, and even to form sentences if they wish.
It doesn’t get any better than this.

This cannot be captured in the statistics in this way, as ‘4’ = (qu) and is therefore already a combination. In addition, ‘o’ is a single character. There are also three other combinations.
This is how it works throughout the entire VM.

(27-08-2026, 06:45 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.I’ll explain it very simply.
You don’t know all the different things ‘qui’ can stand for. See the dictionary.
I’ll also explain to you that you can keep your statistics to yourself, because they’re worthless.
I’ll also explain that you’re just whingeing and arguing against Latin, but never give any examples of German (Bavarian). Apart from your statistics, which are worthless and prove nothing.
What is ‘4o’ supposed to mean in German?
Keep your statistics to yourself and provide facts.

My friend, keep your composure...

This is how people react when they feel deeply unsettled...

If you call such simple statistics worthless, then you're right—we don't need to discuss this any further.

Okay, let’s do it without statistics if you don’t like them  . 

You say “aiin” is “utis” and its variations? That piqued my interest. I therefore counted the endings you appear to be treating as variants, purely as letter sequences and irrespective of grammatical context.

Percentage of all words/tokens in the Latin corpus that end with this sequence of letters.

-itis 354 0.277%
-atis 314 0.246%
-etis 317 0.248%
-ati 196 0.153%
-iti 66 0.052%
-otis 48 0.038%
-uti 39 0.031%
-oti 29 0.023%
-utis 23 0.018%
-eti 15 0.012%

so: In my Latin biblical corpus they occur in 1.10% of all word tokens combined.

As an independent check, I ran exactly the same count on the Latin ITTB corpus (334,471 alphabetic word tokens, largely Thomas Aquinas). There they account for only 0.643%.


Total “aiin” sequence in the VMS: approx. 3,948 occurrences 10.2 % (!!!) of the tokens


Show me a reasonably sized continuous Latin text in which these endings even reach 3% of all word tokens.

(27-08-2026, 08:13 AM)JoJo_Jost Wrote: You are not allowed to view links. Register or Login to view.so: In my Latin biblical corpus they occur in 1.10% of all word tokens combined.

What do you mean by that?
Are you talking about a passage from the Bible?

I thought that someone posted a clip from a manuscript in Spanish, Catalan, Portuguese, or French. But I cannot find that post, so I may have hallucinated again.

Just in case, let me note that "que" is a very common word in those languages -- as a pronoun like "what" or "who" or "that", or a conjunction like "that".

"Quiero que hables" = "I want that you speak"

"Le livre que tu lis" = "The book that you are reading"

"Que sera, sera" = "Whatever will be, will be"

"El hombre que se fue" = The man who left"

"No sè per què" = "I don't know why"

Etc.

(27-08-2026, 09:02 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.

(27-08-2026, 08:13 AM)JoJo_Jost Wrote: You are not allowed to view links. Register or Login to view.so: In my Latin biblical corpus they occur in 1.10% of all word tokens combined.

What do you mean by that?
Are you talking about a passage from the Bible?

No, that was just a part of it: 7,343 verses, ~127,600 words.

I took now the entire Bible (printed in 1454) 35.811 Verse, 612.005 Words

-atis 1,530 0.2500%
-itis 1,360 0.2222%
-ati 1,208 0.1974%
-etis 877 0.1433%
-iti 315 0.0515%
-uti 234 0.0382%
-utis 164 0.0268%
-otis 85 0.0139%
-eti 73 0.0119%
-oti 66 0.0108%
Total 5,912 0.9660%

Nothing changes.

And 334.471 words, mostly Thomas Aquinas (as a “learnblock”), i.e., 13th-century medieval Latin.
Those are the two corpora I have readily available at the moment.

A total of nearly 1 million words

What kind of Latin are you looking for?

But as I said, it would be easier if you gave me an example where the frequency of the glyph sequence you mentioned reaches 10 percent—or even just three...

Because no matter what I give you, you won’t accept it.
So I could give you 1,000 counterexamples, and even then it wouldn’t be conclusive for you.

In short: You should just provide me with a text where your assumption holds true, because so far it doesn’t look good for your “aiin.”

I already knew what you were going to write before you’d finished.
I was really keen to get started, but I held back.
“I will, I am, and I was.”
You’re comparing the text of the Bible to a cookery book.
The Bible tells stories, most of which are set in the past.
Instructions are in the present or future tense.
Didn’t you know that the person, as well as the tense, changes the endings or the whole word?
The human factor.
Who writes the Bible, and who writes a notebook?
Now I’ll just wait and see how you intend to explain that in Bavarian. Because the problem with person, tenses and endings is the same in most languages.
I’m looking forward to it, because the Bible text is also available in German.
Words such as ‘benutzen’, ‘verwenden’, ‘schälen’, ‘kochen’, ‘reiben’, ‘rühren’ and also ‘nimm’, etc., you’ll rarely find in the Bible.
As I said, I’m looking forward to your explanations on the same issue.

@ Aga

realy? That is your defence?
All you do is make accusations and assertions—no evidence. That’s really very scientific!

But okay, you want a typical medieval recipe text. No problem. Let’s take the *Antidotarium Nicolai*. For centuries, this work was considered the standard pharmaceutical reference. It is a recipe book. It was originally compiled in the early 12th century and is attributed to Nicolaus Salernitanus.

In terms of content, it’s a collection of about 150 compound medicines. Typically, an entry consists of two parts: first, it describes the diseases or ailments the remedy is intended to treat, followed by the pharmaceutical instructions for preparation.

So, exactly what you asked for.

-atis  48  0,275 %
-ati   38  0,218 %
-itis   28  0,160 %
-utis  20  0,115 %
-eti   19  0,109 %
-iti    12  0,069 %
-uti   70, 040 %
-etis  40, 023 %
-oti   20, 011 %
-otis        00 %

Summe 178   =  1,019 %   Off by a factor of 10!

The corpus is, of course, much smaller, but again, it’s 1%.

I don’t know, Aga, but honestly—your theory about “aiin” = uti / utis and variants is clearly refuted by this.

You probably have reservations now, too, but since you constantly emphasize the “scientific nature” of your work—from a purely scientific standpoint, it’s now up to you to find a text that reaches 10 percent.

Until then, your theory is considered refuted.

P.S.: Regarding your comments on my Bavarian theory. I'm not going to play along with these little kindergarden games. This thread is about your Latin theory, not my Bavarian one. You claimed that your Latin theory was obvious and scientifically provable—I simply said that it couldn’t be Latin, at which point you started attacking me fiercely. From that point on, I defended myself. Otherwise, I wouldn’t even be going to all this trouble here, because it’s really too obvious that it can’t be a substitution cipher in Latin. (Which everyone who has studied the VMS in any depth knows, except you.)


Source: You are not allowed to view links. Register or Login to view. (The transcription i used is divided into two parts: Dutch and Latin. The Dutch section and other not latin descriptions were removed, leaving only the Latin text.)

[attachment=17448][attachment=17449]
Take a close look.
In the green section, the ending rarely appears.
But it does appear in an instruction, for example a recipe.
And if someone writes in the second person, for example: ‘We did it this way, and you should do it this way,’ this leads to endings appearing where they normally shouldn’t.
It’s exactly where you’d expect to see it. "-us + -tis
I mustn’t lump everything together.

I do hope you’re familiar with the concept of averages. And the VMS is too large—it has too many sections—to attribute this to individual effects. We’re talking about a factor of 10, this isn’t a minor issue.

So can you provide corresponding evidence? A more extensive original Latin text with a similarly high frequency of your words? If not, your theory remains refuted.

I could provide you with plenty more evidence. But you reject statistics, and now you’re obviously rejecting frequency analyses as well. It’s easy to ignore everything just so you don’t have to give up your theory. It’s a widespread phenomenon—you’re not alone in this.

You know, Aga, I’m not here to disprove your theory; I’m not even really interested in it. I have nothing against you, really, I actually think some of your others ideas are good, (but not the latin chiffre). But anyone who attacks others so arrogantly has to expect them to fight back.

As far as I’m concerned, go ahead and keep searching for the Latin theory for the next 20 years. It’s your life. Time you could spend finding a better, more coherent theory. That’s entirely your decision...

But I’m out here now...