The Voynich Ninja

(31-01-2026, 07:17 PM)JoJo_Jost Wrote: You are not allowed to view links. Register or Login to view.Can you elaborate on that? Perhaps that will help me.

Voynichese glyphs have a tendency to appear in words in a semi-ordered manner (defined by a grammar or some probabilistic system), this is a problem if you keep the same order. A limited re-ordering might be possible: a little ambiguity is OK, for example Latin abbreviations are ambiguous but there are rules so it's not too difficult to guess the meaning in most instances.

However, as any language can be writ ten with syl la bles se pa ra ted by spa ces, or 2-3 sylla bles toge ther sometimes to allow longer pseudo-words, and a little phonemic reduction could be done without making the language impossible to understand (again, a little ambiguity is OK), I don't see the necessity of supposing that Bavarian is the target language.

Quote:As already written here, I even suspect that it could be incredibly difficult to decipher VMS, precisely because of the many possibilities and variants of Bavarian, including, among many others, those pointed out by Stefan here

That's fair said.

If you ever try to read Voynichese as Bavarian remember that the most common words should be simple words like "and", "or", "with", "by" and so on.
Also Bavarian should have some kind of der/die/das word like standard German, am I right?
Unfortunately we don't have good candidates for such words.

And remember that 50% of Voynichese words end with "9". You won't find any similar letter in Bavarian, I believe.

I have wondered about it once and some possible option could be tricks similar to Pig Latin which I have already mentioned in this thread.
For example in classical Pig Latin 100% of words end with "y"  

You are not allowed to view links. Register or Login to view.
For words that begin with consonait sounds, these are also known as consonant blends (two letters that make one sound: e.g., black, slack, clown). The initial consonant blend (or two letters) is moved to the end of the word, then "ay" is added, as in the following examples:

• "pig" = "igpay"
  
• "latin" = "atinlay"
  
• "banana" = "ananabay"
  
• "black" = "ackblay"
  

In order to explain the word structure in the Voynich MS text, you need to rely on such major changes to a source text, that the differences between German, Bavarian, and in fact other European source languages fade in comparison.
The low conditional entropy (or the bigram entropy) of the Voynich text is a consequence of this word structure, and it is far too anomalous to be explained by small things like swapping characters or removing distinctions between p and b (and a few more).

(31-01-2026, 08:31 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.If you ever try to read Voynichese as Bavarian remember that the most common words should be simple words like "and", "or", "with", "by" and so on.
Also Bavarian should have some kind of der/die/das word like standard German, am I right?
Unfortunately we don't have good candidates for such words.

These small words almost disappear in Bavarian; they are attached to other words or combined with other small words: Here are some more examples.

High German: Hol die Wurzel,
Bavarian: Hoi d'Wurzn,
Phonetic hoid wurzn.
komt es = komts / kimts /kümts
von dem = vonm / vonem / vonim / vom (wie verwenden „vom“ auch im Hochdeutschen)
so ist = sosts / sost / sist / sois / 
in der = inde / inda /inte

(31-01-2026, 08:31 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.For words that begin with consonait sounds, these are also known as consonant blends (two letters that make one sound: e.g., black, slack, clown). The initial consonant blend (or two letters) is moved to the end of the word, then "ay" is added, as in the following examples:

Yes, I suspect something similar in the VMS text. The y's are vowels maybe "i" or "u" in the word that have been moved to the end.

To reiterate.
There are 3-4 features in the German text that indicate it is High German, Bavarian dialect. Nothing more.
What is High German (Oberdeutsch) (Süddeutscher Sprachraum)? It is the Alemannic and Bavarian dialects.
Around Lake Constance, centred on St. Gallen Abbey, High Alemannic, Swabian and Bavarian.
Nevertheless, I do not consider the encrypted VM text to be German; it lacks structure.
In short, I consider it to be Latin, or close to it.
That is what my studies tell me. After all, I have been studying it for over 15 years.
You really have to work at it. No EVA, no transcription file and no PC. The text is individual.

(01-02-2026, 01:15 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.In order to explain the word structure in the Voynich MS text, you need to rely on such major changes to a source text, that the differences between German, Bavarian, and in fact other European source languages fade in comparison.
The low conditional entropy (or the bigram entropy) of the Voynich text is a consequence of this word structure, and it is far too anomalous to be explained by small things like swapping characters or removing distinctions between p and b (and a few more).

Yes, that may well be the case – I have realized how unusual the VMS is.

At this point, one could actually drop out, because more complex ciphers (such as, in my opinion, the Naibbe cipher) would be anachronistic for this period. This dramatically increases the likelihood of a hoax.

Or you could take a different approach. Thanks to Stolfi, I realised that it might make sense to find a language that is almost monosyllabic. Bavarian is probably the only language in Europe that fits this description reasonably well. And that's quite interesting because the marginalia are probably in Bavarian, and the writers of the marginalia may even have been able to write Voynich (or imagined they could ).

Is this a coincidence? Yes, it could be, but it actually fits too perfectly to really be a coincidence. It is so unusual that it MUST be investigated further. In my opinion. I just don't know if I'm the right person for this job, but I speak German and understand some dialects, and as a journalist and writer with nearly 40 years of experience, I know my language reasonably well and have an advanced understanding of it.

So, in step 2, I want to check whether a cipher that fits the early 15th century can produce some of the anomalies of the VMS, for example the low entropy (this is not yet a solution, but a plausibility check).

And I think I'm at least heading in that direction. It'll just take a little more time.

(01-02-2026, 07:29 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.In short, I consider it to be Latin, or close to it.

@ Aga
One does not know. I dropped Latin at some point because Latin is strongly marked at the end. 
The crucial information (case, number, gender, tense, mood, person) is at the end of the word. In short, the end of a word in Latin is highly informative.

With VMS, it's exactly the opposite. Word endings are extremely monotonous. There are very few ending patterns. Endings are repeated mechanically. There is hardly any variation or combination.

Bavarian fits better, it is the other way around: In Bavarian (as in German in general), the root of the word is recognisable early on.
This means that lexical information (what the word is about) is found at the beginning of the word or very early on, and endings are greatly reduced or neutralised. The variation at the end is small or even optional.

And VMS seems to take this to an extreme.
But: Of course, one can imagine codes that press Latin into a different form – one simply does not know.

@ Aga
Man weiß es nicht. Ich habe latein irgendwan fallen gelassen, weil latein stark end markiert ist.

Die entscheidende Information (Kasus, Numerus, Genus, Tempus, Modus, Person) sitzt am Wortende. Kurz das Wortende im lateinischen ist hoch-informativ.

Beim VMS ist es genau anders herum. Wortenden sind extrem monoton. sehr kleine Menge an Endmustern. Endungen wiederholen sich mechanisch kaum Variation, kaum Kombinationen.

Im bairischen ist es andersherum: Im Bairischen (wie generell im Deutschen) ist die Wortwurzel früh erkennbar. Das heißt: Lexikalische Information (worum es geht) steckt am Wortanfang oder sehr früh und Endungen sind stark reduziert oder neutralisiert. Die Variation am Ende ist gering oder sogar optional.

Und VMS scheint das noch auf die Spitze zu treiben.

Aber: Natürlich kann man sich Chiffren vorstellen, die Latein in eine andere Form pressen - man weiß es einfach nicht.

[attachment=13868]

1. As far as p/b is concerned, Rene is certainly right. There are plenty of examples:
Here, Blau. ‘Pulver’ – what is a “u” and what is a ‘v’?
Die Rabenschlacht, University of Heidelberg. There are so many changes. p/b, d/t, u/v/w. Unbelievable.

2. Without work and trial and error, you'll never get anywhere. You can sit on the merry-go-round and spin around until you feel sick. But you'll still be in the same place.

3. The proof of the pudding is in the eating. Rubbish in, rubbish out until you get a result.
[attachment=13869]

That's a VM word ‘4o89’. How many possibilities are there?

(01-02-2026, 01:15 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.The low conditional entropy (or the bigram entropy) of the Voynich text is a consequence of this word structure

The character entropy is not significant because it depends on the spelling system, which we do not know.  The low entropy does not imply "it is not European".  It could be simply the result of many phonemes being encoded with 2 or 3 EVA characters (like German using "sch" for the "ʃ" sound).  


All the best, --stolfi

Character entropy is a major problem if you can't come up with a spelling system that would account for it.