+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Theories & Solutions (https://www.voynich.ninja/forum-58.html)
+--- Thread: Why and how the text could be Bavarian (/thread-5312.html)
RE: Why and how the text could be Bavarian - ReneZ - 02-02-2026
This is absolutely correct.
Of course there are many possible ways to do this, and doing this causes Voynich words to become considerably shorter, so at best parts of words or syllables.
A recent attempt I made along these lines resulted in character and bigram entropies that are smack in the middle of Latin, Italian and German (but alas, and needless to say, did not result in meaningful text).
RE: Why and how the text could be Bavarian - obelus - 02-02-2026
(01-02-2026, 11:31 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Someone once wrote an entire novel in English (or French, not sure) without using the letter "e" even once."You are not allowed to view links. Register or Login to view." by Ernest Vincent Wright, 1939.
RE: Why and how the text could be Bavarian - JoJo_Jost - 03-02-2026
Cipher Part 2. Work in Progress
Assumptions
Why these cipher structures could also fit Bavarian
Let’s look at the two possible prefixes qo / o. We observe:
After qo, a Gallow follows in almost all cases (under ~90%). After o, this is much less strict (around ~60%).
After qo (and almost always after o), exactly one Gallow follows immediately — never two, almost never three. In general, multiple Gallows within a single word are rare.
Gallows tend to occur near the beginning of words, especially if we treat qo and o as prefixes.
Thesis: In Standard German, articles appear as separate words before nouns. In spoken Bavarian, however, articles are often cliticized to the following word (i.e., written onto it in informal spelling).
Examples: das Haus (“the house”) → s’Haus (with "da" omitted); die Katze (“the cat”, feminine) → d’Katz.
English shows that a single function word like “the” can cover a broad range of contexts.
If qo / o are prefixes, they likely form a function-marker class. This class might cover not only article-like clitics (d’, s’), but also frequent bound elements such as g’ in g’schaut (“looked”), or linking particles / conjunction-like material (e.g., “and”).
This still needs to be disentangled: which marker corresponds to which function.
One possibility is that o is a more generic function-marker, while qo may encode a compound (e.g., “and + determiner”, or a fused prepositional phrase such as in dem / an dem / zu dem / von dem). In English terms: “in the”, “at the”, “of the”, etc.
This kind of fusion is also part of formal German spelling: "von dem" is officially written as "vom" - a good cultural parallel for “function material merging into the word”.
When function words (articles, prepositions, conjunctions) are not written as separate tokens but are attached to the following word, the following happens: the many extremely short words a normal German/Bavarian sentence constantly uses (“der/die/das”, “in”, “zu”, “von”, “und”, “im”, “am”, “vom” …) disappear as a separate word-length class. They are not gone – they are simply embedded in the prefix area of the next word.
Thesis: Gallow structure and consonant clusters
In historical scribal practice (and sometimes also in cipher practice), word-initial vs. word-final positions can be treated differently. A simple illustration is the long ſ vs. short s: long ſ tends to occur at the beginning of words/syllables, while short s often occurs at the end.
Applied to Bavarian/Upper German: many words allow hard consonant clusters at the beginning (which I treat as candidates for some Gallow-classes), while endings are often more reduced / less salient and could be handled by “softer” body ciphers (interestingly, also written smaller), e.g., EVA d y r m, etc.
Typical Bavarian initial clusters include: pl-, kl-, tr-, kr-, dr-, scht-, zt-, pf-, kn-.
Word endings, by contrast, are frequently reduced (phonologically and orthographically).
This aligns well with the idea that Gallows preferentially occur near the start of words (after function-prefixes such as qo/o), and it fits the overall positional behavior seen in the VMS.
Thesis: qo behaves more like an article/determiner or fused prepositional marker, whereas o behaves more like a general linker / function-marker.
And this ties the first and second theses together:
qo (determinative / prepositional complex?) seems to “pull” a hard class-marker (a Gallow) very strongly - (often in noun-like contexts)
o (more general function marker) is more flexible: it often precedes a Gallow (~60%), but it also frequently precedes non-Gallow material (i.e., “softer” word bodies).
RE: Why and how the text could be Bavarian - JoJo_Jost - 03-02-2026
Positive autocorrelation
This relative scarcity of short function-word tokens could also help explain parts of the positive autocorrelation of word lengths in the VMS (long-long-short-short instead of long-short-long-short). In normal texts, articles/prepositions/“and” as separate tokens naturally create a long-short alternation. If such function material is instead attached to the following word (consistent with qo as a function marker, and to a lesser extent o), that rhythm disappears: several words in a row become “bloated” or remain “bare” – which yields length clustering.
Especially since in the standard EVA transcription, higher qo-density is associated with measurably stronger word-length clustering (a positive correlation).
RE: Why and how the text could be Bavarian - JoJo_Jost - 04-02-2026
Here is an observation that also supports the theory that the text could be Bavarian
The actual statement:
As we know, the VMS shows a recurring surface pattern: there seem to be certain word stems that occur with very high frequency, while the beginnings of these words vary (possible prefixes). This leads to visible clusters of related word forms that behave like families.
Interestingly, this also exists in German/Bavarian (and other Germanic languages). One could debate whether it is just as extreme there, but I will leave that aside for now to first clarify what I mean. The following is not intended as a translation, but as a derivation to make it understandable to someone who does not speak German.
Introduction:
I noticed that aiin can also be read as ann: the two ii as strokes of the first n, and the final glyph, as in EVA with its typical curve, also as n → “ann”.
In German (see also Ortloff von Baierland’s text above), "und" (English and) is often written vnnd, with two "n". Since "a" and "u" are not linguistically stable in German, the step to "annd" is small. If one writes annd and shortens it phonetically, one arrives at "ann".
In this respect, aiin could be “und”, which would fit its frequency (note that this is not a claim that this is actually the case, but serves a different purpose).
Then daiin = "dann",( engl: than) the most common word in the EVA transcription.
"Dann" is also used as a sequencing marker in German: "dann mache das, dann koche es, dann seihe es ab, dann …" (engl: then do that, then cook it, then strain it, then.."
It could therefore occur very frequently, especially in recipes.
(But this is not a translation either — it is far too early for a translation; it may be the case, but it may just as well not be.)
If we then look at the most common words containing aiin:
daiin → dann
aiin → ann
raiin → rann
qaiin → qann
olaiin → ol + aiin → ann
okaiin → okann
otaiin → otann
chaiin → chann
shaiin → shann
ykaiin → ykann
oraiin → orann
arain → arain
ykaiiin → ykainn
okaiiin → okainn
daiiin → dainn
qokaiin → qokann
olraiin → ol + raiin → rann
ykaiiin → ykainn
okaiin → okann
This results in many possible German words (without taking possible prefixes into account):
kann, rann, hann, dann, bann, etc.
This is typical for German: the root of the word remains stable, while the initial sound changes, producing a completely different word.
This also applies to other stems. Another strong example is "ein", Bavarian "ain" (Engl: "a"). In this context, "ain = ein". And here too, there are many words built with "ein" (in fact, even more than with ann): ein, dein, mein, sein, kein, nein, rein, fein, klein, gemein, allein, Stein, Bein, Wein, Schein, Schrein, Lein, Hain, Pein, Schwein, gedeihn, leihn, schrein (Again, I do not claim that "ain" = "ein", at least not yet.)
Beyond these, German shows several further highly productive core families built around stable endings with variable onsets or prefixes:
Examples:
-und (und, rund, Mund, Bund, Hund, kund, Pfund, Stund, Schlund, Grund, gesund, wund, Schwund, Befund, Verbund, Fund),
-and (Hand, Land, Sand, Wand, Rand, Band, Pfand, Stand, Brand, Tand, Gewand, Verstand, Bestand, Zustand, Abstand, Beistand, Einwand, Aufwand, Umstand),
-ind (sind, Kind, Rind, Wind, blind, lind, geschwind, Grind, find, bind, wind, schind, empfind, verschwind),
-ang (lang, Gang, Sang, Drang, Klang, Zwang, Fang, Hang, Rang, Strang, Anfang, Empfang, Umfang, Abgang, Gesang, Belang,)
-ing (Ding, Ring, ging, fing, sing, spring, bring, gering).
-och (doch, noch, hoch, Loch, Koch, Joch, roch, kroch, Docht, focht, pocht, horch, Moloch, gepocht, kocht, verfocht),
-acht (Nacht, Macht, Schlacht, sacht, Wacht, Pracht, Fracht, Tracht
-icht (nicht, Gericht, Licht, Dicht, Gicht, Sicht, Wicht, Pflicht, Schicht, Gericht, Gewicht, bricht, spricht, sticht, flicht, schlicht).
-all (all, Ball, Fall, Stall, Knall),
-ell (hell, schnell, Quell),
-eit (Zeit, weit, seit, Streit).
These examples simply illustrate a productive surface mechanism: stable cores supporting large families of forms through variable onsets.
Conclusion
The VMS word structure—strongly regulated suffixes following variable prefixes—fits well with German (and Bavarian) usage, to a markedly greater extent than with other languages, including English, and even more so in comparison with the Romance languages (which operates in the opposite direction).
RE: Why and how the text could be Bavarian - Rafal - 04-02-2026
I would say you sound quite convicing when talking about Bavarian
On the other hand if it is really Bavarian, why nobody was able to crack it?
Especially why nobody at the court of emperor Rudolf was able to crack it. Rudolf lived in Prague but he didn't speak Czech. Rudolf and his entire court spoke Austrian-Bavarian German. And they probably lived less then 200 years after Voynich Manuscript creation and the language hasn't changed so much then.
So that would basically mean that they weren't able to recognise their own language written with another letters.
--------------
Personally I don't believe that it is Bavarian. I am close to nonsense/glossolalia theory, eventually mostly nonsense theory or constructed language theory.
But when it comes to the place of origin I am mainstream, believing it was somewhere around southern Germany, Swirtzerland and norhern Italy. So for me it is quite possible that the scribe knew Bavarian or even spoke Bavarian in his private life.
Therefore if it is gibberish then it could be Bavarian inspired gibberish
Compare Lingua Ignota of Hildegard of Bingen ( You are not allowed to view links. Register or Login to view. ) which is quite clearly inspired by German.
-------------
But I don't discourage you from pursing the Bavarian theory. It is actually quite an interesting read.
RE: Why and how the text could be Bavarian - JoJo_Jost - 04-02-2026
(04-02-2026, 02:22 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.So that would basically mean that they weren't able to recognise their own language written with another letters.
...
But I don't discourage you from pursing the Bavarian theory. It is actually quite an interesting read.
First of all, thank you very much for the positive feedback.
That's rare here in the forum...
I also speak German and can read Bavarian, have extensive knowledge of the German language and still can't decipher it! At the moment, I'm just trying to create an encryption that generates a structure similar to VMS based on a Bavarian text.
However, I think the problem could be the following:
if the gallows stand for groups of letters, then a lot of information is lost. If ‘ch’, “sh” and perhaps even ‘eva d’ and
other characters are structured in a similar way, even more information is lost. I have tried it with a few chiffres. The problem then is the reverse translation... To do this, you need to know:
1. which clusters are determined by the gallows and the other letters.
2. which vowels are swapped and how.
3. understand the trick with qo / o and other prefixes
4. solve the problem with the y at the end.
5. And most importantly: you probably also need to know what the text is actually about – simply because too much information is destroyed.
Not to mention that it may also involve a very specific dialect - including words, we dont know today.
That's also why I'm not sure it can be solved at all. Probably in a few years with a "super AI" that runs through all these possibilities and then finds the most suitable words for the whole book.
I'm afraid this task might be too complex for a human being.
However, due to a structure similar to Bavarian, I dont believe it is a hoax. But of course, I cannot rule it out yet.
RE: Why and how the text could be Bavarian - Petrasti - 04-02-2026
What currently speaks most strongly against an artificial language or a string of meaningless letters and words for me is that we find prefixes in the manuscript such as d, ch, which in turn combine with the base words, thus creating words like ddor, chcheey. However, we don't find this even once with the gallow signs. There is no kk, pp, tt, ff. Why would one deviate from a system if one wants to create meaningless words? Instead of writing the gallows symbol twice, spellings like kaiin to kokaiin are used to place the gallows sign before the base word without creating a double letter sequence.
RE: Why and how the text could be Bavarian - JoJo_Jost - 04-02-2026
(04-02-2026, 09:26 PM)Petrasti Wrote: You are not allowed to view links. Register or Login to view.What currently speaks most strongly against an artificial language or a string of meaningless letters and words for me is that....I agree, for I also notice peculiarities from time to time that argue against a random combination of glyphs. I should write them down and collect them...
For example, this arrangement of prefixes and word stems is too similar to language (maybe german) to be a coincidence.
RE: Why and how the text could be Bavarian - DG97EEB - 04-02-2026
(04-02-2026, 09:26 PM)Petrasti Wrote: You are not allowed to view links. Register or Login to view.What currently speaks most strongly against an artificial language or a string of meaningless letters and words for me is that we find prefixes in the manuscript such as d, ch, which in turn combine with the base words, thus creating words like ddor, chcheey. However, we don't find this even once with the gallow signs. There is no kk, pp, tt, ff. Why would one deviate from a system if one wants to create meaningless words? Instead of writing the gallows symbol twice, spellings like kaiin to kokaiin are used to place the gallows sign before the base word without creating a double letter sequence.
It very clearly has very strong grammar and positional elements. You can computationally map which elements go together and which are essentially forbidden combinations. Add to that features like LAAFU and the Currier distinction, and it's clear there's something going on (Stolfi is the expert here).. but none of that means it's language. There are too many other confounding factors, and it's not necessarily a cipher either. I think every cipher since 1400 has been thrown at this and nothing works. I've built 30 generators that can get very close (including the deep grammar), but none based on plain text and all based on pure generation. I've even simulated mixing Cardan grilles and volvelles taking Rene's paper as a starting point or Grescko's, and nothing touches it... So my conclusion is it's highly likely not to encode plain text and is simply generated.. but I can't prove it any more than anyone else. Sadly statistics alone only take you so far.