The Voynich Ninja

(28-03-2021, 04:47 AM)mscheo Wrote: You are not allowed to view links. Register or Login to view.Well, I've just finished my first round transcription of the VMS, with many "a/o" and "r/s" marked in red. (Gosh, some of them are really difficult to differentiate.) And whenever I hit a group of duplicates, triplicates or even quadruplicates vords, I marked them in green and wondered - what words could they be? 

Duplicates are easier to explain. Besides the is-is, that-that, so-so, it also happens whenever a word (particularly a noun) ends a sentence and starts a new one, like "It has to be You/Elephants. Elephants/You are the greatest."

Hi mscheo,
of course, in most languages it is possible to create sentences with repeated words. The problem is quantitative. Duplicates amount to almost 1% of Voynichese. In actual European texts, they are about one order of magnitude rarer or more.
This plot shows the % of exact reduplication and quasi-reduplication (consecutive words which only differ by 1 character) in Voynichese EVA transliteration (Currier A/B) vs various European languages from tYou are not allowed to view links. Register or Login to view.:
CZE Czech
ENM Middle English
FIN Finnish
FRM Middle French
GMH Early New High German
GRC Ancient Greek
ITA Italian
LAT Classical Latin
OCI Old Occitan
ORV Old East Slavic
OSP Early Modern Spanish

Brian's corpus and the interesting case of the Finnish Kalevala were pointed out by Jonas Alin You are not allowed to view links. Register or Login to view..

[attachment=5413]

As you can see, the distance is abysmal.
Vietnamese (VIE) shows that some Eastern languages are better fits for reduplication in the VMS (though not for quasi-reduplication).

Of course, similar comparisons only make sense under the assumption that each Voynichese word corresponds to a word in some plain language. As discussed by You are not allowed to view links. Register or Login to view., this assumption is supported by evidence like Zipf's law and MATTR values, but such evidence is not conclusive and it could be that Voynichese does not work that way.

(28-03-2021, 04:47 AM)mscheo Wrote: You are not allowed to view links. Register or Login to view....

I believe Voynichese will have its fair share of (ancient) idioms/figure of speech/metaphor/hyperbole. Unless a Rosetta Stone with Vords is discovered, I doubt the VMS could be decoded fully by one person. Instead, through the works of many, it would come in fragments, dozens of words here and there, then hundreds and then thousands, over the course of time, by different folks. So, along the course of your work, you might have translated/substituted/decoded some vords that seem strange or out of place. Don't discarded them yet. Those vords could be figure of speech or hyperbole that need to be understood in their proper context. Share them here and let everyone tinker on them.

Also, it would be helpful to find out what are the common figure of speech/metaphor/hyperbole used in botany/herbal/medical/astrology in the 14th/15th century Latin world. Any one got any to share?

While I agree that Eastern languages could explain the high reduplication rate, I think that Latin cannot be word-by-word mapped into (or from) Voynichese.
One of the problems is that the most frequent word in Voynichese (daiin) often occurs in consecutive repetitions. On the other hand, the most frequent Latin word 'et' can never appear consecutively: it can only appear between two words/phrases belonging to the same category ("creavit deus caelum et terram") or at the start of the sentence ("Et tu Brute?").

A kind of repetition that you sometimes find in medieval Latin texts are magical/religious invocations (some examples were listed You are not allowed to view links. Register or Login to view. by Searcher). But these:

• typically involve rare words
  
• are not nearly as frequent as what we see in the VMS
  

You probably know about that exchange involving You are not allowed to view links. Register or Login to view.:

Quote:Question (Speaker not identified): How do you account for the full-word repeats?
Currier: That’s just the point — they’re not words!

I partially agree with his position:

• either Voynichese words do not correspond to words in a natural language
  
• or the underlying language is not an ordinary (Latin / Saxon / Slavic / Greek) European language
  

Differently from Currier, I tend to favour the second option. Since you appear to be familiar with Chinese and other Eastern languages, I hope you will help us in the search for texts that parallel these fascinating features of Voynichese.

So I stick to the linguistic templates.

"da da da"
It looks funny, but it's used every day in Alemannic.
For example:
du da da, de da da, di da da, das da da.
That's how it's called:
Du da = You there
di da = this da, these da, or " these there, now the "da" can stand for da, there or here.
de da = the da, this da.
So "da da da" means exactly this here, or there but also da

And when I see Chinese, it seems Spanish. Then something has gone extraordinarily wrong.


So halte ich mich an die sprachliche Vorlagen.

"da da da"
Sieht zwar komisch aus, ist im alemannischen täglicher gebrauch.
So auch:
du da da, de da da, di da da, das da da.
So heisst:
Du da = Du da, oder dort oder hier.
di da = dieses da, diese da, oder "diese dort, jetzt kann das "da" für da, dort oder hier stehen.
de da = der da, dieser da.
So heisst " da da da" genau dieses hier, oder dort aber auch da

Und wenn ich chinesisch sehe, dann kommt wir das spanisch vor. Dann ist nämlich etwas ausserordentlich schief gelaufen.

Aus dem deutschen Lehrbuch:
Du da, der Du die da zur Frau nehmen willst, und Du da der Du den da zum Manne nehmen willst.

From the German textbook:
You there who want to take her as your wife, and you there who want to take him as your husband.

(28-03-2021, 08:16 AM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.

• either Voynichese words do not correspond to words in a natural language
  
  
• or the underlying language is not an ordinary (Latin / Saxon / Slavic / Greek) European language
  
  

The third option, of course, is that the natural flow of the plain text is simply not preserved in the ciphertext.

(28-03-2021, 11:03 AM)Anton Wrote: You are not allowed to view links. Register or Login to view.

(28-03-2021, 08:16 AM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.

• either Voynichese words do not correspond to words in a natural language
  
  
• or the underlying language is not an ordinary (Latin / Saxon / Slavic / Greek) European language
  
  

The third option, of course, is that the natural flow of the plain text is simply not preserved in the ciphertext.

And the manner of that disruption (either by necessity (preferred!) or scribe’s choice (ugh!)) results in the observed repetitions.

I just discovered this thread while opening a new one on repetitions, on which you dealt intensively.

My question has first an assumption made: what if the repeating sequences relate to usage of numbers like a thousand thousand, or a million million, or more appropriately something like two two to say twenty two?

In this respect, as a set, repeating words may exhibit an affinity one to the other and two sequences XX and YY would make it more likely that XY appears in the text, that is, if X and Y are digits. How would you approach affinity between two words that tend to repeat and happen to come next to each other given the context?

(28-02-2023, 08:08 AM)Arichichi Wrote: You are not allowed to view links. Register or Login to view.In this respect, as a set, repeating words may exhibit an affinity one to the other and two sequences XX and YY would make it more likely that XY appears in the text, that is, if X and Y are digits.

The most frequently reduplicated words are chol and qokeedy, but these words appear next to each other only twice:
1 qokeedy chol
1 chol qokeedy

For the next few words on the list of most reduplicated, the numbers of appearances next to each other are not particularly high, with some exceptions:
-y q- (there is a known affinity between these words)
chol daiin
chedy daiin

(28-02-2023, 11:03 AM)nablator Wrote: You are not allowed to view links. Register or Login to view.For the next few words on the list of most reduplicated, the numbers of appearances next to each other are not particularly high, with some exceptions:
-y q- (there is a known affinity between these words)
chol daiin
chedy daiin

Thanks for the quick reply. I want to add that in this case, it wouldn't be valuable to disregard cases that appear only once and am ready to look into it further, maybe find a way to group words based on shape and see if there are clues for a system that uses repeating 'i' as a counting system.

(28-02-2023, 08:08 AM)Arichichi Wrote: You are not allowed to view links. Register or Login to view.what if the repeating sequences relate to usage of numbers like a thousand thousand, or a million million, or more appropriately something like two two to say twenty two

I think that's unlikely. Quanitites like one million (a thousand thousand), let alone greater than that, would hardly be expected in any treatise of this time. Anything like "two-two standing for four" would be expressed as two-by-two or two-and-two or twice-to, and thus would not result in exact reduplication. "Two two for twenty two" - I think this positional expression is out of scope of the medieval mindset.

By now I'm inclined to think that reduplication is likely to be a product of some kind of shuffling. If I'm not mistaken, this aspect has not yet been researched in any detail. Let's say transpose the VMS text paragraph-wise and see what happens with reduplication and word pair affinities in general - has anybody done at least an exercise that trivial?

[attachment=7215]

You are always thinking too far ahead.
If you would just look at simple applications, the word "daiin" can also be read as "d'aiin".
The "8" is not only a letter, it also shows an article.
In Alemannic, but also in French, the article is sometimes attached to the word. In the French case, it is now separated by an apostrophe (d'blabla). It would be interesting to know when French introduced the apostrophe.
In Alemannic, the grammar distinguishes it (dBlabla).
So you only see it when you understand the language, or 2x the 88 occurs at the beginning.  German. Dachs, dDachs. / dAchs.
Goes off like dSau.

The same goes for "o". It can be the beginning of a word, or an article. "afrow" today, a woman. dfrow, today the woman or this woman.
Definite and indefinite article.

The VM follows precise rules! You just have to understand it.


Ihr denkt einfach immer viel zu weit.
Würde man einfach einmal auf einfache Anwendungen schauen, so kann das Wort "daiin" auch als "d'aiin" gelesen werden.
Die "8" steht nicht nur als Buchstabe, sie zeigt auch einen Artikel.
Im alemannischen, aber auch im französischen ist der Artikel manchmal an das Wort gebunden. Im französischen Fall, wird es heute durch ein Hochkomma getrennt (d'blabla). Wäre noch interessant zu wissen wann das französisch das Hochkomma eingeführt hat.
Im alemannischen unterscheidet es die Grammatik (dBlabla).
So sieht man es erst wenn man die Sprache versteht, oder 2x die 88 am Anfang vorkommt.  Deutsch. Dachs, dDachs. /  dAchs.
Geht ab wie dSau.

Das gleiche gilt für "o" Es kann der Anfang eines Wortes sein, oder eben auch ein Artikel. "afrow"  heute, eine Frau.    dfrow, heute die Frau oder diese Frau.
Bestimmter und unbestimmter Artikel.

Das VM folgt genauen Regeln! Man muss es nur verstehen.