The Voynich Ninja

dvallis Wrote:I did not at the time, LoL.

The VMS definitely has its share of learning curves. It takes a while to become familiar with all the different versions and interpretations.


I'm also interested in seeing where the CUVA version ends up. I look forward to seeing the graph.

dvallis, any chance you can add Romani to the input languages? I know it's going to be tricky to find text in Romani (and there are multiple dialects), but it looks like you are good at finding text written in obscure languages 

(11-08-2020, 08:08 PM)Gioynich Wrote: You are not allowed to view links. Register or Login to view.dvallis, any chance you can add Romani to the input languages? I know it's going to be tricky to find text in Romani (and there are multiple dialects), but it looks like you are good at finding text written in obscure languages 

I accept the challenge! :-) Will add Romani to the next run.

Hey, here's the latest.

I used You are not allowed to view links. Register or Login to view. to convert EVA to CUVA.

As a sanity check, here's the 1st line of You are not allowed to view links. Register or Login to view.

fachys ykal ar ataiin shol shory cthres y kor sholdy sory ckhar or y kair chtaiin shar are cthar cthar dan syaiir sheky

And the 1st line of You are not allowed to view links. Register or Login to view.

fasyc ykal ar atam zol zory tsrec y kor zoldy cory ksar or y kair stam zar are tsar tsar dai cyanr zeky or ykam zod tsoa

Also hunted down Romani, combined and transmorgafied it with You are not allowed to view links. Register or Login to view. and and included You are not allowed to view links. Register or Login to view. in the latest run.

And the results? EVA and CUVA are very close on the plot, and still smack in the middle of Caucasian. Romani is out in the wilderness around Turkish, Kurdish and Finnish and Macedonian.

So, no wild changes in data with small tweaks in letters. That's good.

Image is flipped around because I don't know how to alter the viewpoint of a PCA plot. You get what you get. If anyone knows how to drive R-Plot and is good with PCA, have at it. Latest data is You are not allowed to view links. Register or Login to view..

Very interesting, again, Darrin.



Okay... is there a chance you could do Scenario 3, which expands some of the most common Latin scribal abbreviations into their most common expansions? I know it is more work than the others to convert it, but here are some reasons it might be worth the time:

• Latin was the lingua franca of literate people.
  
• Some of the glyph-shapes in the VMS are morphologically similar to Latin scribal abbreviations.
  
• It was not uncommon for 15th-century ciphertext to include some of the most common abbreviation symbols, especially y and the "-rum" symbol (in fact, there are examples on the diplomatic ciphers thread that include these).
  
• Some of the glyphs, like y and m are positioned as one might expect them in a medieval manuscript written in one of the languages that used Latin scribal conventions.
  
• Latin is one of the most common languages selected by "solvers", but not only by solvers, also by some of the more cautious researchers (those who are not saying they have solved it but who feel there are reasons to believe it might be Latin).
  

I'm agnostic about whether the VMS is meaningful and also about which language it might be if it is, but it seems prudent to see if the scribal patterns that were prevalent at the time will shift the position to any significant degree.


Another thought... should you include a sample of random text (one with spaces) to see where it falls on the graph?

(11-08-2020, 05:30 PM)dvallis Wrote: You are not allowed to view links. Register or Login to view.that EVA-Sh and EVA-sh are not the same thing

They are the same thing.

EVA-Sh and EVA-sh are not the same thing.

This is EVA-S on the left (in magenta)                                                      and EVA-s on the right:

                          [attachment=4662]

• EVA-Sh Sh is common in the VMS.                                                   •  EVA-sh sh is essentially non-existent in the VMS.
  
• [font=sans-serif]EVA-S S attaches to [font=Tahoma, Verdana, Arial, sans-serif]EVA-h [/font][/font]h                                                                  •  EVA-s s does not attach to EVA-h h
  
• [font=sans-serif]EVA-S [font=sans-serif] S[/font] [/font]does not stand alone.                                                             • EVA-s  s frequently stands alone. 
  

  

People who create their own transcripts (or modify existing transcripts) based on the Basic EVA chart that is widely available on the Web and who compare this to the VMS text do not treat them as the same.

Here is the Basic EVA chart linked from your Website:



This chart has been reposted as source material on a number of sites that discuss the VMS glyphs or who are doing computational attacks.


Also, as per voynich.nu Website:

Quote:The Eva alphabet, primarily in its basic form, found great reception in the community, and it was used by the Japanese Takeshi Takahashi to produce a new transliteration, also based on the Yale "copyflo" You are not allowed to view links. Register or Login to view., which was the first that could be considered essentially complete. He used 'capitalised Eva' to represent all benched or pedestalled characters, e.g. 'Sh' for  Sh.

This statement is consistent with what we see in the chart. T K P F are benched characters and  EVA-S S is part of a bench character (left side + cap). In contrast, EVA-s (lowercase) s is not part of a bench-character+cap. EVA-s is a somewhat 2-shaped character that behaves quite differently from the left side of a capped bench.

Hi Darrin,
Thank you for the update!
I was thinking that Dim1 could correlate with the number of vowel occurrences, with Abjads on the left. It seems that Eva and Cuva fall at the boundary between Abjads and non-abjads with many consonants e.g. Dutch, Polish. Languages with many vowels like Italian are on the right-hand side.

Edit: Adyghe seems to contradict this idea. Do you think you could also share your original files?

@dvallis
For me it is interesting that you take into account Mandarin, Hindi, Japanese, but the languages where in the VM text are not present.
Maybe you should realize that today's spoken and written German is there to help us understand each other.
The text passages in the VM point in Alemannic towards the Bavarian language area. Which is supported by the image of the crown.
To understand the differences better, I have here 3x link. German, Bavarian, Alemannic.
You can also just click on wiki the different languages back and forth.
Do you see the differences ?
See link:
deutsch:         You are not allowed to view links. Register or Login to view.
alemanisch:    You are not allowed to view links. Register or Login to view.
bairisch:         You are not allowed to view links. Register or Login to view.

nur alemanisch:   You are not allowed to view links. Register or Login to view.

Everyone who can speak German, is welcome to read along.

Translated with You are not allowed to view links. Register or Login to view. (free version)

It would be interesting to see the different German dialects, but it's probably lower priority. They will probably cluster almost on top of each other on the graph.

Also, even though some of the marginalia is German, that does not mean that the main text is German. In this particular analysis, Voynichese does not come anywhere near German on the graph.


I was going to suggest adding Swedish or Icelandic, so one of the older Germanic languages would be represented, but they are so similar to other Germanic languages, it likely will come out near German and English and the additional information probably won't be very useful.


Darrin, I want to thank you for posting all this. It has increased my awareness of Caucasian languages. I have looked at them briefly, but not as much as some of the other language groups.