Was Voynichese meant to be pronounced?

[MarcoP]

It seems the thread has turned into brainstorming decipherment ideas, something we have done numberless times.
I think this is a waste. Stephen proposed a specific question that I don't think has ever been discussed and is independent from the possibility of deciphering the text.
As Stephen wrote You are not allowed to view links. Register or Login to view., positional rigidity is not necessarily a problem for pronounceability.

[Koen G]

I think the most important aspect is the alternation of apparent vowels and consonants. Various analyses have shown that Voynichese likely does something like this. 

We can't know if it was explicitly designed with pronouncability in mind, but it certainly did turn out that way, while it could have turned out differently as well. 

Furthermore, its shapes are familiar enough to Western eyes, which is again a huge boost to pronouncability.

[ReneZ]

if one applies a tool that separates vowels and consonants to a transcription of the Voynich MS, then this tool will provide an answer. It just applies a pre-defined logic and it won't say: "sorry, I can't do it".

Of course, that would have been possible, in principle, but is more difficult.

From my own analysis, I do not see a clear separation between vowels and consonants, as shown You are not allowed to view links. Register or Login to view. .

[-JKP-]

1) If we assign sounds to the glyphs that "look" like consonants and 2) if we assume the round shapes are vowels ("a", "o" "c" and maybe "9") then sure, it's pronounceable. Other combinations would also be pronounceable, depending on how they are sound-associated.


But we don't KNOW if the vowel shapes represent vowels (they do not behave like vowels in natural languages with regard to position), which makes it hard to determine whether it was meant to be pronounced.

If the VMS is a cipher, then maybe the vowel-looking shapes are nulls. Nulls are not intended to be pronounced, except maybe to make it easier to remember the components (even nonsense syllables are easier to remember if they can be pronounced, as is evidenced by songs with nonsense syllables).

[-JKP-]

By assigning the Voynich glyph e to the letter "e" on the keyboard, the EVA system has, in a sense made Voynichese more pronounceable.

Imagine trying to say chedy to someone at a Voynich conference without converting the c-shape to a vowel.


People claiming decipherments of the VMS in both western and some of the eastern languages (not the abjads), almost always assign ch as a consonant and e as a vowel, and usually also assign y as a vowel. That's because it makes sense to us that way.

Now imagine if all of them were consonants, it would be hard to pronounce (and may not have been designed to be pronounced) or, alternately, if the vowel-consonant shapes were reversed (so that chedy were something like "amis") then some of the tokens would be pronounceable but they would have to be evaluated differently.

[Searcher]

It seems the thread has turned into brainstorming decipherment ideas, something we have done numberless times.
I think this is a waste. Stephen proposed a specific question that I don't think has ever been discussed and is independent from the possibility of deciphering the text.
As Stephen wrote You are not allowed to view links. Register or Login to view., positional rigidity is not necessarily a problem for pronounceability.

I agree, in general, but the verbal ability and memory of people make them choose and form pronounceable constructions in the limits of their language traditions. We judge by majority of the VMS words which can be easily pronounced, but there are some words that make problems with this. If we change sounds for glyphs in the same proportion, with the aim to find every  letter of Latin alphabet, we will get a pronounceable majority of the text again, and again some part of the words will be hard for pronounce. So, is it not a waste?

[MarcoP]

if one applies a tool that separates vowels and consonants to a transcription of the Voynich MS, then this tool will provide an answer. It just applies a pre-defined logic and it won't say: "sorry, I can't do it".

Hi Rene,
you are right of course. The level of the alternation in the resulting classification depends on properties of the input data, but this is probably irrelevant.

For instance, using Sukhotin's algorithm, one can replace supposed vowels with 'o' and supposed consonants with 'C'. For the first lines of Dante's Divina Commedia, one gets:

nel mezzo del cammin di nostra vita
mi ritrovai per una selva oscura
che la diritta via era smarrita

CoC CoCCo CoC CoCCoC Co CoCCCo CoCo
Co CoCCoCoo CoC oCo CoCCo oCCoCo
Coo Co CoCoCCo Coo oCo CCoCCoCo

For the first lines of the VMS using CUVA:

FASYC YKAL AR ATAM ZOL ZORY Y KOR ZOLDY
CORY KSAR ORY KAIR STAM ZAR ACE TSAR TSARDAI
CYANR ZEKY OR YKAM ZOD TSOARY TSEC DARAM CY

CoCoC oCoC oC oCoC CoC CoCo o CoC CoCCo
CoCo CCoC oCo CoCC CCoC CoC oCo CCoC CCoCCoC
CooCC CoCo oC oCoC CoC CCooCo CCoC CoCoC Co

For a random text:

ztcuij noelu cgmjgqbv dt fpdevas zz rimziojupb
pb elc xfmykgph uqdrwqwzqlpl vqpnpdjb exs zretxvyf
fzid lqybaxlxaifu kfkebwpdspyj cw ygdhlufcb ubgjcq

ooCCoC ooCCC CCCCCooC oo oooCCCC oo ooCoooCCoo
oo CCC CoCCoCoo CoooCoCooCoC CoooooCo CCC ooCoCCCo
oooo CoCoCCCCCooC oooCoCooCoCC CC CCooCCoCo CoCCCo

For a random text, one does not get a high rate of o-C alternation: results show a higher number of consecutive 'oo' and 'CC'. I guess this just means that Voynichese words are not "random", something that is well known.

From my own analysis, I do not see a clear separation between vowels and consonants, as shown You are not allowed to view links. Register or Login to view..

This is a strong point against the idea that Voynichese could have been meant to be pronounceable.

[MarcoP]

Furthermore, its shapes are familiar enough to Western eyes, which is again a huge boost to pronouncability.

I am not sure. Why should familiar shapes suggest pronounceability, when they obviously don't have their usual meaning / phonetic value? They suggest that Voynichese was designed to look like language.

[Koen G]

They help pronounceability for those used to such glyphs, even if the pronunciation may not be the "correct" one. I will demonstrate this by converting a piece of Voynichese in EVA to the font Wingdings:

   

Edit: to clarify, I mean that it helps, not that it was necessarily intended.

[ReneZ]

To be more precise, being able to separate vowels and consonants, and being able to pronounce, are not the same. Depending on one's origin, one can pronounce just about anything.

Not many people would be able to comfortably pronounce the dutch words "angstschreeuw" or "koeieuier", and Czech is famous for being able to use "r" and "l" as vowels, leading to various more or less meaningful sentences that do not seem to have any vowels.

The problem with Voynichese is that it isn't really working either way.
Even if Eva is largely pronounceable, there are many words that aren't, at least comfortably. While the opening lines of You are not allowed to view links. Register or Login to view. (in Cuva, as shown by Marco) seem to be OK, there is already a word ending in -NR that is uncomfortable.

There are also such words as chcthy / chcThy and oeeos / oeeos

The characters / sequences ee and ch seem to be able to replace each other, but the first is usually classified as a vowel (pair) and the second as a consonant.