Do you believe Voynichese is a substitution code for a natural language?

[-JKP-]

The majority of solutions proposed for the VMS are substitution codes for natural languages (with the most common choice being Latin, but with most other common languages like Czech, English, Italian, Greek, etc., also being proposed).

Many of these are one-to-one substitution codes, some are substitution codes that depend on expanding some of the glyphs (one-to-one substitution combined with one-to-many substitution). Recent examples of one-to-one combined with one-to-many include Lockerby and Gibbs (Latin).

A few have also suggested transposition codes (codes that include substitution, but also transposition of the letters, as with anagrams). Sherwood and others have suggested this.


I would like to restrict this poll to substitution codes that do not involve transpositions (in other words, this poll is not asking about solutions that involve changing relofw into flower). It's probably better to ask that as a separate question.

[Koen G]

I think that if Voynichese is a natural language, one to one is unlikely. I rather believe it's several-to-several.

For example, I think 9 is just a with a flourish. And gallows are expressed by different glyphs in the middle of words. So one sound maps to several glyphs depending on its location. 

Likewise, one glyph might express different sounds depending on its surroundings. Think biglyphs. For example, in French, 'u' has a different sound value when it's preceded by a, and different again when preceded by o.

So several to several. That's the only way I can see it work.

[Anton]

I think options 3 and 4 are somewhat confusing between them. Basically it's better to specify whether we are speaking of the overlay (the script as it appears) or of the underlay (that which the observed script is utilized to convey).

For example, for the "hoax theory": the overlay is not a natural language, and there is no underlay at all.

[davidjackson]

I don't think the questions make sense. If we agree it's a one-to-one code, then it's not a natural language; and so the fourth answer makes no sense.

And the third option doesn't gell with your explanation - it should read more like "underlying natural language with anagram code obfuscation" or something. It's fine saying

I would like to restrict this poll to substitution codes that do not involve transpositions (in other words, this poll is not asking about solutions that involve changing relofw into flower).

But if you don't give those who think this is the answer an opportunity to vote, you're skewing the result.

[Koen G]

Would the third option mean a complex cipher for example?

[-JKP-]

I don't think the questions make sense. If we agree it's a one-to-one code, then it's not a natural language; and so the fourth answer makes no sense.

And the third option doesn't gell with your explanation - it should read more like "underlying natural language with anagram code obfuscation" or something. It's fine saying

I would like to restrict this poll to substitution codes that do not involve transpositions (in other words, this poll is not asking about solutions that involve changing relofw into flower).

But if you don't give those who think this is the answer an opportunity to vote, you're skewing the result.

1. By substitution code, I mean that the VMS glyphs are substituted for letters (or groups of letters) in a natural language.


2. Adding transposition to the question makes it too complex. One then has to consider one-to-one substitution with transposition, one-to-many substitution with transposition, no substitution with transposition. I really think it would be better to deal with tansposition separately.

If there are people who think it's a substitution code with transposition (or some other kind of code with transposition) then they can abstain or suggest another poll (or suggest better wording for this poll, but I can't see a way to do it without adding at least three more options and I'm not sure that would work).

[Emma May Smith]

Would the third option mean a complex cipher for example?

Surely the third option means that it's a natural language written in the plain without any kind of cipher?

[-JKP-]

Would the third option mean a complex cipher for example?

Surely the third option means that it's a natural language written in the plain without any kind of cipher?

I meant the third option to be pretty broad, so yes, it could mean a natural language written in the plain (with glyphs that may be unfamiliar to us but not necessarily to them), but it would include other kinds of cryptography (such as steganography in which the overall shapes or the darks and lights in the inks, or other forms of hiding information are used rather than substituting glyphs for letters).

[Davidsch]

None of the above fits exactly.

My current answer to the question is:  the Voynich text is a ciphered or coded message. The decoded message is probably a natural language, with strong preference for Latin.

[Koen G]

I don't know whether my view qualifies as a code. I'm more inclined to see the VM as a cultural product, though I'm not sure about the script. I do lean towards the possibility that this script was used among a group of people. 



Basically my question is: what is the diffetence between a script and a substitution code (of any type)?