The Voynich Manuscript is a cipher because...?

[-JKP-]

It was not uncommon for short passages to be anagramed in a way that was not completely apparent because the assumption was that if it wasn't too long an anagram, it could be worked out, even if there were several possible solutions. Many crossword slots have more than one possible answer, but once you get the right context (and the right potential words running 90°), you can usually figure it out.


The situation is different, however, with long passages of text. If it's anagramed with no specific pattern, it becomes a one-way cipher... useless to anyone (sometimes even the original writer) because it can't be decrypted.

[Anton]

Thread moved.

As to the subject matter of the thread: first of all, the distinction between "positive" and "negative" arguments looks to me artificial. When we speak of difference between two objects, that means that one object has or possesses something that the other one lacks, and vice versa. That's just the essence of difference. Hence if one argues that a language does not normally behave in that way or another - this is as "positive" an argument toward a cipher as it is a "negative" one.

I would like to mention some of arguments in favour of the cipher hypothesis not mentioned above in the discussion.

1) Here and there the text bears evidence of having been put down not in a single pass. For some examples and discussion refer e.g. to You are not allowed to view links. Register or Login to view..

2) There is that curious phenomenon of "gallows coverage" there. Refer to You are not allowed to view links. Register or Login to view..

3) There is at least one label consisting of only one character. Refer to You are not allowed to view links. Register or Login to view..

It can be difficult to separate a cipher from a constructed language.

For that I proposed the criterion of the narration structure. Refer to You are not allowed to view links. Register or Login to view..

[Emma May Smith]

As to the subject matter of the thread: first of all, the distinction between "positive" and "negative" arguments looks to me artificial. When we speak of difference between two objects, that means that one object has or possesses something that the other one lacks, and vice versa. That's just the essence of difference. Hence if one argues that a language does not normally behave in that way or another - this is as "positive" an argument toward a cipher as it is a "negative" one.

Except that there are not only two possibilities. Also, negatives and positives are never black and white, more like arguments for and against to be weighed and considered. A lot of arguments against a linguistic solution can be put forward as problems that need to be solved, but they may in no way add up to an argument for a cryptological solution.

I really think the positive and negative split is crucial if we're not simply going to get lost in a limbo of "I think it is a cipher because it doesn't look like any language I know!" It's a logical fallacy of lack of imagination.

Of your three points in favour of a cipher, I agree that point 2) is good. But point 1) is ambiguous, and point 3) is incredibly thin.

[Anton]

Of your three points in favour of a cipher, I agree that point 2) is good. But point 1) is ambiguous, and point 3) is incredibly thin.

Point 3) is not that thin when we begin to consider possibilities of what this label may stand for. Every possibility yields quite strange outcome. Like, if it stands for a number, then from the viewpoint of an unknown language we should admit not only unknown script, but also an unknown numbering system. If it is a noun, then we have a one-letter noun (is that at all possible in a non-hyeroglyphic script?) And so on.

[Oocephalus]

Why should a one-letter noun be impossible? If French was written phonetically, "eau" would be one. Indeed, the vowels a, e, i, o, u are all nouns in some language according to Wiktionary.

[-JKP-]

Of your three points in favour of a cipher, I agree that point 2) is good. But point 1) is ambiguous, and point 3) is incredibly thin.

Point 3) is not that thin when we begin to consider possibilities of what this label may stand for. Every possibility yields quite strange outcome. Like, if it stands for a number, then from the viewpoint of an unknown language we should admit not only unknown script, but also an unknown numbering system. If it is a noun, then we have a one-letter  noun (is that at all possible in a non-hyeroglyphic script?) And so on.

I don't know if the behavior of the labels in the VMS will be a good model for how the main text works. They may use a different system.

[Emma May Smith]

Of your three points in favour of a cipher, I agree that point 2) is good. But point 1) is ambiguous, and point 3) is incredibly thin.

Point 3) is not that thin when we begin to consider possibilities of what this label may stand for. Every possibility yields quite strange outcome. Like, if it stands for a number, then from the viewpoint of an unknown language we should admit not only unknown script, but also an unknown numbering system. If it is a noun, then we have a one-letter  noun (is that at all possible in a non-hyeroglyphic script?) And so on.

Many scripts at this time used characters to represent numbers. Being both an unknown script and an unknown numeral system would be pretty much normal.

Many languages have nouns which have a single sound.

Most importantly, it's not hugely convincing that this single stroke is a label.

[-JKP-]

Many scripts at this time used characters to represent numbers. Being both an unknown script and an unknown numeral system would be pretty much normal.

...

I have to agree with this. Quite a few languages didn't have separate characters for numbers. The letters were used and it was known by context (or sometimes by a symbol like a line above the numbers) that it was supposed to be read as numbers.

[Anton]

Many scripts at this time used characters to represent numbers

Yes, I am well aware of that. Old Russian did exactly that.

Being both an unknown script and an unknown numeral system would be pretty much normal

What is not normal is that this supposed number is represented by a symbol which is extremely rare in the corpus.

Many languages have nouns which have a single sound.

Here we have a single letter. Is that OK?

Most importantly, it's not hugely convincing that this single stroke is a label.

What's it, otherwise?

[ReneZ]

What is not normal is that this supposed number is represented by a symbol which is extremely rare in the corpus.

I am not so sure that it isn't simply an Eva-s, which occurs stand-alone elsewhere in the MS.
The bottom part of the s is barely visible but it seems to be there...