20-06-2026, 02:08 PM
Thanks @Rafal for giving me this tip. His thread: You are not allowed to view links. Register or Login to view.
In fact, the numeric cipher has a large number of Voynichese characters.
In his thread, Rafal suggests that the a symbol is not an a, but an ei written too close together. Throughout the manuscript, its spelling is stable and distinct from Latin a, it consists of writing e, and then i close to it on the right. For the type of difference and feature of the symbol a, let's compare it with the Latin a from the inscription on f116v:
The difference is insignificant, but it does occur - in Voynichese a, sometimes there is a gap between part e and part i, and in general it seems that it takes two strokes of the pen to write it. In the example from the word "oladabad" (which is more noticeable in the middle a), the letter is written with one stroke of the pen, and there is no pronounced space or gap between its left and right parts.
Remembering my syllable system, we get the fact that we do not have the syllables ai/al/ar, there are syllables ei, i(n) and the symbols of its group (l, r, m).
It turns out that we can identify the following system of minimal semantic units:
1). Units: i(n), l, r, m
2). Tens: e, s, d, y, ch/sh
3). Vagues: all gallows, x, q, o
4). Syllables: ei, eir, eil, oi, in, ir, iin, ee, eee,
Total: 25 "letters" (there are 26 of them in the Latin alphabet, if U = V, then 25)
Most of the words in Voynichese take the form of a series of "tens and ones". (I doubt that e means the same as C, i.e. a hundred). In this form, it looks like the classic numbering of letters a la A = 1, B = 2, C = 3... The first letters of the alphabet will always appear at the end of the word, and the last ones at the beginning.
The trick of the Voynichese alphabet is that it does not encode all letters as eei (21) or eii (12), but replaces some values with individual characters, making them difficult to calculate.
It's more like a numeric cipher than anything else.
However, this does not give us much. To detect letters in such a way that they may turn out to be (maybe not A = 1 Z = 26, but A = 26 Z = 1). But if we recognize the letters, then the output will definitely not be a perfect translation, but rather a set of letters where we will have to identify the nulls and the "splitting" of the word.
In the meantime, study, study, and only study...
In fact, the numeric cipher has a large number of Voynichese characters.
In his thread, Rafal suggests that the a symbol is not an a, but an ei written too close together. Throughout the manuscript, its spelling is stable and distinct from Latin a, it consists of writing e, and then i close to it on the right. For the type of difference and feature of the symbol a, let's compare it with the Latin a from the inscription on f116v:
The difference is insignificant, but it does occur - in Voynichese a, sometimes there is a gap between part e and part i, and in general it seems that it takes two strokes of the pen to write it. In the example from the word "oladabad" (which is more noticeable in the middle a), the letter is written with one stroke of the pen, and there is no pronounced space or gap between its left and right parts.
Remembering my syllable system, we get the fact that we do not have the syllables ai/al/ar, there are syllables ei, i(n) and the symbols of its group (l, r, m).
It turns out that we can identify the following system of minimal semantic units:
1). Units: i(n), l, r, m
2). Tens: e, s, d, y, ch/sh
3). Vagues: all gallows, x, q, o
4). Syllables: ei, eir, eil, oi, in, ir, iin, ee, eee,
Total: 25 "letters" (there are 26 of them in the Latin alphabet, if U = V, then 25)
Most of the words in Voynichese take the form of a series of "tens and ones". (I doubt that e means the same as C, i.e. a hundred). In this form, it looks like the classic numbering of letters a la A = 1, B = 2, C = 3... The first letters of the alphabet will always appear at the end of the word, and the last ones at the beginning.
The trick of the Voynichese alphabet is that it does not encode all letters as eei (21) or eii (12), but replaces some values with individual characters, making them difficult to calculate.
It's more like a numeric cipher than anything else.
However, this does not give us much. To detect letters in such a way that they may turn out to be (maybe not A = 1 Z = 26, but A = 26 Z = 1). But if we recognize the letters, then the output will definitely not be a perfect translation, but rather a set of letters where we will have to identify the nulls and the "splitting" of the word.
In the meantime, study, study, and only study...