Discussion of "A possible generating algorithm of the Voynich manuscript"

[Torsten]

Torsten: re-reading your 2019 article and looking at your source code, you term certain high-frequency pairs of glyphs (such as ‘ol’ and ‘dy’) as ‘ligatures’, and yet treat them in the same way as ‘o’ and ‘l’ (i.e. individual glyphs).

* Why was it necessary for you to treat these ‘ligatures’ in a special way in your paper?

The answer to your first question is given in a blogpost at You are not allowed to view links. Register or Login to view.:
In this blogpost you have pointed out that it is not easy to transcribe the VMS. There are different transcriptions alphabets available and none of them is perfect. One transcription alphabet is EVA. Since it is stroke based it is necessary to parse EVA. You also argue that EVA isn’t final and that it would be an error to treat EVA as an exact representation of voynichese (see You are not allowed to view links. Register or Login to view.). 

As an example for parsing a word in EVA you suggest that 'qokeedy' should be parsed into [qo][k][ee][dy] (see You are not allowed to view links. Register or Login to view.). For parsing EVA you obviously define tokens like 'qo', 'ee', or 'dy'. Nothing else happens in the text generator. The program also defines tokens and it even parses EVA 'qokeedy' in the same way as you do. The only difference is that you explain rare words as copying errors whereas I explain the whole text as generated by modifying copied words.

The answer to your question is therefore, that a computer readable representation of the VMS text was necessary and we have chosen EVA for this purpose. In our eyes EVA is the best representation available today. But since EVA is stroke based it is necessary to parse this strokes into ligatures or tokens.

They are not consistently physically joined on the page, so are not actual ‘ligatures’.

Remains the question if stroke groups should be called ligatures. We have already used the term token to distinguish between word-types and word-tokens. Therefore we used the word 'ligature' here. Some of the stroke groups look like ligatures. This is for instance the case for EVA 'ch', 'sh', 'ckh', 'cth', 'cph', and 'cfh'. But since 'ch' consists of two 'e'-strokes connected by a dash it would also be possible to interpret 'ch' as two 'e'-characters. This way it would also be possible to interpret groups like 'chh', 'ckhh' or 'cckhh'. The goal behind EVA was to represent such groups. I therefore would agree that a definition for the term 'ligature' would be helpful.

* Why would the original person doing the auto-copying have treated them in a special way?

Your question suggests that it is possible to distinguish between normal and special behavior. I would argue that there is no such difference. For each common word there is at least another one differing from it by only a single quill stroke. For example, in addition to the word 'daiin' also the words 'dain' and 'daiiin' are present in the text. This means the script to write the VMS uses similarly shaped glyphs since the shape of a glyph is important (see Timm 2015, p. 36).

* Why are there so many manually added substitution cases in your code on github?

It is possible to generate other words, which exist in the VMS, by replacing similar shaped glyphs (see Timm 2015, p. 5). This is in my eyes the main idea behind the VMS and it is the main idea for the text generation method used by the self-citation text generator. The substitution cases define which glyphs or tokens are similar to each other. There is no other way to make the information about similar shaped glyphs available. Even if someone would write a OCR program for the VMS this program would use some definition of how to handle shape differences.

Your other two questions are only used to transport your personal opinion. Even if we disagree about arguments it should be possible to discuss them based on facts.