+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: News (https://www.voynich.ninja/forum-25.html)
+--- Thread: [Article] Category-based dynamics in the Voynich Manuscript (draft) (/thread-4858.html)
Category-based dynamics in the Voynich Manuscript (draft) - Addsamuels - 08-08-2025
I have a new draft paper,
I don't think it's well written, but I think the findings are useful, especially on categories.
You are not allowed to view links. Register or Login to view.
RE: Category-based dynamics in the Voynich Manuscript (draft) - RobGea - 08-08-2025
Quote:The VM is generally considered to be divided into around several parts with illustrations,These are considered to be
Herbal (H) , Astronomical (A), Cosmological ( C), Zodiac (Z),
Balneological / Biological (B), Pharmaceutical (P), Stars/Recipes (S), and miscellaneous Text (T).
Quote:The categories are themselves divided into the subcategories
(AC, BB, CB, CC, HA, HB, HC, PA, SA, SB, TA, TB, TC, ZC),
where the first letter is the corresponding category letter and the second letter is the Currier language.
Hmm, then, SA would be the Stars / Recipes section ( Quire20 ) and it would be written in Currier Hand A.
But the entire Stars / Recipes section is written in Currier Hand B.
How did you get the data for the SA subcategory ?
RE: Category-based dynamics in the Voynich Manuscript (draft) - Addsamuels - 08-08-2025
(08-08-2025, 09:03 PM)RobGea Wrote: You are not allowed to view links. Register or Login to view.My source was this You are not allowed to view links. Register or Login to view. in which one folio is SA:Quote:The VM is generally considered to be divided into around several parts with illustrations,These are considered to be
Herbal (H) , Astronomical (A), Cosmological ( C), Zodiac (Z),
Balneological / Biological (B), Pharmaceutical (P), Stars/Recipes (S), and miscellaneous Text (T).
Quote:The categories are themselves divided into the subcategories
(AC, BB, CB, CC, HA, HB, HC, PA, SA, SB, TA, TB, TC, ZC),
where the first letter is the corresponding category letter and the second letter is the Currier language.
Hmm, then, SA would be the Stars / Recipes section ( Quire20 ) and it would be written in Currier Hand A.
But the entire Stars / Recipes section is written in Currier Hand B.
How did you get the data for the SA subcategory ?
<f58r> S A
<f58v> S A
RE: Category-based dynamics in the Voynich Manuscript (draft) - RobGea - 08-08-2025
Nice, i stand corrected.
RE: Category-based dynamics in the Voynich Manuscript (draft) - Addsamuels - 01-09-2025
If anyone has any feedback, please give it! Especially on the results. I think there definitely exist categories in the text, which could be used more?
RE: Category-based dynamics in the Voynich Manuscript (draft) - oshfdk - 01-09-2025
I think the paper is too technical for me. Is it possible to have some high level summary of the findings? I've tried checking the abstract and the conclusion, but still don't understand the significance of the result.
Quote:In this paper, I have hoped to show many words are “dominant” in a certain category, even with Currier languages. Additionally, many words are formed of an affix and a common dominant word, and there is relatively high amount of self-redirection with descendant words. I’ve also shown that Currier languages are highly linked to main (i.e. first) PCA Component on a PCA of the VM.
Could you explain what this means in practical sense? I'm mostly interested in anything that could lead to deciphering of the manuscript (if it can be deciphered). The fact that some sections of the manuscript have significantly different vocabulary (if this is what the results show) could mean many things: semantic differences or stylistic differences dictated by the topics in the text, differences in encryption or just different authors with their own idiosyncrasies.
Quote:For next steps, deciding the process which words are not easily connected to dominant words are formed, since rarer words are more unusual, would be helpful at working a provisional translation.
I'm not sure I understand this part at all, but I'm intrigued by "working a provisional translation". Could you elaborate?
RE: Category-based dynamics in the Voynich Manuscript (draft) - Addsamuels - 03-09-2025
(01-09-2025, 07:59 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.I think the paper is too technical for me. Is it possible to have some high level summary of the findings? I've tried checking the abstract and the conclusion, but still don't understand the significance of the result.
Quote:In this paper, I have hoped to show many words are “dominant” in a certain category, even with Currier languages. Additionally, many words are formed of an affix and a common dominant word, and there is relatively high amount of self-redirection with descendant words. I’ve also shown that Currier languages are highly linked to main (i.e. first) PCA Component on a PCA of the VM.
Could you explain what this means in practical sense? I'm mostly interested in anything that could lead to deciphering of the manuscript (if it can be deciphered). The fact that some sections of the manuscript have significantly different vocabulary (if this is what the results show) could mean many things: semantic differences or stylistic differences dictated by the topics in the text, differences in encryption or just different authors with their own idiosyncrasies.
Quote:For next steps, deciding the process which words are not easily connected to dominant words are formed, since rarer words are more unusual, would be helpful at working a provisional translation.
I'm not sure I understand this part at all, but I'm intrigued by "working a provisional translation". Could you elaborate?
Dear oshfdk, thank you for message. First of all, this is my first paper, so the results might not be actually that significant, hence the vagueness!
Quote:In this paper, I have hoped to show many words are “dominant” in a certain category, even withCurrier languages. Additionally, many words are formed of an affix and a common dominant word,and there is relatively high amount of self-redirection with descendant words. I’ve also shown thatCurrier languages are highly linked to main (i.e. first) PCA Component on a PCA of the VM.
So basically, I thought that categories are underresearched, the divisions in the text weren't really looked by C. Bowern and L. Lindemann in the Voynich linguistics paper (and their additional paper also seemed inconclusive). Although L.F. Davis has the hand system, I wanted to use traditional categories, based on a feeling, that text seems to match up with illustrations generally (I'd assume in general with manuscripts too).
So, using a "dimensional reduction analysis [that] allows for the reduction of the dimensionality of complex multivariate data while preserving patterns, which allows for easier visualisation and interpretation between variables.", which showed, like S. Goslee and other before, that the folio pages form nice clusters when separated for "Currier Language". Therefore the categories themselves actually are really separated (especially compare HA to BB), or even to ignore Currier languages, the Herbal and Biological/Balneological folios are too.
Therefore, I would assume that the work is divided into categories, and the text and images support this.
I then, like Goslee before, stated that the main (or first) PCA component of the You are not allowed to view links. Register or Login to view.correlates well with the Currier language distinction, i.e. the A folios are on the left, the B on the right, and the C in the middle.
Then, I tried to find which words are related to each category. To calculate this, I counted all of the words, and then worked out if one category had more than a certain amount of words in them. Then if it was 50% of more, I would count that word as being dominant in that category, as it seen mostly in that category. Because of the difficulty of reading the manuscript,I also added a minimum count (and to avoid hapax legomena etc). Therefore if a word is dominant in one category, it would make sense to link that word to a category. Thereby, you can start to link words to categories, and have a partial "translation" or tuning of each word.
I then, added certain affixes to words, and interestingly the words seem to fall in similar categories, to their starting category. I'm not sure exactly why this is.
I suppose, further work could normalise the values, so the larger categories are seen as less dominant, (since they naturally can contain more words, they'll dominated the dominant words), or look at the un-dominant words, which are very common, (again potentially also with this normalisation). This could lead to some information being generated.
Overall, I think it's really getting the point out some of the divisions in the manuscript!