The Voynich Ninja

I always describe the EVA-t as being like "4p" as the left loop is angular like a right-angled triangle and the right loop is curved more like a circle.

I think of EVA-k as being "1p" or "lp". Then with with EVA-f the right shape is what I think of as a curvy "p" or an "sp", so I tend to write EVA-f as "ls" to distinguish it from the other p.

It is not common practice to split the EVA gallows characters in this way, but it makes a lot of sense to me to do so. It also fits with the appearance of the "4" in "4o".

In truth, because Voynichese doesn't use the Roman alphabet, it can't really be classified using Latin paleographical terminology. We can  talk about tendencies and overall stylistic similarities, but I think we have to be cautious about trying to pin it down using terms developed for a completely different set of symbols. I do think it's valuable to consider secondary terms like bookhand, cursive, and hybrid, however. Because of the lack of ligation, I wouldn't classify Voynichese as a cursive and, without having access to both a bookhand AND a cursive version of the script, we can't define a hybrid version. The exception to the lack of ligation is [qo] which, as I have argued, may be a single character since the symbols are (nearly) always ligated. The same might be argued for [ch], although since we just don't know the function of [ch] and its gallows-interpolated versions, it's hard to classify (something else I'm working on from a paleographical perspective).

The fact that [q] and [o] can be (but admittedly are rarely) separated by a gallows ([qpo] on 101r.1, [qfo] on 82v.33 (the best example), [qko] on 7r.1 and elsewhere) suggests that indeed [ch] should, by analogy, also be considered a single character whether benched or not.

I agree with your observation that Voynichese, not using the Roman alphabet, can't really be classified using Latin paleographical terminology. However, it's undeniable that the paleographical approach offers unique insights, even if they have to be tailored to this unique manuscript.

The classification of [qo] as a single character is intriguing, and I'm eager to learn more about your research on this. The occasional presence of separations by a gallows between [q] and [o] is indeed a point of interest, as is the exact nature of [ch]

[attachment=7543]

Once it is clear that EVA "ch" is a combination between 2 characters (already discussed), I can put whatever I want in between.
But now two spellings are also possible.
Variant A+B. Appended or not.
What each writer can now decide for himself, is understandable for the other, but remains a mystery for us.
  From a crypological point of view.

Thank you Aga - I haven't seen these pictures before. It's strange how intentional it seems that the letters are not connected.

Thank you for your insights into the interpretation of EVA "ch" as a combination of two characters. It's fascinating how this opens up the possibility for different variations in spelling. The choice between Variant A and Variant B, whether appended or not, indeed adds an element of intrigue to the deciphering process

I know my opinion doesn't count much, however I would like to point out that it would be much more sensible to search for the proper transliteration alphabet, because even if the VM language was suitable for computer analysis (which I don't think it is) the insufficient transliteration alphabet will only give insufficient results. However, since I am able to read at least 150 sensible words transcribed with EVA, I believe that the alphabet has matched more than half the VM correctly to the Latin letters. Most of the VM glyphs can be found in 15th century manuscripts, except the four tall glyphs and somewhat reshaped letters s, p, b.
When the vernacular languages switched to the Latin script, they used various ways to adopt Latin script to their particular sounds. 
1. Semivowels were often dropped, making writing, particularly Slavic, often look like abjad;
2. Latin letters were slightly changed - like German  z (z with a downward tail), w (two v together) for u,  or VM EVA t (as two t written together with one stroke to make two loops at the top), EVA s - as c with a flourish, a mirror image of s, b as a mirror image of b and p as a mirror image of slightly redesigned p.
3. Two or more Latin letters were used for one sound - German sch, English sh, or Italian ch, German tsch, Slovenian zh. 
4. Diacritic markers were used - it is not clear if the plumes over ch in the VM is a marker over c or z (like in the Czech alphabet developed by J. Hus) to for the Czeck sounds č, š, ž. Without diacritic markers over the vowels, there could be many words written the same, but pronounced differently.
5. New letter forms were invented - Latin letter C was problematic mainly because it was pronounced as c, ch, k or z. The Germans invented cumbersome form of K, so EVA k might be just more practical invention for the letter k. EVA has a designation for b, but in transliteration, it is seldom used, because the b is often connected with c, which can be confused with EVA sh. Since EVA p does not correspond to Latin p, some detective work is required to find the p. The Latin letter p is much better fit for the 4-like VM glyph designated in EVA as q. In this way, po- makes a perfect syllable which is a prefix in many Slavic languages.

The best manuscript to study Voynich Manuscript is the manuscript Tractatus de Husitis, which can be found on You are not allowed to view links. Register or Login to view./.  It was written between 1428 and 1450 by 5 different hands in Latin language with five folios in Slovenian. It reflects the VM style (no capitals, no punctuation), religious situation. Unlike VM, the Tractatus is written in Gothic book cursiva and reflects German writing convention which was not used in the VM, but in Slovenian protestant writing in the 16th century. 

Because the original transcription of the Slovenian pages of the Tractatus is available on line, it can help us understand the minims in the VM, so that we no longer refer to them as just  in, iin, iiin, but also as n, m, in, im, nu, un, iv, iw, wi, as Dr. Bax suggested.  Such changes would make the VM more readable, will improve vowel-consonant ratio, and increase the vocabulary.
Also, there are also various flourishes over m, n, w, which seems to be more of a decorative nature.

In Tractatus, various scribes used different German letters in the Latin text, as well as in Slovenian. In the VM, all scribes use the same letters and letter combination throughout the text, no German letters, but rather invented or re-designed VM glyphs and ligatures. 

Currier was wrong in assuming that the Voynich Manuscript was written in two different languages. He based his theory on the most frequent words in different sections, which can be a reflection of different subject matter, the grammatical style of writing and different grammatical style. For example, in Slavic languages, the suffix -am is used for 1. person singular, and -dy or -y for 2. pers. singular imperative mood. Both are important in poetry, instructions, and personal communication.

Another interesting manuscript that can be found on the same internet site is Latin Alphabetarium divinis amoris by Johannis Nyder, which also alludes to the Swabian connection and the spiritual movement Friends of God, which was considered 'pagan'. This manuscript also contains upward and downward flourishes at the end of w, n and m, or even without any flourish. 

The computer analysts have detected the prefixes, roots and suffixes.  Timm's analysis and his conclusion that the VM is a nonsense is based on wrong transcription alphabet. Improving EVA transliteration alphabet would enable much better reading and the recognition of the language.


Cvetka

I want to express my gratitude for your insightful contribution to the discussion. Your perspective enriches our understanding of the challenges associated with interpreting the Voynich Manuscript.

I agree with your proposition that a more in-depth search for a suitable transliteration alphabet could significantly impact our results. Your pertinent observation regarding the correspondence between EVA and Latin letters highlights how certain combinations of letters from that era, even when modified, can enhance transcription accuracy.

The various approaches adopted by vernacular languages to adapt the Latin alphabet to their specific sounds are truly fascinating. The examples you provided illustrate how these adaptations led to variations in writing and pronunciation.

The link you establish between the Tractatus de Husitis manuscript and the style of the Voynich Manuscript is intriguing. Comparing similar manuscripts can undoubtedly provide valuable clues to decoding the hidden meaning within the VM.

Your suggestion to consider minims in a new light for improved readability is ingenious. Similarly, your emphasis on decorative elements above certain letters could offer insights into the writing style and intentions of the scribes.

It's evident that your insights are grounded in a deep knowledge of linguistics and the history of languages. Your analysis of possible errors in previous transcriptions is crucial for advancing our understanding.

I thank you for sharing your findings with us, and I hope that these exchanges will continue to enrich our collective comprehension of the Voynich Manuscript.

Best regards,

As long as one is describing the shapes of what one observes in the writing, there is no issue. So, Eva "ch" 'looks like' a combination of two forms that can also appear separately, though that is quite rare. The intruding gallows 'look like'  a combination of a gallows and the Eva "ch". One can split the gallows into parts and then write all gallows as combinations of these.

The problems start as soon as one tries to interpret.
Dies ch a single character? What about the version with a little hook on top: Eva sh?
They are not written as single strokes but that does not mean anything.
Etc. etc 

Looking for the 'real' transliteration alphabet implies that this should exist. This then implies that the text should be a more or less direct rendition of some written plain text.
While that is a very good working assumption, everything that has been attempted on translating the text indicates that this does not work.