Calgary engineer believes he's cracked the mysterious Voynich Manuscript

[MarcoP]

It is the small clues that make it so difficult to believe that it is Turkish.
Example:
Why is the Latin alphabet used when in the Eastern Roman Empire, Greek was normal.

Hi Peter,
You are not allowed to view links. Register or Login to view. mentioned Codex Cumanicus You are not allowed to view links. Register or Login to view. a few years ago.
From the Wikipedia page:

It consist of two parts. The first part consists of a dictionary in Latin, Persian and Cuman written in the Latin alphabet, and a column with Cuman verbs, names and pronouns with its meaning in Latin. The second part consists, Cuman-German dictionary, information about the Cuman grammar, and poems belonging to Petrarch.

The Codex uses the Latin alphabet and contains a mix of Romance languages (Latin and Italian), German, Cuman (a Turkic language) and Persian (which had a considerable influence on Turkish). I see no reason to exclude Turkish languages from the candidates. In his 2014 paper You are not allowed to view links. Register or Login to view., Stephen Bax mentioned reduplication in Turkic languages as a relevant feature. He also wrote:

the language in the VM could be borrowed in part from Indo-European languages such as Latin, and could be acting in part as an Abjad, like Arabic and other Semitic scripts, the underlying language could nevertheless be from a completely different language family again, such as Turkic.

You are not allowed to view links. Register or Login to view. also mentioned that Turkic languages are good candidates. She also pointed out some simple You are not allowed to view links. Register or Login to view..

Expanding on Emma's observations, Ahmet believes that the Turkish 'k' is represented by EVA:m m. He seems not to be aware that EVA:m almost exclusively appears at the end of lines: it looks like a special feature of the writing system. In my opinion, the best explanation is that it is an abbreviation symbol, but there may be other good explanations.
Anyway, EVA:m clusters at line ends while Turkish:k is an ordinary character. The idea that the two  are the same thing is obviously wrong.

I hope the attached image makes clear both the line-effect for EVA:m and the higher frequency of Turkish 'k'. 
EVA:m occurs in 2.6% of word-tokens, Turkish:k in 23.9% (I used the Zandbergen-Landini transliteration ignoring uncertain spaces and the Turkish file in You are not allowed to view links. Register or Login to view.). The difference is huge, almost one order of magnitude. It's really strange that Ahmet did not notice.

Almost all (95.8%) Voynichese words which  include EVA:m have the character at word-end. Turkish 'k' has no preference for the word-ending position. Also from this side, things definitely don't work.


Turkish k words%: 23.9% word-final: 25.2%
Voynich m  words%:  2.6% word-final: 95.8%

We have seen so many people make the same mistakes that it's only natural to be put off. But Ahmet's efforts may still have the effect of stimulating more research on the Turkic-language hypothesis and maybe others with a greater interest for evidence will look at this. Not all evil comes to harm.