Phonology of Voynichese

[MarcoP]

Geoffrey forgot to link the source of the fragment on which he grounds his assessment of the statistics of Greek: boring. 
It comes from this You are not allowed to view links. Register or Login to view..

αντικείμενο κ.λπ.), με συντακτικές δομές (π.χ. κανόνες συμφωνίας, σχηματισμός της παθητικής φωνής, αποτύπωση των υποχρεωτικών συμπληρωμάτων ενός ρήματος στο σχηματισμό της πρότασης κ.ά.) και με σημασιολογικές αξίες (π.χ. χρονική τοποθέτηση ή τρόπος παρουσίασης ενός γεγονότος, σχέση κτήσης κ.λπ.). Αναφέρονται επίσης επιλεκτικά ορισμένες πραγματολογικές συνθήκες που

Slightly edited Google translate (the fragment selected by Geoffrey in bold):

In this way, the different forms that a word takes in inflection are combined with grammatical functions (e.g., subject, object, etc.), with editorial structures (e.g. agreement rules, formation of the passive voice, imprinting of mandatory complements of a verb in the formation of the sentence etc.) and with semantic values (e.g. time position or presentation of an event, acquisition relationship, etc.). Certain factual conditions are also mentioned that selectively determine the use of alternative formulas (e.g. courtesy, style levels, etc.).

Some observations:

• For no apparent reason, Geoffrey has excluded from his count the Greek forms corresponding to "etc" and "e.g.". If one includes them, words ending with -ς make up 46% of the text.
  
• The fragment is not grammatical: it is made of the end of a first long sentence and the beginning of a second one. Both are incomplete. 'Αναφέρονται' (are mentioned) was the single verb selected by Geoffrey (out of the four that appear in the two complete sentences). The result is that only 2% of the words are verbs.
  
• This fragment is mostly made of lists of items of the form "X of Y" (formation of the passive voice, imprinting of mandatory complements of a verb in the formation of the sentence) where X is in the Nominative case and Y in the Genitive (both cases correlating with -ς as mentioned by Geoffrey)
  
• In a grammatical modern Greek text (the four paragraphs before the table of content of the wikipedia You are not allowed to view links. Register or Login to view.) 22% of the words end with -ς (which is by far the most frequent word-ending character). This is less than half the value in the fragment chosen by Geoffrey.
  
• The frequency of word-final -y in Voynich Currier B (45%) is comparable to the value observed by Guy in the two pages he examined (50%).
  

TLDR the frequency of -ς in a grammatical Greek text is half that of -y in Currier B. It is slightly more than half the frequency of -y in the whole manuscript (22% vs 41%). Guy's conclusion that -y cannot correspond to -s / -ς is confirmed.

If one wants to believe that the VMS is a direct encoding of Greek, one is forced to go down the road chosen by Geoffrey and interpret the text as ungrammatical lists of words.

PS: I suggest we stay on topic and adhere to the assumptions that David stated in the first post, in particular:

• Voynichese will be a language that shares enough phonemes with our languages to be recognised in the modern era.
  

The idea of ungrammatical word-lists with arbitrarily skewed character distributions can be discussed elsewhere.