Syllabification

[crezac]

Hmm, if we take a script for its face value, then we expect that the chunks of symbols represent words, do we not? Because that's the function that a script fulfils in language.

Those chunks may be words abbreviated or non-abbreviated, but they should be words, within the framework of the "face value" assumption.

That is not necessarily true.

You can't use a qualified statement to dispute a statement of fact and then follow it up with a contradictory statement of fact.  It doesn't make your fact correct.  Anton's statement is not necessarily true but "A Voynich word is not a word in a linguistic sense, whatever it may be" is not necessarily true either. 

Rather than just make the statement  could you explain why you are certain that a VMS word is not a linguistic word?  An abbreviated word is still a word in a linguistic sense - if I call someone a tard it isn't polite or P.C., but it has semantic content and is an abbreviated form of another word (perhaps P.C. would have been a better example). 

You want to read VMS as abbreviated 15th century Latin.  When (or if) it is read, it will be read as whatever it is