The predictability of glyphs in labels agrees with that of vords

[david]

For instance, in English, "faster" can be broken into "fast" and "er", "singer" can be broken into "sing" and "er", but "lumber" can't be broken into "lumb" and "er" - it's just one morpheme, "lumber".

This is true if you have a knowledge of English. However, if you don't have an understanding of the meaning of the words but are simply looking for patterns to clarify the rules of the language (imagine an alien trying to dechiper English, or JKP trying to understand Voynichese ) then the -er suffix becomes a morpheme that fits in nicely with your list of rules (in the same way that we say 4o- tends to be a prefix in Voynichese). Because it's a common suffix that appears all over the place in clearly definable rules (it's a suffix that is the regular formative of agent nouns, or designator of nouns from occupation or characteristic).

But to take another example, a bigram such as "ez" isn't a morpheme for our alien because it's usually an n-gram pattern such as in sneeze, breeze, jeeze, tweeze, freeze, etc (unless he postulates that the 4gram "eeze" is a morpheme suffix!).

I agree with Sam, that's a good point.

A term is required here which does not have linguistic flavour.

I couldn't think of one, any suggestions? ngram and its derivatives could be an alternative, but fail to convey the meaning of letters with a potential sense meaning.

Basically we get 30% unique vocabulary and more than three times the amount of grammar that doesn't match.

I'm going to copy and paste the comments of Prof Stolfi here:

Abnormal words
The words that do not fit into our paradigm are collected in the gramamr under the symbol You are not allowed to view links. Register or Login to view.. These words comprise 1295 tokens (3.7%) in the main text, and 127 tokens (12.4%) in the labels. The vast majority are rare words that occur only once in the whole manuscript. They were manually sorted into a few major classes, according to their main "defect" as we perceived it:

• You are not allowed to view links. Register or Login to view.: words that do not have a properly nested layer structure, and seem to be two more normal words joined together (716 tokens, 55% of the abnormal words). These can be subdivided into:
   
  • You are not allowed to view links. Register or Login to view.: words with two or more gallows (208 tokens). The most common is oteotey (3 occurrences).
     
    
  • You are not allowed to view links. Register or Login to view.: words with crust letters surrounded by core or mantle letters (278 tokens). The most common are chodchy and cholky (4 occurrences each)
     
    
  • You are not allowed to view links. Register or Login to view.: words which contain the A.IN groups in non-final position (206 tokens). The most common are daiidy and dairal (5 occurrences each).
     
    
  • You are not allowed to view links. Register or Login to view.: abnormal words which contain the y letter in non-final, non-initial position; or the letter q in non-initial position (24 tokens). The most common is oykeey (2 occurrences).
     
    
    
• You are not allowed to view links. Register or Login to view.: this class was defined by John Grove, who noticed that the rare words often found at the beginning of lines, such as polchedy, could be interpreted as normal words prefixed with a spurious gallows letter. Of the abnormal tokens in the text, 213 (16%) fit this description.
   
  
• You are not allowed to view links. Register or Login to view.: the remaining 366 abnormal tokens (28%) are not easily interpreted as joined words or Grove's gallows-prefixed words. We have sorted them into:
  • You are not allowed to view links. Register or Login to view.: words that have one of the letters m or g not preceded by a circle (57 tokens). Apart from the letter m by itself (13 occurrences), the most common is dm (4 occurrences).
     
    
  • You are not allowed to view links. Register or Login to view.: words that contain letter i in any context other than an IN group (68 tokens). The most common is dairin (2 occurrences).
     
    
  • You are not allowed to view links. Register or Login to view.: abnormal words that contain isolated e after an s (28 tokens). The most common is shese (3 tokens).
     
    
  • You are not allowed to view links. Register or Login to view.: abnormal words that did not seem to fit in any of the above categories (213 tokens). Apart from isolated letters like v (7 tokens) and c (4 tokens) --- mainly in the circular text on page You are not allowed to view links. Register or Login to view. --- the most common are da (6 tokens), ackhy, sa, and sha (3 tokens each). Note that the latter are probably the result of misreading y as a in otherwise normal (and common) words.
     
    
    

It is quite possible that, when the VMS is deciphered, we will discover that some of these abnormal words are in fact quite "normal". Indeed, although most "abnormal" words occur only once, some classes of abnormal words may be sufficiently frequent and well defined to deserve recognition in the grammar. One such candidate, for example, is You are not allowed to view links. Register or Login to view., the set of words that have A.IN groups in non-final position.
Conversely, the grammar is probably too permissive in many points, so that many words that it classifies as normal are in fact errors or non-word constructs. See the section about circle letters, for example. For instance, there must be many apparently "normal" tokens which are in fact "Grove words". These could result from prepending a spurious gallows letter to a crust-only normal word (e.g. p + olarar = polarar), or prepending a spurious non-gallows letter to a suitable normal word (e.g. d + chey = dchey). Indeed, it is quite possible that most ot of the normal-looking line-initial words are in fact such "crypto-Grove" words.