Hi all,
firstly I apologise for not citing any references to previous emails as I cleared out my folders on Yahoo yesterday - then realised that I had a couple of points to make.
Re. the code book approach where there might be a sort of tree structure to words, imperfectly remembered as
animal a, mammal b, horse c
so to say a horse use the word abc. One point that was stated was that this doesn't account for the repetitions of words. There is an answer here. If the maker of the code book wanted to make comparison, e.g. good, better, best s/he could either have separate words for all of these, or, have a special suffix or prefix to use, or, just repeat the word, so if fgh = good then fgh fgh = better and fgh fgh fgh = best.
This code book approach could also account for some of the features of the vms that linguists find difficult to explain in terms of languages. These were enumerated recently (sorry, haven't got the email) and Frogguy made a sterling effort to find examples in different languages. Now, it could be that the vms writer decided to pick a language or invented one that had all of these unusual features or perhaps a simpler explanation is that these are all an artifact of the code book approach. Some of these features are the way certain letters are found word initial like 4 and 40 and others word final like Currier K, etc.
One thing about Currier K, where it is almost always found as word final, it is also mostly line final and I cannot see how this would fit with the code book approach.
D'Imperio has an example of a similar type of code on page 118 as the Sylvester Code.
Regards,
Brett Cotton