Re: VMs: Word Endings

[Gordon Rugg <g.rugg@xxxxxxxxxxxxxx>]

Dear Glen,

Interesting.

I suspect that my previous post wasn't as clear as I'd intended (pressure of
admin...) and that your results might be strengthened by what I intended.

In Latin, word endings in some contexts are completely predictable, and in other
contexts are not predictable. After the preposition "ex", for instance, the noun
in question will end in the ablative. What I was trying to say was that a single
character, such as "y", could in that setting substitute for all the relevant
ablative endings, such as "o", "a", "e" "is","ibus" etc (assuming
singular/plural being already known from surrounding context). That would allow
"y" to subsitute unambiguously for at least half a dozen different characters or
syllables at the end of some words. The "y" could also be used unambiguously to
represent a completely different set of characters or syllables at the start of
some words (e.g. "con", "com", "co"), or more creatively as a marker of the
number and case of the ending, to record information about the key being used,
etc. The "y" could in addition be used for a free standing word (e.g. "et").
Each of these uses would be completely unambiguous in its context. However, in a
different context the same grammatical ending could be ambiguous/unpredictable,
and would need to be spelled out in full as "ibus" or whatever.

I continue to have reservations about this as an explanation - as others have
pointed out, the medial stretches of the Voynichese "words" are rather short -
but it might help someone towards a solution.

Best wishes,

Gordon


GC wrote:

> Hey all,
>
> Pardon me if all this has been done before in numbers, but I'd like to
> repeat the process using "pictures" (glyphs) so pure numbers don't lose
> anybody who isn't fluent in EVA (me especially). :-)
>
> Someone was asking if a study had been done on word endings.  I think I saw
> something on Jorge's site on this awhile back, and there have been numerous
> papers and writings on the subject taken from one vantage or another, but I
> think it might do good to revisit some of these again, as the concept of
> word beginnings and endings are an important feature on which we base a lot
> of our arguments.
>
> Because of recent comments to the nature of "the VMS has so few word
> endings", I've posted three short studies of these in separate files.
> There's a lot to consider on this topic, so I'll just touch on a few and see
> if we can draw enough different perspectives to look into this a little
> deeper.  Follow me to http://voynich.info/vgbt/xcrptn/first_last.pdf
>
> I record 92 separate glyphs in the first six quires, and have already dealt
> with my perceptions of their classifications.  (I know some are probably
> part of the standard set, etc.)  Out of that 92 however, 66 (72%) are found
> at least once at the beginning of a word. Out of these 66, 44 occur more
> than once at the beginning of a word (66.6%).  Moreover, we have what
> appears to be four separate statistical breaks in the data, which might
> allow us to investigate a little further.  This data is far too simple to
> need a graph, but deeper studies would certainly benefit.  Enough of the
> beginnings however, as I only have time in one evening to explore the
> question of word endings in any detail, and this is certainly not enough
> time to cover even the basics.
>
> Go to the bottom of this page, and you'll see a count of word-ending glyphs.
> Here we have 58 glyphs that serve as word terminators, and 44 of these (76%)
> occur more than once.  Talk about your statistical breaks, though, here we
> have a big one.  Out of 8,175 words, 6,392 (or 78%) end in one of four
> glyphs.  These individual numbers are important if we are forming some
> perception about beginning, median, and terminating "strings".
>
> Follow me now to http://voynich.info/vgbt/xcrptn/last2.pdf where we have a
> listing of the last two glyphs of every word.  Words two or more in length
> have 275 unique word endings when considering the last two glyphs.  Our
> three big ones are am, 89, oe, and os (don't rightly know the EVA for "os",
> but here you can see it visually ).  After this set we have another major
> statistical break, which we might have to take back and compare to the
> single-glyph examples in order to determine if these are "suffixes".
>
> >From the first file we've seen that there are 1,097 words ending in m.
> There are 994 words ending in am, which is 12.2% of all words, and 90.6% of
> all words ending in m.  We have a positive ID on correlation between a and
> m.  76 occurrences of am are stand-alone words, so the count of am as
> "suffix" would be 918.  How a "suffix" is also a stand-alone word is one I
> haven't quite yet grasped.
>
> To continue this vein I'll jump ahead to
> http://voynich.info/vgbt/xcrptn/last3.pdf where we find that our top runner
> is Nick's 8am <daiin>, with a whopping 484 count as a word-ending string. To
> reiterate, we have 994 words ending in am, which is 90.6% of all words
> ending in m.  44% of all words ending in m are words also ending in 8am.
> The problem here is that 366 of those 484 (about 76%) of the occurrences of
> 8am are stand-alone words.  When looking for "suffixes", only 118, or 1.4%
> of all the 8,175 words have 8am as a "suffix".  (yes Nick, I know this helps
> your theory.)
>
> I only have time tonight to look at the next runner in the series of word
> endings, the 89.  Our 9  tops the list of word terminators, at 2,930
> occurrences, or 36% of all word terminators.  515 of these are stand-alones,
> and if you will look at the first page of first and last glyphs, you'll see
> that 515 words also begin with 9, meaning that 9 occurs 515 times as a
> single-glyph word, but NEVER begins a compound word in these 6 quires.
> (Finally I get to say NEVER when talking about the VMS! 9 has a few other
> "nevers" attached to it as well). The ending 89 occurs 964 times.  This is
> 33% of the words ending in 9, or if we negate the 515 stand-alones, this is
> 40% of all words two or more in length that end in 9.  It's not the 90.6% of
> the am words, but it's a pretty good chunk. 120 of the 964 occurrences of 89
> are two-glyphs words, or 12.6%, so our number goes down a bit.  We jump
> ahead and look at the list of the last three glyph occurrences, and c89 is
> right up there below 8am.  There are exactly -0- stand-alone occurrences of
> the word c89.  The next on the list which includes 89 is o89, which has only
> 13 occurrences, a statistically insignificant number.  89 rightly qualifies
> as a "suffix" or "significant group", as does am.  8am, if we consider word
> construction, is rightly a word of its own, or it can also be considered
> that 8+am is a consistent and common word used throughout these six quires.
> That is, until you look at the occurrences of 8am at the beginning and in
> the middle of words.....
>
> That's all I can do tonight, but I hope this helps add some definition, and
> I hope it also tempers the common perception that the VMS word beginnings
> and endings are "very few".  Currier, Tiltman, D'Imperio and others did some
> major work in this area, but it certainly doesn't hurt to revisit it once in
> awhile to demonstrate that the problem is more complex. Ah shoot.. I was
> going to finish proofing and post quire 7 tonight, but I got caught up in
> this train of thought.
>
> I'd like to continue in this, but with some definitions set in place.  What
> makes something a "prefix" or "suffix"?  By definition, these should not
> occur in the middle of words.  (Yes, I have considered words like "Augustus"
> in making that statement.)  What lines do we draw to determine beginning,
> median, and word endings?  And my big question - why do these so-called
> "prefixes" and "suffixes" operate well in pairs of glyphs, but when they
> reach three-glyph length, either fall off or turn into a complete word by
> itself, like 8am does? (Yes Nick, I know, but humor me anyway!).
>
> GC
>
> ______________________________________________________________________
> To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
> unsubscribe vms-list

______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list