The Voynich Ninja

I know that we have a You are not allowed to view links. Register or Login to view. on gallows characters but I have a theory about benched gallows (cKh, cTh, cPh, cFh) that has statistical support.

In my You are not allowed to view links. Register or Login to view. I showed that <qo> is followed by <k> or <t> 80% of the time in the VMS corpus (4 out of 5 times). It is followed by <l> only 5% of the time, <e> 2% of the time, and <ch> less than 1% of the time. So, there is a clear pattern:  <qo> comes in front of a gallows character.

There are a very small number of times where <qo> comes before a benched gallows like [cKh] or [cTh]. So there are three options for interpretation:

• 1) The benched gallows can be deconstructed as <qokch> or <qotch>, which makes statistical sense (qo before gallows is normal).
  
• 2) The benched gallows can be deconstructed as <qochk, qocht> or <qoeke, qoete>, but all of these combinations together only appear 9 times in the VMS.
  
• 3) The benched gallows are independent sounds, but they rarely follow <qo> (about 1.7% of the time, or less than 1-in-50 times, that <qo> is used).
  
• EDIT: Anton proposed a fourth system You are not allowed to view links. Register or Login to view. (which may or may not work in conjunction with #1 or #2)
  

The first option is the most statistically likely, and in addition, there are plenty of times where a benched gallows and its [gallows+ch] equivalent are interchangeable, which also lends support to option 1:




So, I would conclude that the benched gallows are really [gallows followed by ch], but I could be wrong and I'm open to counter-arguments  

(02-10-2016, 02:25 AM)ThomasCoon Wrote: You are not allowed to view links. Register or Login to view.I know that we have a You are not allowed to view links. Register or Login to view. on gallows characters but I have a theory about benched gallows (cKh, cTh, cPh, cFh) that has statistical support.

In my You are not allowed to view links. Register or Login to view. I showed that <qo> is followed by <k> or <t> 80% of the time in the VMS corpus (4 out of 5 times). It is followed by <l> only 5% of the time, <e> 2% of the time, and <ch> less than 1% of the time. So, there is a clear pattern:  <qo> comes in front of a gallows character.

There are a very small number of times (less than 100) where <qo> comes before a benched gallows. So there are three options for interpretation:

• 1) The benched gallows can be deconstructed as <qokch> or <qotch>, which makes statistical sense (qo before gallows is normal).
  
• 2) The benched gallows can be deconstructed as <qochk, qocht> or <qoeke, qoete>, but all of these combinations together only appear 9 times in the VMS.
  
• 3) The benched gallows are independent sounds, but they rarely follow <qo> (about 1.7% of the time, or less than 1-in-50 times, that <qo> is used).
  

...

Consider this as part of your analysis... a letter is sometimes inserted under the bench, between the gallows and the end of the bench. It doesn't happen often, but it happens often enough that I'm pretty sure it's deliberate rather than some kind of error correction.

The scribe made a deliberate effort to write either kch or cKh, so I don't see why the distinction should be considered meaningless.  In fact it seems that the reason why the gallows letters are tall and why ch has the shape that it does is precisely so that the combination cKh will be possible.  That the script was apparently designed with this combination in mind suggests that it's significant.

Also, you cite the fact that they often do not occur in the same place as evidence that they are equivalent, but then also cite the fact that they sometimes do occur in the same place as evidence that they are equivalent... it's hard to imagine what evidence that they are not equivalent would look like.

(02-10-2016, 03:59 AM)Sam G Wrote: You are not allowed to view links. Register or Login to view.The scribe made a deliberate effort to write either kch or cKh, so I don't see why the distinction should be considered meaningless. 

I'm not sure I agree - because someone made an effort to write something differently does not conclude that there is a distinction. This is observable: for example in English we can write <it's> or <it is> with the same meaning, but with distinct formats.

Quote:Also, you cite the fact that they often do not occur in the same place as evidence that they are equivalent, but then also cite the fact that they sometimes do occur in the same place as evidence that they are equivalent... it's hard to imagine what evidence that they are not equivalent would look like.

Sorry if that was vague, maybe this will clear it up: the benched gallows appear very rarely after qo- (they follow <qo> about 2% of the 5300 times that <qo> is used), but when they do appear, very often a <gallows>+<ch> parallel word can be found (maybe even every time - I'll have to check).

-JKP- Wrote:Consider this as part of your analysis... a letter is sometimes inserted under the bench, between the gallows and the end of the bench. It doesn't happen often, but it happens often enough that I'm pretty sure it's deliberate rather than some kind of error correction.

Thanks I'll have to check that out - I am not sure about these situations, though I see them every-so-often when I'm transcribing.

Sam G Wrote:In fact it seems that the reason why the gallows letters are tall and why ch has the shape that it does is precisely so that the combination cKh will be possible.

That is an interesting observation, but respectfully I don't think I agree - I don't see the same deliberate creation.

The set of options listed in the title post misses the option of complex glyphs being a superposition of simpler glyphs.

Imagine that c stands for number X, h stands for number Y, k stands for number Z. Then ckh could just stand for number X+Y+Z (or X*Y*Z, or otherwise). Something like the numeric monastic cipher described in the book by D. Kahn.

By the way, in the second screenshot of the title post it is yodaiin, and not qodaiin.

They could also be phonetic digraphs - double letters which are considered to be a single letter in the orthography.

Examples from modern languages would be the (old) double ll or the ch from Spanish, both of which were considered to be single letters in the alphabet; or ij in Dutch.

If we take k and ch to be distinct phonetic letters, then there could be a mixture of phonetics to allow the glyphs to be combined in a certain fashion. That also accounts for their regular placing if we assume Voynichese to be a transliteration of a language - or the phonetics being somehow beyond the encryption process. But given that all glyphs in Voynichese appear in regular patterns, we're only considering a subset of the language when looking at these bench glyphs.

Back again. I looked up all the unique [qo+benched gallows] vords to see if there are parallel [qo+gallows+ch] vords.
Here are the results: there are 41 unique [qo+bench] vords in the VMS, and 32 of them have [qo+gallows+ch] parallels - that is a 78% match:

qocthy -->     qotchy
qocthol -->    qotchol
qoctholy -->    -------
qocthey -->    qotchey
qocthedy -->    qotchedy
qoctheol -->    qotcheol
qocthey -->     qotchey
qocthes -->     qotches
qocthdy -->    qotchdy
qoctho -->    qotcho
qocthsy -->    -------
qocthody -->    qotchody
qocthor -->    qotchor
qoctheody -->    -------

qockhy -->    qokchy
qochhdy -->    qokchdy
qockhhdy -->    -------
qockhey -->    qokchey
qockhom -->    -------
qockhor -->    qokchor
qockho -->    qokcho
qockheol -->    qokcheol
qockhhy -->    --------
qockhedy -->    qokchedy
qockhol -->    qokchol
qockheey -->    qokcheey
qockhos -->    qokchos
qockhoy -->    --------
qockhol -->    qokchol
qockhody -->    qokchody
qockheor -->    qokcheor
qockhal -->    qokchal
qockhed -->    qokched
qockheos -->    --------
qockheedy -->     qokcheedy

qocphy -->    qopchy
qocphody -->    qopchody
qocphdy -->    qopchdy
qocphee    -->    -------
qocphey -->    qopchey
qocfhey -->    qofchey

So, while this doesn't conclusively prove that [benched gallows] = [gallows+ch], I can offer three pieces of objective, statistically-backed evidence that support this interpretation:

1) There is a clear statistical tendency for a gallows to follow [qo]
2) 78% (almost 4 out of 5) [qo+benched gallows] vords have a [qo+gallows+ch] parallel.
3) [qo+benched gallows] words can appear in similar environments as [qo+gallows+ch] words.

Anton, you are correct that that is a possibility - I will edit my post. Also, what you propose may not necessarily be mutually exclusive to options 1 and 2 (both things could occur simultaneously). Also, in the screenshot, the second <qodaiin> word is somewhere in between <y> and <q> - it doesn't have the characteristic tail of <y> nor the characteristic loop of the <q>. So, it is a bad example, and I should have found a better one.

david, I vaguely account for this in option 3: that two separate characters can be written together to represent a unique sound. Was this what you were going for when you said "If we take k and ch to be distinct phonetic letters, then there could be a mixture of phonetics to allow the glyphs to be combined in a certain fashion"?

(02-10-2016, 05:33 PM)ThomasCoon Wrote: You are not allowed to view links. Register or Login to view.

(02-10-2016, 03:59 AM)Sam G Wrote: You are not allowed to view links. Register or Login to view.The scribe made a deliberate effort to write either kch or cKh, so I don't see why the distinction should be considered meaningless. 

I'm not sure I agree - because someone made an effort to write something differently does not conclude that there is a distinction. This is observable: for example in English we can write <it's> or <it is> with the same meaning, but with distinct formats.

Perhaps, but certain things about the VMS like the low entropy suggest that there's not a lot of orthographic variation going on in general.  And I still think my point about the script apparently being designed with cKh in mind is a strong indicator that it's meaningful.

Quote:

Quote:Also, you cite the fact that they often do not occur in the same place as evidence that they are equivalent, but then also cite the fact that they sometimes do occur in the same place as evidence that they are equivalent... it's hard to imagine what evidence that they are not equivalent would look like.

Sorry if that was vague, maybe this will clear it up: the benched gallows appear very rarely after qo- (they follow <qo> about 2% of the 5300 times that <qo> is used), but when they do appear, very often a <gallows>+<ch> parallel word can be found (maybe even every time - I'll have to check).

I still see the same problem - a similarity is cited as evidence of equivalence and a difference is cited as evidence of equivalence.  What would be evidence that they are not equivalent?

I agree that there is probably some kind of relationship between the two - that much seems obvious - but I do not see a reason to think that they are exactly the same.

---

Also, off the top of my head, words like cThol, cThor, and cThy are going to be way more common than tchor, tchol, and tchy, especially in the Herbal A section.

(03-10-2016, 04:53 AM)Sam G Wrote: You are not allowed to view links. Register or Login to view.Perhaps, but certain things about the VMS like the low entropy suggest that there's not a lot of orthographic variation going on in general.  

That is true, there is low entropy. I believe that can be attributed to the You are not allowed to view links. Register or Login to view.. In addition, benched gallows may have even contributed to low entropy if those statistics were calculated with an EVA transcription: [ckh, cth, cph, cfh] would be seen as predictable 3-letter combinations. But I'm not sure how entropy was calculated (maybe not with EVA), so I could definitely be wrong.

Quote:And I still think my point about the script apparently being designed with cKh in mind is a strong indicator that it's meaningful.

That is an interesting observation. I think on your side, one thing which may be hard to prove is that the author didn't stumble upon the cKh combination after he began using script  - there are many examples of signs ending up fitting together which were never intentionally designed to be (Mayan Hieroglyphs give many examples). I believe the VMS author designed the two signs separately and later realized he could fit them together if he elongated the ch a little.

Either way, both of us seem to be arguing from observation and conjecture, and this is problematic. Without stone-cold numbers to support or deny what we see, we will continue to go in circles forever and the VMS will never be solved ("I think that [a] and [o] are the same" - "I think that [l] is a form of [y]" - "I think that [ch] and [k] weren't intentionally designed to make [ckh]").

We need statistics

Quote:I still see the same problem - a similarity is cited as evidence of equivalence and a difference is cited as evidence of equivalence.  What would be evidence that they are not equivalent?

That is fair. I see what you are saying. I have an idea to solve this problem but I'll have to do some manuscript-wide distribution searches and get back later.

Breakdown by Currier A and B might (or might not) be suggestive. (Whether one of them mostly uses composite gallows while the other does not.)