+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Analysis of the text (https://www.voynich.ninja/forum-41.html)
+--- Thread: The Impossibility of Double Gallows (/thread-129.html)
RE: The Impossibility of Double Gallows - -JKP- - 06-02-2019
There are many many "q" abbreviations (words starting with q appended by an abbreviation symbol) in languages that use Latin abbreviations. The Latin language has so many short "q" words that stand alone or at the beginnings of other words (or which stand alone but are sometimes written attached to the next word in some documents).
quia, quo, que, quis, quibus, qui, quodam, etc. There were slightly different abbreviations for each one, but I've noticed some scribes were a bit indiscriminate in which ones they chose and some of them used the same abbreviation for several different "q" words. quam and quod are sometimes abbreviated like the example in the clip.
I've written some blogs on 4o. Here are the two that are most relevant:
Variations in the 4o glyph pattern: You are not allowed to view links. Register or Login to view.
Was the 4o glyph in the Diplomatic Ciphers unique? or was it a common Latin convention: You are not allowed to view links. Register or Login to view.
RE: The Impossibility of Double Gallows - Davidsch - 09-02-2019
@JKP: Thank you for your posting on this subject. Since your posting from 2018 has a very recent follow-up and because in my research I do not agree with your conclusions, it seems good to share some points.
I use the text without labels. I've counted
q: 5433 x
qo: 5300 x = 98%
q not qo: 133 x = 2% !
Of the last 133x [q] [not qo] is followed by :
[e] 66 x
[c] 19 x
[k] 19 x
and then 29x on other different letters.
Therefore my conclusions are:
the [qo] is dominant and is a coherent enitity (probably represents in plain text 1 value)
the [q] can not be separated form [qo]: The 133 times are errors and bad transcripts. (there are more arguments but too extensive for here)
That observation is important because from there we can go forward. Keeping all options open, for me that is, does not lead to a specific direction and it is impossible to follow up on all possible options. Anyway that is my way of working.
Then, you speak of [qo] followed by another [o]. Yes that seems to be true in 32 cases.
That is only 0,6% ! and therefore (there are more arguments but too extensive for here)
are also mistakes and have no specific meaning to me. (Which could be made deliberate, but that's another theory)
RE: The Impossibility of Double Gallows - -JKP- - 09-02-2019
Quote:davidsch wrote: ...the [q] can not be separated form [qo]: The 133 times are errors and bad transcripts. (there are more arguments but too extensive for here)...
I respect your point of view, but I don't think the places where q is placed ahead of something different than "o" represent transcription errors (if it does, that suggests there are many transcription errors in the VMS).
As you know from my more recent posts, I think the "o" prefix is very frequently prepended by q but not always (that it is not a requirement for an "o-token" to follow the q[font=Sans-serif] or for [font=Eva]q to only precede "o-tokens").[/font] q doesn't even have to be at the beginnings of words.
In English, we have combinations like "th" or "qu" that are very frequent, but it does not mean that "u" is required to follow "q" ([font=Eva][font=Sans-serif]loanwords like faqir, shish qabab, or alternate spellings like cinqfoil have "q" not followed by "u".[/font]), it just does so most of the time.[/font]
If [font=Eva][font=Sans-serif][font=Eva]q[/font][/font]o were a coherent entity (and ONLY supposed to be a biglyph), then I don't think the macron glyph-shapes would have been added next to q in several places (if the VMS macron has the same function as in Latin languages, then it represents omitted letters between q and "o").
I am always willing to change my mind if I see good evidence to the contrary... but for the moment[font=Sans-serif], I'm comfortable taking the position that [font=Eva][font=Sans-serif][font=Eva]q[/font][/font] is distinct to itself and [/font][/font][/font]what we see is [font=Eva][font=Sans-serif][font=Eva][font=Sans-serif][font=Eva][font=Sans-serif][font=Eva]q[/font][/font][/font][/font][/font][/font] prepended to "o-tokens" rather than [font=Sans-serif][font=Eva][font=Sans-serif][font=Eva][font=Sans-serif][font=Eva][font=Sans-serif][font=Eva]q[/font][/font][/font][/font][/font][/font][/font][/font]o prepended as a unit to a shorter token.
There's always the possibility that it's both, just as "th" in English is sometimes treated as a unit (in fact, it used to have its own glyph) and also as separate characters.[/font][/font]
RE: The Impossibility of Double Gallows - DONJCH - 10-02-2019
The question of low percentage variants being errors has come up before.
I have also asked what percentage is to be used below which a variant can be considered to be an error.
There was no response.
In a medical laboratory setting, in which I worked as a scientist for 40 years, transcription error rates are intensively studied.
Error rates in the order of 2% could not be tolerated. When one is handling hundreds or thousands of figures per day, the death rate would be horrendous.
RE: The Impossibility of Double Gallows - Davidsch - 10-02-2019
The afore mentioned counts and percentages can be discussed, but really, it is an accurate count you can crosscheck.
Because the counts and very very low percentages are the basis and fundaments of my current conclusions,
it would be possible to discuss from that position on forward that
1. It could be established that the scribe made much errors and mistakes, if you are open for it. Then specific textparts need to be discussed.
2. this is only useful if there are mutal agreeable assumptions (read: starting points).
3. And then there is the method used for comparing textparts must be accepted on every end.
RE: The Impossibility of Double Gallows - nickpelling - 11-02-2019
(09-02-2019, 08:34 PM)-JKP- Wrote: You are not allowed to view links. Register or Login to view.[font=Sans-serif]If [font=Eva][font=Sans-serif][font=Eva]q[/font][/font]o were a coherent entity (and ONLY supposed to be a biglyph), then I don't think the macron glyph-shapes would have been added next to q in several places (if the VMS macron has the same function as in Latin languages, then it represents omitted letters between q and "o").[/font][/font]
If the macron-like overbar is typical of those used in the fifteenth century, it indicates that the letter following (not preceding) the o has been omitted / nasalized / whatever. So my own suspicion (before even The Curse of the Voynich) was that this might be a case where the writing system had somehow placed EVA o after word-initial EVA qo-, and so what we may well be looking at in those cases was a contracted EVA 'qoo'.
However, whatever is going on with the macron-like bar over the o in qo in those very few instances (aren't there just two in the entirety of the VMs corpus? You are not allowed to view links. Register or Login to view. and one other?), it's a very small sample to be using to try to draw inferences about the overall writing system.
For what it's worth, my various estimates for the scribal error rate (e.g. when aiin ends up oiin, etc) all sit relatively consistent at about one mistake per line of text, which makes me pretty comfortable with the suggestion that EVA qo is a genuine bigram and that the remainder of the q[^o] pairs are scribal copying errors.
RE: The Impossibility of Double Gallows - -JKP- - 11-02-2019
Why do they have to be copying errors?
In many languages, there are combinations that are common and variations of the same combinations that are rare. It doesn't have to be one or the other. It might be a common combination that also has some rare variants.
Take for example, the loanwords I mentioned in the previous post. In English "qu" is a common combination, but there are loanwords with "q" without "u". Loanwords were not at all uncommon in medieval texts and spelling variants were very common. Plus, the way things were abbreviated, common combinations might turn into rare combinations if they were abbreviated only once in a while (this is not uncommon either, the word might be spelled out most of the time and then suddenly the scribe abbreviates it, either due to haste or lack of space).
Even if it were a synthetic language, common combinations with rare variants might occur.
RE: The Impossibility of Double Gallows - Wladimir D - 11-02-2019
Nick.
In my opinion, it is necessary to take into account that when writing “4o” different diacritical marks are used, and in different positions, which can mean different meanings. I tend to support the point of view of JKP.
RE: The Impossibility of Double Gallows - nickpelling - 11-02-2019
(11-02-2019, 08:24 AM)Wladimir D Wrote: You are not allowed to view links. Register or Login to view.In my opinion, it is necessary to take into account that when writing “4o” different diacritical marks are used, and in different positions, which can mean different meanings. I tend to support the point of view of JKP.
For the qo + flat macron/overbar, there are only two definite examples (the third is debatable): that's not a lot of data to draw inferences from, considering how many qo's there are in the text.
For the qo + curved macron/overbar, the four examples are all in Currier A pages (as I recall), so it seems likely to me to be an early writing system experiment that was dropped along the way. If we knew the order in which the pages had originally been written, that would be very helpful.
If it was a universal mark or part of the system (as you seem to assume), we'd expect to see it throughout the text: but we don't.
RE: The Impossibility of Double Gallows - nablator - 11-02-2019
(11-02-2019, 01:16 PM)nickpelling Wrote: You are not allowed to view links. Register or Login to view.For the qo + curved macron/overbar, the four examples are all in Currier A pages (as I recall), so it seems likely to me to be an early writing system experiment that was dropped along the way.I count a few more on f. 5v.3, 19r.7, 22v.1, 49r.2, 49r.14, 49r.16, 49v.10 (or 20 if the first column is counted separately), 87r.15, and only one in Currier B: You are not allowed to view links. Register or Login to view..
Quote:If it was a universal mark or part of the system (as you seem to assume), we'd expect to see it throughout the text: but we don't.If the purpose is not to confuse would-be decipherers - there is an even more dubious round macron on daiin on line f49r.15 - rarity would be a problem only if there were any reason to assume that these rare occurrences all represent a common pattern in the underlying language and that this pattern can only be expressed in this way, or, if several ways are possible, that equal (or a fair share of) opportunity would be given to all of them... That's a lot of assumptions.
These rare occurrences are going to be solved from context when the rest is understood, so they don't really matter.