+- 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: Gallows distribution (/thread-2051.html)
Gallows distribution - davidjackson - 16-08-2017
Can anyone out there confirm or deny this?
Stolfi mentions in a You are not allowed to view links. Register or Login to view. that he found the following gallows distribution:
Quote: I just noticed a curious coincidence:I can't be bothered to dig into this right now - does anybody have the stats to prove or disprove this antique statement?
>
> total *occurrences* of words (tokens) with
>
> 0 gallows .... 17363 (49.4%)
> 1 gallows .... 17443 (49.6%)
> 2 gallows .... 323 (0.9%)
> 3 gallows .... 3
> Many (if not all) of the 2- and 3-gallows words are probably due to
> omission of word spaces by the transcribers. Other data errors may
> have injected a few percent of noise in these figures.
>
> Still, the coincidence is intriguing. It seems safe to assume that a
> "correct" Voynichese word can have at most one gallows; so we have
> almost exact 50-50 split between 0-g and 1-g words.
Even curiouser:
w/o gallows with gallows
+--------------+--------------+
w/o tables | 8772 (25.2%) | 9016 (25.9%) |
+--------------+--------------+
with tables | 8591 (24.7%) | 8423 (24.2%) |
+--------------+--------------+
These are counts of tokens (word instances) in the whole majority-vote
transcription; minus key sequences, labels, unreadable/contentious
tokens, and the 326 tokens with two or more gallows.
The "gallows" are the EVA letters [ktfp], including any platforms
("ct", "cth", "ith") and isolated "e" suffixes ("te", "cthe", etc.).
The "tables" are the letters "ch", "sh", "ee", and any isolated "e"s
that are not attached to a gallows letter.
RE: Gallows distribution - -JKP- - 16-08-2017
Quote:...
> Still, the coincidence is intriguing. It seems safe to assume that a
> "correct" Voynichese word can have at most one gallows; so we have
> almost exact 50-50 split between 0-g and 1-g words.
...
I can't give examples while I'm working, but I would strongly disagree with this statement.
Voynichese tokens can certainly have more than one gallows and incidences of those with more than one are not transcription errors. I also believe that combination gallows (which are done in a very systematic way) may be a "shorthand" way of combining two gallows into one.
RE: Gallows distribution - Emma May Smith - 16-08-2017
I think Stolfi's definition of 'gallows' and 'tables' is not typical or useful beyond his research at the time. They are specific groupings related to the paradigm he was working on. He mentions later in the same message that different figures could be obtained were the groups to be defined differently.
RE: Gallows distribution - Anton - 17-08-2017
I recall that the question was discussed somewhere in the forum (should check the main gallows thread). I agree with JKP - there definitely are vords with more than one gallows. Vords exhibiting t coverage would also count for 2 gallows in transcriptions, but I don't mean them.
RE: Gallows distribution - MarcoP - 18-08-2017
I computed some numbers that are certainly different from what Stolfi intended: not all his definitions are clear to me.
I considered all the words in Takahashi's transcription (labels included).
I counted as a table all sh,ch,e sequences: I tried excluding those adjacent to gallows, but this produces much lower numbers for "words with tables".
The four sets described by Stolfi have similar numbers, but in Currier A gallows words are less common than words with no gallows, while in Currier B they are more common (47% Currier A, 53% Currier B).
In all cases, the presence of tables seems slightly correlated with that of gallows.
Code:
______________________________ all CUR.A CUR.B
(blue) tables gallows 12079 3229 7782
(orange) noTables gallows 7255 2206 4799
(yellow) tables noGallows 8825 3072 5435
(green) noTables noGallows 8946 3069 5692