+- 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: Repeating words across line boundaries (/thread-2408.html)
Repeating words across line boundaries - bi3mw - 27-06-2018
Examining the structure of the lines raised the question of whether word repetitions occur across line boundaries. I mean the following structure (one letter for a word):
Line 1: a b c
Line 2: c d e
Line 3: e f g
...
I have examined 5 folios for the sample; f 46r, f56r, f78r, f84r, You are not allowed to view links. Register or Login to view.
There were only two hits, both on folio You are not allowed to view links. Register or Login to view. ("qokeedy" and "dar"). In total, all reviewed pages together have 1249 words. For 37043 words in the manuscript 59 hits can be expected according to the pattern described above. These are not many, but my expectation that such a recurrence never occurs has been refuted.
You are not allowed to view links. Register or Login to view.
RE: Repeating words across line boundaries - nablator - 27-06-2018
Interesting. I did some stats on my own EVA transcription.
I found 8 on the main text paragraphs only, without labels, circles, isolated lines:
13r.2
47v.11
78v.22
82r.3
84r.10
84r.27
87r.13
90r2.5
On the other hand there are about (it depends on the choice of transcription of course) 260 word repetitions inside the 4045 lines from these same main text paragraphs.
If the proportion were the same, and since there are 30211 word boundaries in these paragraphs, there should be about 35 repetitions across line boundaries. Maybe the fact that the first and last words of lines are statistically different from each other and from the other words (first is longer, last more often contains EVA-g or m) explains the gap: they are less likely to match.
RE: Repeating words across line boundaries - bi3mw - 27-06-2018
Thanks nablator, the magnitude of word repetition of this type seems to be correct. Your comment that it might be due to the line structure is an interesting point.
RE: Repeating words across line boundaries - Emma May Smith - 27-06-2018
This kind of thing fascinates me. I've studied the line break statistics in the past with Marco Ponzi, and we've found so many interesting things (far more than have been described). I would be very interested in anything you find in the future.
Idea: have you tried matching the last word of a line minus the last glyph with the first word of a line minus the first glyph?