+- 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: Can LAAFU effects be modeled? (/thread-4869.html)
RE: Can LAAFU effects be modeled? - ReneZ - 04-09-2025
(04-09-2025, 02:06 AM)RobGea Wrote: You are not allowed to view links. Register or Login to view.It may be about time for a reassessment of LAAFU. what it entails and how it is defined.
I think so too. There are some contradictory results in the above posts.
One has to be careful with comparison texts that are printed, because these have been spell checked and typeset, so are the results of completely different processes and rules.
RE: Can LAAFU effects be modeled? - anyasophira - 04-09-2025
Yay more line talk! Does this mean that it would take several voynich vords to create one plaintext word?
-Anya
RE: Can LAAFU effects be modeled? - oshfdk - 04-09-2025
(04-09-2025, 12:33 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.One of the problems with the Voynich MS text is its relative lack of repeating sequences of words. This is aggravated in case one assumes that the text is a verbose cipher.
I think a verbose cipher that uses some source of entropy (dice, cards) to add random nulls (Mark Knowles style) or pick one of the few ciphertext variants (the Naibbe) won't necessarily produce extra repeating sequences and, on the contrary, can effectively hide existing ones. Only one-to-one context-free verbose ciphers would need to explain the lack of repeating sequences.
RE: Can LAAFU effects be modeled? - MarcoP - 04-09-2025
(03-09-2025, 12:42 PM)pfeaster Wrote: You are not allowed to view links. Register or Login to view.2. Are line-end features abbreviations employed when the writer was running out of space? Maybe -- but my sense is that, in practice, abbreviations didn't typically cluster at line-end in manuscripts of the period, so this would be a stranger explanation than it might seem at first glance.
Years ago, I mentioned Vat.Lat.410 on You are not allowed to view links. Register or Login to view.. The example is from You are not allowed to view links. Register or Login to view..
My impression is that, in manuscripts with only a few abbreviated words, they do tend to cluster at line end. But most early 15th century manuscripts are heavily abbreviated and I agree that there usually is no clear clustering. Anyway, my opinion is that "abbreviation clustering" could play a role with EVA:m,g.
(04-09-2025, 02:06 AM)RobGea Wrote: You are not allowed to view links. Register or Login to view.It may be about time for a reassessment of LAAFU. what it entails and how it is defined.
For example, if my understanding is correct, the long 1st word and short last word of vms lines, according to Elmar Vogt can be also be found in 'Tomsawyer'.
You are not allowed to view links. Register or Login to view.
This passage from You are not allowed to view links. Register or Login to view. could explain what happens with English:
Quote:Word lengths in most meaningful texts are negatively autocorrelated: that is, long words tend to be interspersed with short words (long-short-long-short). By contrast, the VMS exhibits positive autocorrelation (long-long-short-short).Short words can easily fit at the end of lines, which causes the following longer word to appear at the start of the following line. What happens with Voynichese is less clear.
Another great line-effect is what Tavie discussed at Voynich Day 2024.
(08-08-2024, 01:41 AM)tavie Wrote: You are not allowed to view links. Register or Login to view.Scribe 1 in Herbal A really dislikes q-q vertical pairs despite loving q as a Line Start initial. Scribe 2 in Balneological also shows a distaste for q-q.
Basically, though line-initial q- is crazily frequent, it almost never occurs on consecutive lines, so that alternating vertical patterns can be observed. E.g. check You are not allowed to view links. Register or Login to view., or the You are not allowed to view links. Register or Login to view..
RE: Can LAAFU effects be modeled? - ReneZ - 04-09-2025
(04-09-2025, 12:33 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.These are roughly as follows:
for first lines, 85% of cases start with one of 'p t k f' (in descending order of frequency).
For all other lines, 85% of cases start with one of 'd s y o q' (in descending order of frequency). Note that the 's' combines the stand-alone s and the left half of Sh, which is not optimal. (I computed these stats using the RF1b file, and both numbers were in fact 84% when rounded to zero decimal places).
There is a bit more to this. The frequency distribution of the first group goes down rather steeply, while the frequency distribution of the second group is almost flat.
Since You are not allowed to view links. Register or Login to view. was brought up just above, the first five characters in the sequnce in the margin are almost the same as these five characters, the only difference being that the y is replaced by a second s.
This has had me scratching my head for the last four days.
Of course, there is a problem: the fact that s also includes the cases of Eva-sh, and the sequence on You are not allowed to view links. Register or Login to view. also has that one a bit further down...
RE: Can LAAFU effects be modeled? - ReneZ - 04-09-2025
(04-09-2025, 08:07 AM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.I think a verbose cipher that uses some source of entropy (dice, cards) to add random nulls (Mark Knowles style) or pick one of the few ciphertext variants (the Naibbe) won't necessarily produce extra repeating sequences and, on the contrary, can effectively hide existing ones. Only one-to-one context-free verbose ciphers would need to explain the lack of repeating sequences.
With or without a verbose cipher, one would need something that reduces the repeating phrases, and one can think of numerous examples. With a (consistent) verbose cipher, the problem is just a bit bigger.
Problems is, that the things that break up consistent word sequences will also tend to increase entropy.
- Spelling variations (it should not be too difficult to do an experiment with that)
- Homophonic cipher
- Arbitrary nulls
(note that consistent use of nulls - if ever done - defeats their purpose and turns them into a verbose cipher.
We'd need a well-balanced mix of things that reduces entropy and also breaks up standard phrases.
RE: Can LAAFU effects be modeled? - Mauro - 04-09-2025
(04-09-2025, 10:54 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.We'd need a well-balanced mix of things that reduces entropy and also breaks up standard phrases.
A possibility which came to my mind recently is a many-to-many substitution cipher. I don't like it at all, first because it's a practically unfalsifiable theory, and second because it makes the VMS undecipherable by construction (even if it would not be gibberish, after all). However I'd expect it to reduce n-grams and word entropies (say, 'et' 'aut' 'ab' could all be ciphered as 'aiin') and to break up standard phrases (because the same source word could be ciphered in many different ways). Just a fringe idea.
RE: Can LAAFU effects be modeled? - Jorge_Stolfi - 04-09-2025
(04-09-2025, 12:33 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.One of the problems with the Voynich MS text is its relative lack of repeating sequences of words.
Compared to what text? What are the top "repeating sequences of words" in that text?
These are the most common two- and three- word sentences in the Herbal section of Culpeper's Herbal (17th century English). The phrases do not span paragraph or section boundaries. Capitalization was ignored. The counting was stopped after collecting the first 10000 phrases:
136 of.the | 17 vertues.and.use
73 in.the | 14 of.the.liver
43 it.is | 13 the.liver.and
30 and.the | 11 of.the.body
26 for.the | 10 liver.and.spleen
26 the.leaves | 9 the.juyce.of
24 at.the | 8 the.decoction.of
21 vertues.and | 7 about.the.edges
19 a.little | 7 an.herb.of
19 of.a | 7 and.a.little
19 the.body | 7 and.of.a
19 the.liver | 7 decoction.of.the
19 the.root | 7 is.good.to
19 to.the | 7 it.is.an
18 is.good | 7 obstructions.of.the
18 they.are | 7 pains.in.the
17 and.use | 7 the.root.is
17 of.them | 6 distilled.water.of
17 the.decoction | 6 end.of.the
17 the.juyce | 6 in.wine.and
Here is the same for the novel I Promessi Sposi (modern Italian), same conditions:
42 don.abbondio | 8 di.don.abbondio
29 che.non | 7 amor.del.cielo
19 non.si | 7 per.amor.del
17 con.un | 6 l'illustrissimo.ed.eccellentissimo
16 e.di | 5 ed.eccellentissimo.signore
15 e.poi | 5 qua.e.là
12 a.un | 4 bravi.e.vagabondi
12 e.che | 4 che.don.abbondio
11 e.il | 4 don.abbondio.non
11 in.un | 4 e.poi.e
11 per.la | 4 eccellentissimo.signore.il
10 con.la | 4 non.so.niente
10 il.mio | 4 poi.e.poi
10 quello.che | 4 posso.aver.fallato
9 a.me | 4 signore.il.signor
9 non.ci | 3 a.don.abbondio
8 ciò.che | 3 a.ogni.passo
8 con.voce | 3 a.segno.che
8 del.cielo | 3 a.un.galantuomo
8 di.don | 3 a.un.tratto
And here is the same for the Vulgate Genesis (Latin), same conditions:
35 super.terram | 21 annis.et.genuit
32 et.genuit | 17 filios.et.filias
27 in.terra | 14 et.genuit.filios
21 ad.eum | 14 genuit.filios.et
21 annis.et | 10 annos.et.genuit
19 dominus.deus | 10 et.mortuus.est
18 est.in | 10 inter.me.et
18 et.filias | 9 et.facti.sunt
18 filios.et | 9 et.factum.est
16 cum.eo | 9 sunt.omnes.dies
16 me.et | 9 super.terram.et
16 postquam.genuit | 7 facti.sunt.omnes
15 et.ait | 6 dixitque.ad.eum
14 genuit.filios | 6 est.ad.eum
13 et.dixit | 6 est.vespere.et
13 terram.et | 6 et.mane.dies
13 usque.ad | 6 et.vidit.deus
12 caeli.et | 6 locutus.est.ad
12 est.ad | 6 me.et.te
12 nomen.eius | 6 triginta.annis.et
I conclude that
(1) English has a lot more repeated two-word phrases than Latin because of its grammar: use of prepositions rather than cases, mandatory pronouns, articles (only one of each)...
(2) Italian has fewer repeated two-word phrases than English, probably because it has inflected articles and relies on verb inflections rather than mandatory pronouns; but still has more than Latin.
(3) The amount of repeated three-word phrases, and of repeated two-word phrases excluding function words, is totally dependent on the text, not on the language.
How does Voynichese compare with those languages?
All the best, --jorge
RE: Can LAAFU effects be modeled? - Jorge_Stolfi - 04-09-2025
(04-09-2025, 10:42 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.for first lines, 85% of cases start with one of 'p t k f' (in descending order of frequency).
IIRC, the first word of each paragraph generally does not occur elsewhere in the book, not even on the same page. If that is because if p t k f are "capitalized" versions of other letters, could it be that variants of that word with the first gallows replaced by some non-gallows letter do appear elsewhere?
All the best, --jorge
RE: Can LAAFU effects be modeled? - nablator - 04-09-2025
(04-09-2025, 02:10 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.IIRC, the first word of each paragraph generally does not occur elsewhere in the book, not even on the same page.
It does happen:
You are not allowed to view links. Register or Login to view.
You are not allowed to view links. Register or Login to view.
You are not allowed to view links. Register or Login to view.