The Voynich Ninja

Dictionaries contain words in alphabetical sequence, so if the VMs is some kind of glossary or dictionary, it must contain absolutely similar sequential words at the beginning, at least, of a paragraph, but it doesn't. As for grammars, it seems to be possible.

This is a popular late medieval chant "Virgo Virginum":
Veni virgo virginum veni lumen luminum veni vena veniae  
Veni salus hominum veni splendor ordinum caelestis militiae  
Consolatrix inclita veni vide visita certantes in acie  
Nos rege nos excita nos fove nos suscita de lacu miseriae  
Veni Jesse virgula veni rosa primula rosa carens carie 
 Peccatorum vincula rumpe prece sedula praesentis familiae  
Magne major maxima reple cordis intima caelesti temperie  
Consolatrix optima esto nobis proxima rogans regem gloriae  
Ut nos jungat superis dans nobis in dexteris post spem frui specie  
Quae regina diceris miserere miseris virgo mater gratiae.

The firtst line is especially interesting with its repetitions:
Veni virgo virginum veni lumen luminum veni vena veniae
The part [ven] has been repeated here 5 times in the three different words (veni - come; vena - stream, vein; veniae - of grace), veni itself is repeated 3 times. As well, there are two word pairs with the same root: virgo virginum, lumen luminum. Also we can find such kinds of word pairs: carens carie, miserere miseris. The word "nos" is repeated 4 times in the fifth line. Also, I marked part "ma", as it is repeated quite often, for example, in the line "Magne major maxima reple cordis intima caelesti temperie" - 5 times. Dividing some words we can get such a string:
magne ma jor maxi ma reple cor dis inti ma. 

What do you think about such chants, could they be a content of the VMs?
Playing with division of the words in the first line "Veni virgo virginum veni lumen luminum veni vena veniae" and substituting letters with the Voynich glyphs, I've got such a string:
ven      ivir       govir      ginvm   venil     vmen   lv   minvm  ven  i  vena     ven     iae
ched ochor tachor toiin chedol chked lch koiin ched o chedy ched og

(18-11-2022, 01:03 PM)Searcher Wrote: You are not allowed to view links. Register or Login to view.Dictionaries contain words in alphabetical sequence, so if the VMs is some kind of glossary or dictionary, it must contain absolutely similar sequential words at the beginning, at least, of a paragraph, but it doesn't. As for grammars, it seems to be possible.

This is a popular late medieval chant "Virgo Virginum":
Veni virgo virginum veni lumen luminum veni vena veniae  
Veni salus hominum veni splendor ordinum caelestis militiae  
Consolatrix inclita veni vide visita certantes in acie  
Nos rege nos excita nos fove nos suscita de lacu miseriae  
Veni Jesse virgula veni rosa primula rosa carens carie 
 Peccatorum vincula rumpe prece sedula praesentis familiae  
Magne major maxima reple cordis intima caelesti temperie  
Consolatrix optima esto nobis proxima rogans regem gloriae  
Ut nos jungat superis dans nobis in dexteris post spem frui specie  
Quae regina diceris miserere miseris virgo mater gratiae.

The firtst line is especially interesting with its repetitions:
Veni virgo virginum veni lumen luminum veni vena veniae
The part [ven] has been repeated here 5 times in the three different words (veni - come; vena - stream, vein; veniae - of grace), veni itself is repeated 3 times. As well, there are two word pairs with the same root: virgo virginum, lumen luminum. Also we can find such kinds of word pairs: carens carie, miserere miseris. The word "nos" is repeated 4 times in the fifth line. Also, I marked part "ma", as it is repeated quite often, for example, in the line "Magne major maxima reple cordis intima caelesti temperie" - 5 times. Dividing some words we can get such a string:
magne ma jor maxi ma reple cor dis inti ma. 

What do you think about such chants, could they be a content of the VMs?
Playing with division of the words in the first line "Veni virgo virginum veni lumen luminum veni vena veniae" and substituting letters with the Voynich glyphs, I've got such a string:
ven      ivir       govir      ginvm   venil     vmen   lv   minvm  ven  i  vena     ven     iae
ched ochor tachor toiin chedol chked lch koiin ched o chedy ched og

That is very useful and suggestive Searcher. Many thanks. The Voynich needs comparison with late medieval inventory literatures. It is beyond my skills and resources to make such a study so I appreciate contributions in that direction. 

On appearances, the text is very chant-like isn't it? Try it in EVA. It works as a chant - resonant, rhyming, strongly alliterative, elongated vowels. I asked the question: if not prose, what? The obvious reply would be: poetry. The possible poetic nature of the text tends to be overlooked. Amongst other things, the purpose of poetry (as opposed to prose) is mnemonic. 

Nor need that be incompatible with an inventory: a chanted inventory, made to be easy to remember entries, is possible especially in a pre-printing press context.

What type of inventory might it be? The safest and most likely answer is: one relevant to the illustrations. 

In this respect, I note that the nymphs in the ms. are depicted taking measurements.

Further adventures with Voynich Vord Paradigms, once again applied to vords in the context of LINES.

There are many possible decompositions of a LINE of Voynich text, each privileging one or other feature. My methods are, perforce, non-statistical: paradigms and patterns. I am trying to find useful  methods for analysing lines and better understanding Voynichese in context.

This current method I call ASSEMBLY ANALYSIS. It supposes that longer vords are composed of shorter vords, that vords are assembled from basic elements within the line as a container.

To illustrate this I divide the vords of a line into four groups:

1. Those with gallows.
2. Benched, but without gallows. ([ch] and [sh] vords) + [q]
3. The [daiin] group
4. Particles and sundries. (neither benches nor gallows).

Vords with gallows are marked with bold.
Benched vords without gallows - [ch] and [sh] vords – are marked in italics.

The [daiin] group are marked green.

The particles and sundries are left unmarked.

The categories go in that order. If a vord might be in two groups, it is marked in the higher.

There is an implied hierarchy or assembly order, with the order culminating in vords with gallows.

The order reveals the phenomenon of promotion or ELEVATION. A vord is promoted or ELEVATED if it has a benched glyph (or q-) added to it, and it is further promoted or elevated if it then has a gallows glyph added to it.

Thus: aiin might be elevated or promoted to become chaiin, which may be promoted to become kaiin.

There is an implied order, simplicity to complexity, shorter to longer (as well as shorter to taller according to a vertical decomposition of the script.)


As a focus of analysis I have also marked the vord in each line that is nearest to either of two default vords: CHOLDAIIN or QOKEEDY. A vord near to QOKEEDY is marked in red and a vord near to CHOLDAIIN is marked in blue.

The proposal is that vords in a line, in their assembly from smaller units, mutate towards one of the paradigm vords. This is a method for showing this process. 

The paradigm vords were identified by Patrick Feaster as the default vords of Currier A and B respectively, so the comparison made in this method bares upon the operations of the two ‘languages” dicernible in the Voynich text as they manifest at the level of LINES.


We disassemble lines into these four groups, their assembly order:

<f101r1.P.6;H>   

okeeol.sho.shody.sho.shol.okeeeol.cheos.sheokeey.sheeor.chchy.chodaiin.sheeckhey.teeol.scheol.sar.oeeor=

sar.
oeeor

sho.
sho
shol
shody
cheos
chchy
scheol
sheeor
chodaiin

teeol
okeeol.
okeeeol.
sheokeey.
sheeckhey.

Here it seems the benched group moves towards CHOLDAIIN but the gallowed group move towards QOKEEDY.

We can see motifs being developed. The motif of a string of vowels is already present in the ‘Particles’ group: oeeor. This is elevated to [sheeor] and then to the three [-eeol] forms in the gallows, with [r] and [l] glyphs interchangable or mutating.

As we see in a Vord Paradigm we quickly observe that glyphs mutate. Some mutations are more common than others, but the general rule is consonants-for-consonants and vowels-for-vowels.


Another instance:

<f34v.P.7;H>     

olchdaiin.chedy.chey.keedy.chy.kedy.dy.qokedy.okey.sair.chkain.otain-

dy

sair

chy
chey
chedy
olchdaiin

kedy
okey
otain
keedy
chkain
qokedy

Again, the benched vords culminate in a vord near to the CHOLDAIIN paradigm and the gallows in a vord near to QOKEEDY.

We can see development within groups - [chy] becomes [chey] becomes [chedy] – but also elevation or promotion – [chedy] becomes [kedy].

Note well that in [olchdaiin] the deviation from the paradigm is the reversal of elements in [chol] – it becomes [ol + ch], i.e. backwards. We find this again in other lines.


<f100r.P2.7;H>   

saiichor.sheor.qockhody.odeor.yksheey.chol.sheody.sai.cheol.raiin-

sai.
odeor.

raiin

chol.
cheol.
sheor.
sheody.
saiichor.

yksheey
qockhody

In many cases we see how longer vords are assembled from smaller units – hence the ‘assembly order’.

Note how the particle [sai] is a constituent part of [saiichor.] 

In fact, [saiichor]  might be seen as being assembled entirely from smaller elements. The double [ii], for instance, might be taken from [raiin].

Note, though – importantly - [saiichor] resembles the default vord backwards: choldaiin = chor - saii. = saiichor

Again, we see the gallowed vords being developed towards QOKEEDY and the benched-no-gallows vords developing towards CHOLDAIIN.

Again we find a case of the reversal or inversion of component parts.


A case where CHOLDAIIN dominates:

<f32v.P.10;H>     

sho.keol.chor.chol.daiin.cphol.cthol.da.ar

da
ar

daiin

sho
cho
chol

keol
cthol
cphol

This line is entirely concerned with CHOLDAIIN - the gallowed vords play with elevations of “chol”. 

Typically, different forms and combinations are rehearsed: [sho] and [cho].

Notice here how the benched gallows are a combination of the two paradigms: the gallows intrude into the [ch] of [chol]. Elevation can happen by the [ch] being replaced with [k] or the [k] (or other gallow) being added to [ch] to become [ckh].


And another:

<f111v.P.13;H>   

sheaiin.okaiin.chckhy.cheol.kedy.chetar.okaiin.shal.lchdal.tedy.tar.amd-

amd

shal
cheol
lchdal
sheaiin

tar
tedy
kedy
okaiin
okaiin
chetar
chckhy

This is another CHOLDAIIN dominant line but where that paradigm extends into the gallowed vords. [kedy] is the nearest to QOKEEDY.

The intriguing thing is that many vords and patterns are best explained as a meeting of the two paradigms. Thus: [okaiin] might be [ok] + [aiin] or [q-ok] + [d – aiin].

Note inversions and reversals. There can be no doubt that [lchdal] is based upon the CHOLDAIIN paradigm but note how [chol] becomes [lch].


<f99r.P3.10;H>   

sol.sheol.keshey.qokeeey.chs.chey.dolchey.ctheey.daiin.cheom-

sol.

daiin

chs.
chey.
sheol.
cheom.
dolchey.

ctheey.
keshey.
qokeeey.

Another case where the most complex of the benched-but-no-gallows vords shows inversion. CHOLDAIIN becomes [dolchey] with the [ch] and the [d] of the paradigm being swapped.


This is enough to demonstrate the method. It continues from other methods I've outlined already.

It is my observation that there are not two languages or even ‘dialects’, but one cloth woven of two threads. In some lines of text one thread dominates, and in others the other thread dominates. But they interweave and overlap. A simple binary – A and B – does not adequately describe what we see. Indeed, the interweaving of the two threads seems to drive the mutations of the vords; it is the dynamic of the text.

The two threads meet in an obvious way in the benched gallows.

(19-11-2022, 10:10 AM)Hermes777 Wrote: You are not allowed to view links. Register or Login to view.The possible poetic nature of the text tends to be overlooked.

To my knowledge, most researchers regard poetry as a possibility. For instance:

• In their statistical analyses, 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. included both poetry and prose.
  
• In his 2014 paper, You are not allowed to view links. Register or Login to view. mentioned that "although such repetition is an unusual feature in natural languages, it is not unknown in particular genres (e.g. poetry and incantation)".
  
• You are not allowed to view links. Register or Login to view. compared poetry and features of Voynichese.
  

I guess that the reason why the subject is not explicitly mentioned more often is that, both linguistically and statistically, poetry is very similar to prose: they both are language. Measures like character conditional entropy and MATTR do not vary considerably between poetry and prose: this tells us that Voynichese cannot be a simple substitution of prose or poetry written in any ordinary European language (e.g. Latin, Italian, German, English). It would be great if there was a longer text of any kind that exhibits the same rates of perfect reduplication and partial reduplication that we see in the VMS, but as far as I know nothing really close has been found yet. For instance, the Finnish poem Kalevala (collected by Brian Cham and pointed out by Jonas Alin) features both phenomena, but the rate is very far from Voynichese (check FIN vs EVA in You are not allowed to view links. Register or Login to view.).

Here I put together a few plots comparing verse (V) and prose (P). I included the "Veni Virgo" poem pointed out by Yulia and the Donatus text pointed out by stenog. It should be noted that "Veni Virgo" is too short for reliable statistics. Also, I could have made errors somewhere along the process. The texts are:

P-DE-weltchr Nuremberg Cronicle
V-DE-tristan Tristan
P-ENG-g.herb Grete Herball
P-ENG-kjames King James Revelations
V-ENG-shake Shakespeare's Sonnets
P-LAT-don.all The whole Donatus text
P-LAT-don.pron Donatus "Pronoun" paragraph (including the quote You are not allowed to view links. Register or Login to view.)
P-LAT-pliny Historiae
V-LAT-veni Veni Virgo
V-LAT-virgil Aeneid
P-ITA-mach Machiavelli: Il Principe
V-ITA-dante Divina Commedia
EVA transliteration of the individual sections

Character entropy and character conditional entropy
[attachment=6993]
Character conditional entropy is possibly the best known statistics. The VMS has consistently low values, also using different transliterations. Yulia and stenog's texts appear to "move" in the right directions: this is partially due to the fact that they are short, but the repetitiveness contributes as well. The whole Donatus text is much  closer to ordinary Latin. In general, there is no apparent systematic difference between Verse and Prose.


Mattr5 and Mattr20
[attachment=6994]
Similar plots were You are not allowed to view links. Register or Login to view.. This is one of those plots that makes Voynichese look almost normal: the difference is that it falls above the regression line because MATTR5 has lower values than normal due to the frequent identical repetition of words. Donatus, VeniVirgo and King James' Revelations are outliers with particularly low MATTR5 values. In the case of King James, this is not due to the consecutive repetition of words, but to the same word occurring at a distance of less than 5 words, eg:

and riches and wisdom and strength and honour
of god and of the testimony of jesus
had a face as a man

Again, the fact that the text is verse or prose does not appear to have a clear effect.


Perfect reduplication and partial reduplication (1 character difference)
[attachment=6992]
The bulk of verse and prose samples fall near the origin with low values for both measures. VeniVirgo and (more clearly) Donatus come closer to Voynichese. I suspect that the spacial distribution of reduplication is different in Voynichese and Donatus. In the VMS the phenomenon is more or less uniformly distributed, in Donatus I guess it is concentrated in paragraphs where declensions are listed.


In conclusion, I find that considering both poetry and prose is the safest course of action. The assumption that the text is poetry does not explain much by itself. We don't know if the text is meaningful and, if it is, how much of the weirdness is due to the underlying text and how much is due to the writing system, so it's impossible to say anything conclusive about the nature of the underlying text.

The following statistics would help us understand what the Gallows are (ordinary letters, some kind of key or something else).
We take all the words from a two-story bench with a gallows.
We remove the gallows from the second floor. And we calculate as a percentage how many received "normal" (existing) words.
Do this for languages A and B, and separately for each gallows.
Do the same without counting unique words.
Can anyone make these calculations?

(21-11-2022, 11:37 AM)Wladimir D Wrote: You are not allowed to view links. Register or Login to view.The following statistics would help us understand what the Gallows are (ordinary letters, some kind of key or something else).

We take all the words from a two-story bench with a gallows.

We remove the gallows from the second floor. And we calculate as a percentage how many received "normal" (existing) words.

The percentage, something like 32% in TT, doesn't mean much because of the high percentage of hapax legomenon, 71%, among words with benched gallows. Also many of the un-gallowed words can be re-gallowed somewhere else to get an existing word.

This is an attempt to integrate two (or four!) different Vord Paradigms into a single model, a foolhardy attempt to forge a global template. It requires a dual paradigm. The two paradigms are:

CHOLDAIIN
QOKEEDY

Both, however, have variants:

DARCHOL
QOKEDAR

There are thus no less than four paradigmic vords. I have attempted to reconcile all four at common points in order to reveal a more comprehensive system.

The system has two main features that permit this:

1. It treats the vord paradigms as CYCLES.

2. It treats vord BREAKS as the critical points in the cycles.


The view of the text implicit in this is that there is a repeating base cycle (the CHOLDAIN or CH cycle) that is like the warp of a weave. A second cycle – a weft – ( the QOKEEDY or Q cycle) then intrudes and is intermixed with the base cycle.

(The variants DARCHOL and QOKEDAR actually arise out of the intermixing.)

This CYCLIC and INTERWOVEN nature of the text is the reason any single paradigm will be inadequate. The text is woven of two main threads.The bulk of vords are hybrids of the two.

The main distincton between the two threads is that the QOKEEDY (Q) cycle carries or contains gallows glyphs. The CHOLDAIIN cycle contains benched glyphs but no gallows. The gallows glyphs are the whole issue.

The intermingling of the two cycles comes about when the gallows glyphs are inserted into the repeating base cycle.

Concerning this, namely the way the two threads or streams of text converge, the invariable rule is:

Gallows glyphs are inserted into the SPACES of the CHOLDAIIN cycle.

Those are the points where the two threads are linked together. That is the rivet. That is where one cycle is anchored into the other. Other points of contact follow from that anchor.

Note that the base cycle has both soft and hard breaks. There is a soft break between CHOL and DAIIN. There is a hard break before CHOL and after DAIIN. The base sequence goes:

** CHOL*DAIIN ** CHOL*DAIIN ** CHOL etc.

Looking closer at this underlying churn, CHOL and DAIIN are versions of the same vord, and DAR is a variant of DAIIN.

Secondarily, or simultaneously, the same cycle goes:

** DAR*CHOL ** DAR*CHOL ** DAR*CHOL ** etc.

In the end, therefore, everything comes down to a threefold pattern of glyphs:



This is at the very foundation – the atomic level – of the non-gallows text. We cannot reduce it any further.


All the permutations of glyphs familiar to us from earlier vord paradigms are not shown in this current presentation.

[ch] can become [sh], for instance, and [n] can become a range of final glyphs.

In this core formation [ch] becomes [d], [o] becomes [a] and [l] becomes [r], and vice versa. These are the most basic permutations in the underlying CH-cycle.

Ignoring such permutations, vords can be returned to their basic forms. Thus [shol] is considered a permutation and can be returned to its basic form: [chol].


It is evident that the purpose of the QOKEEDY cycle is to bind together the cycling components of the CHOLDAIIN cycle.

The purpose of the gallows glyphs is to join, bind, bridge. They are placed in the breaks (spaces, pauses) of the CHOLDAIIN cycle. (To continue a weaving analogy, they are knots - a weaving being an assembly of knots.)

The CHOLDAIIN cycle is short phase. The QOKEEDY cycle is long.

Note that this iteration of the Vord Paradigm specifically permits vords with two gallows: these are cases of vords extending across two phases of the cycle.

What is called Currier A is text in which the CH-cycle dominates whereas in Currier B the Q-cycle overwhelms the underlying CH-cycle.


I must say that developing vord paradigms (as study tools) in this direction inevitably starts to suggest analogies with music.

What behaves like this: cycles of looping and interweaving refrains that seem mathematical in some way but are subject to endless variation? It very much resembles a (polyphonic) musical texture in this respect. 

On entropy and musical notation I note this study:

You are not allowed to view links. Register or Login to view.

(26-11-2022, 10:29 AM)Hermes777 Wrote: You are not allowed to view links. Register or Login to view.This is an attempt to integrate two (or four!) different Vord Paradigms into a single model, a foolhardy attempt to forge a global template. It requires a dual paradigm. The two paradigms are:

CHOLDAIIN
QOKEEDY

Thanks for this latest installment in your ongoing vord paradigm investigations.

One question I'd meant to pose earlier was whether you found your [qokeedy] paradigm to be as good a fit for Currier A as for Currier B.  Before, when I proposed the idea of [qokeedy] and [choldaiin] loops in a different context (You are not allowed to view links. Register or Login to view.), the [qokeedy] loop was based on Currier B statistics, while the [choldaiin] loop was based on Currier A statistics.  The [qokeedy] loop can also be inferred from statistics for the text as a whole, but that's mainly (I assume) because there's so much more Currier B text than Currier A text.

Those loops are based on connecting the most probable individual glyph-to-glyph transitions (making some working assumptions about what counts as a glyph, of course), and in that context, the reason why the [choldaiin] loop doesn't contain a gallows is that there's no glyph in Currier A whose "first choice" of following glyph is a gallows.  In Currier B, by contrast, there is -- most notably, [o] is followed by [k] 30.31% of the time, well ahead of the next option of [l] at 23.25%.

Since the [choldaiin] loop itself is based on Currier A statistics, it might be interesting to look at where gallows most often appear specifically in Currier A.  I'm still absorbing the intriguing observations Tavie made at the Voynich conference yesterday, which involved some notable differences in gallows distribution among scribes -- and hence also Currier languages -- that could likely shed some light on this.

(26-11-2022, 10:29 AM)Hermes777 Wrote: You are not allowed to view links. Register or Login to view.Gallows glyphs are inserted into the SPACES of the CHOLDAIIN cycle.

Here are the most common "nested" environments for unbenched [k] and [t] in Currier A -- i.e., trigrams where the middle glyph is an unbenched gallows, ignoring word breaks:

[oka] 244, [okch] 232, [oko] 210, [oke] 175, [ykch] 117
[otch] 275, [oto] 156, [ota] 153, [oty] 114, [ytch] 107

I suppose [oka] and [ota] could be analyzed as fitting a gallows into the "soft break" in [chol,daiin]; the other cases might get a bit more complicated.

On the other hand, benched gallows seem to fit neatly *into* the [ch] of [chol,daiin], as e.g. with the sequences [cKhol] and [cThol] -- that is, if we interpret these as gallows inserted into a bench, and if we're thinking of gallows being "inserted" anywhere, that seems like the most conspicuous situation in which that could be happening.  The benched gallows would seem to occupy the "space" only if [cKh] = [kch], as has sometimes been speculated.

The two vord paradigms I’ve presented in this thread as a dual paradigm model of the Voynich text – developed from Patrick Feaster’s study of “loops” – whereby the text is an intermingling of two streams or cycles, can be reduced down to two glyph sets.

We might call a vord paradigm a telos. It is an end towards which the text moves. But the Voynich text is formed in the tension between two teloi, CHOLDAIIN and QOKEEDY. There is a spectrum between two poles. Words are formed in a tension between two stillpoints, two points of resolution.

Here I consider the teloi just as glyph sets.

The paradigm CHOLDAIIN obviously covers [ch], [o], [l], [d], [a], [i] and [n].

The second paradigm QOKEEDY shares [o] and [d] but provides [q], [e] and [y] and the gallows [k].

What then of the other glyphs/letters that constitute the text?

Some of them are demonstrably variations on the glyphs of CHOLDAIIN. The plainest example is [sh] which is a modification of [ch].

Modifications of CHOLDAIIN yield [sh], [s] and [r], along with the rare finals [m] and [g].

There are no such modifications in QOKEEDY except, of course, that there are four variations of the gallows. So as well as [q], [e] and [y], QOKEEDY provides: [k], [t], [f] and [p].

This leaves only the benched gallows which by this analysis are a special case because they are a combination of the two paradigms.

We note that CHOLDAIIN contains a bench but no gallows. Conversely, QOKEEDY contains a gallows but no bench.

In the benched gallows the two streams overlap and compound. The gallows intrudes upon, or into, the bench.

Thus, by these two vord models, or templates, we can account for the full glyph set as observed in the running text, barring rare glyphs.

The two glyph sets in this breakdown are:

Ch O L D A I N Sh S R M G
Q K T P F E Y

The benched gallows are separate being, literally, an overlapping of the two sets using glyphs already covered.


In the following depiction of text (lifted from Takahashi - p. 19v and 112v) I have highlighted the glyphs belonging to QOKEEDY in red. The glyphs in the CHOLDAIIN set, CHOLDAIIN and its variant glyphs, remain plain.