The Voynich Ninja

(01-11-2022, 12:19 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.For me, this is not a problem. If the MS text was the result of a truly arbitrary process, then we would expect to be able to predict such frequencies, but in case it is meaningful language, then not.

Compare English:

bat - bet - bit - bot - but
cat - cet - cit - cot - cut

Tehre is no pattern in the frequencies
(Dislcaimer: just the first example I could think of).

Words in the Voynich text doesn't behave like in the "bat - bet - bit - bot - but" example. There is not only a small 2 dimensional grid but a whole network:
qokeedy (305) qokeedy (305) qokeedy (305)    qoteedy (74) qoteedy (74) qoteedy ( 74)
qokedy  (272)  qokeey (308)  okeedy (105)    qotedy  (91)  qoteey (42)  oteedy (100)
qokdy   (  4)   qokey (107)   keedy ( 53)    qotey   (24)   qotey (24)   teedy ( 13)
qo      ( 29)      qy (  3)      ky ( 25)    qo      (29)      qy ( 3)      ty ( 16)

To achieve something similar longer words and multiple changes would be required. Something like:
   acaut  aceut  aciut  acout   acuut
    acat   acet   acit   acot    acut
     cat    cet    cit    cot     cut
     bat    bet    bit    bot     but 
    abat   abet   abit   abot    abut
   abaut  abeut  abiut  about   abuut



There are patterns in the frequencies, e.g. <ee> is more common than <e> and <eee> and <ok> tends to be more common than <ot>. The general rule is that "high frequency tokens also tend to have high numbers of similar words" [see Timm & Schinner 2019, p. 6].

qokeeedy (  5)  qoteeedy (  3)
 qokeedy (305)   qoteedy ( 74)
  qokedy (272)    qotedy ( 91)

 qokeeey ( 26)   qoteeey (  4)
  qokeey (308)    qoteey ( 42)
   qokey (107)     qotey ( 24)

 okeeedy (  9)   oteeedy (  3)
  okeedy (105)    oteedy (100)
   okedy (118)     otedy (155)

(02-11-2022, 04:53 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.1. words with multiple gallows are present.

The question is can I judge them as one word or are there two?

I have a correction in one gallows. Are they two letters in combination. Why would a split work here. Does a meaningless text work here?

3. obviously two of the same gallows in a row. Are they really the same or is it just a deception.

When I use EVA all the words look the same. But they are not. There are more differences than first apparent.

Do you know the difference between an f and a t. v or u.

It's not what it seems.

Words with two gallows glyphs do exist, but why do so many occur at the start of a line? And many others in diagrams rather than running text? 

And why are they all so rare? According to Stolfi, no word with two gallows occurs more than three times.

While these words are interesting, a model which excludes them may be more enlightening than one which includes them.

(02-11-2022, 03:04 AM)Emma May Smith Wrote: You are not allowed to view links. Register or Login to view.A good model explains common observations: why only one gallows glyph per word? why is /d/ before a gallows so rare, but a /d/ after a gallows so common? why is /ch/ so common, but not after /o, y/, except at the beginning of a line? And when we say "explain", we mean not that it provides a cause, but rather it provides a rule. So we might say, in answer to the foregoing questions: "there is only one normal slot for a gallows in a word", and "the slot for a gallows precedes the slot for /d/", and even "/o, y/ before /ch/ is abnormal and not part of regular word formation". These answer still require more explanation, but they take us forward: glyph have slots! slots have order! the final word may be the result of multiple processes!

Not every pattern must be based on rules. Also a repeating process can cause resonance effects which appear as patterns. For a resonance effect even small changes in the start configuration can result in different outcomes. In the case of the VMS properties like the long-range correlations, the binomial-like word length distribution, and the word frequencies indicate that a resonance effect is in place [see Timm & Schinner 2019].

There is no rule like "there is only one slot for a gallow" as the existence of words like <keoky> or <okeokeokeody> demonstrates. There is only the general principle that "high frequency tokens also tend to have high numbers of similar words" [see Timm & Schinner 2019, p. 6].

The rule behind "/o,y/ before /ch/ are rarely used" is a positional one. /o/ added to a string starting with /ch/ or /d/ normally results in <ok> [see [ch + ol] in Timm 2014, p 70]. This means Instead of <ochol> or <odol> normally <okol> is used. Therefore the word <ochol> is rare even though <chol> and <dol> are quite common. If you prefer you could also interpret /d-/, /ch-/, /ok-/, and /qok-/ as interchangable prefixes [see Table 3 in Timm & Schinner 2019, p. 10]:

Prefix |    aiin     ol       dy
none   |    aiin     ol       dy
d-     |   daain    dol       dy
ch-    |  chaiin   chol    chedy
ok-    |  okaiin   okol    okedy
qok-   | qokaiin  qokol   qokedy

(02-11-2022, 04:53 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.1. words with multiple gallows are present.
The question is can I judge them as one word or are there two?

Two. They can always be split before the second gallows to get two normal-looking words: it is not a coincidence. Gallows should not be included in word models anyway. Their placement is basically free, with strong preferences and a few exceptions.

Quote:When I use EVA all the words look the same. But they are not. There are more differences than first apparent.

The small differences in the loops and curvature could be important, but there is a continuum in the way loops are drawn (more or less rounded) so they are not likely to express something meaningful. The same observation applies to the size of the loops and just about any part of the glyphs. The writing skill is not so great that such minutiae should be taken into account in my opinion.

The vord paradigm I presented in this thread (based on that of Thomas Coons) can be further simplified. There is the temptation to make it more complicated to accommodate exceptions, but the objective is to draw out the persistent, pervasive, simplest structure.

To reduce the paradigm to the simplest formula:

Every vord must conform to the default vord QOKEEDY as nearly as possible, consonants with consonants and vowels with vowels, tripartite, with a hard boundary at the gallows.

This, in fact, reduces Voynichese down to a single rule: every vord is based upon the template of QOKEEDY and approximates it more or less.


Every vord consists of a STEM and an ending, and a great many have a prefix added, the prefix being provoked by a gallows in the stem.

Every vord has a subject. It is the element of the stem around which other elements revolve. In the default vord QOKEEDY the subject is [k]. The [k] is the core of the stem that is modified by the other elements. Specifically, it is modified by [ee] and by [qo]. 

Every extra part of a vord should be understood as an element that modifies (acts upon) the subject. Elements might be:

*additional consonants in the stem that create compounds with the subject (eg. [pchd] = the subject is [p] with [ch] and [d] as compounds. A compound consonant.

*different arrangements of vowels in the stem in the position [ee]. (eg. eeee, ai, aiii)

*prefixes before the subject (or a compound containing the subject.)

The subject of a vord is usually a gallows, or bench, but in any case almost always a consonant corresponding to the [k] in QOKEEDY (and only rarely a vowel corresponding to [ee] with the consonant empty.)

The default ending is [y] but often endings are formed by hard consonants [m], [g] etc corresponding to the [d] in QOKEEDY.

Finals provoke a word break.

Ultimately, any consonant can go in any consonant slot and any vowel in any vowel slot, this being determined by a scale of probabilities rather than hard rules. 


Applying this streamlined paradigm to LINES rather than isolated vords. Here is a breakdown of Line 10, Page 39r (selected at random):


We can consolidate this for clarity into the tripartite system of three compartments, A, B and C and see the STEMS, Endings and Prefixes.


To parse it long-hand:

pchdar = pchda – r
shedy = she – dy
ar = a – r
aiir = aii – r
okair = o – kai – r
ykeols = y – keo – ls
shedy = she – dy
qockhdy = qo – ckh – dy
laiin = laii – n
syky = sy – k – y

The subject of each vord is always the foremost element of the stem:


The principle is: where there are any dilemmas in parsing, choose the formation that most nearly corresponds to QOKEEDY, consonants with consonants and vowels with vowels. Parse the vord to as near to QOKEEDY as you can.

I know the theory that words start at the gallows and go to the next gallows, regardless of the space. That is not new.
But this presupposes that there should be about 26 different types of gallows. But this is not the case. Therefore, the gallows characters would have no value.

But now, if I delete the gallows, the already monotonous becomes even stranger.
That tells me I'm working in the wrong direction. It has to become more.

That's the word QOKEEDY. What is QO?
Is there really no difference from 4o to qo ?
For the PC, EE is simply 2x E-pulse.
You can describe something with it, but you can't solve it.

How should I feed the PC programme ?

Are these characters really all the same? Do I only see C?
From the disc.
[attachment=6914]
Translated with You are not allowed to view links. Register or Login to view. (free version)

We always need to remember (as I am sure people do) that the Voynich text was not created by a machine, but by a human being, and that many hundreds of years ago. There will be no mathematical rule that can apply. Only fuzzy rules. (*)
Humans will make mistakes, are prone to whims, may forget things etc. etc.

Therefore, rules of the type:
- .. almost always ...
- .. almost never ...
is what we should expect.

Therefore, that there are exceptions to the almost hard rules that:
 - q is alwazys followed by o
 - there is only one gallows per word
should not be a concern.
No system will perfectly.

--
(*) Indeed, "fuzzy set theory" is a part of mathematics since the mid 60's

(03-11-2022, 01:14 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.We always need to remember (as I am sure people do) that the Voynich text was not created by a machine, but by a human being, and that many hundreds of years ago. There will be no mathematical rule that can apply. Only fuzzy rules. (*)

Humans will make mistakes, are prone to whims, may forget things etc. etc.

It is true but, tempting as it is to explain (way too many) irregularities as mistakes, qko and multiple gallows tell us more about the system than dysfunctional word models ever will.

(03-11-2022, 01:14 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.We always need to remember (as I am sure people do) that the Voynich text was not created by a machine, but by a human being, and that many hundreds of years ago. There will be no mathematical rule that can apply. Only fuzzy rules. (*)
Humans will make mistakes, are prone to whims, may forget things etc. etc.

Therefore, that there are exceptions to the almost hard rules that:
 - q is alwazys followed by o
 - there is only one gallows per word
should not be a concern.
No system will perfectly.

That something rarely occurs doesn't mean that it must be an error. For a text representing a natural language it is very likely that rare word variants are errors. However, in the case of the VMS the whole text consists of word variations. For instance beside <daiin> also <dain>, <daiiin>, and <dan> exists:

daiiin ( 17)
daiin  (863)
dain   (217)
dan    ( 20)

Are <daiiin> and <dain> misspelled variants of <daiin> and what is with <dan>? In 97.4 % of the cases EVA-n occurs after EVA-i. EVA-q is followed in 97.5 % of the cases by EVA-o. This means EVA-n without EVA-i is similar unlikely as EVA-q without EVA-o. So if  the sequence /qe/ is in your eyes an error what is with sequences like /an/ or /aiiin/?

(03-11-2022, 10:01 AM)nablator Wrote: You are not allowed to view links. Register or Login to view.It is true but, tempting as it is to explain (way too many) irregularities as mistakes, qko and multiple gallows tell us more about the system than dysfunctional word models ever will.

I am not at all sure of anything. For the things we observe I can usually think of several explanations, and I am not ready to decide on one or the other. In fact, I find it hard to understand how (some) people are convinced about certain explanations for features in the text, since all we have are opinions, and there is no evidence.

For the case of Eva-qko, for me it is just as wrong to assume that this is an error, as to assume that it is part of a model or a ystem that we need to be able to explain. We just don't know.
In fact, the whole situation that there is a word paradigm, or several more or less successful ones, is unexpected, and this is not the case for 'normal'  encrypted documents of the time.

We don't know how the MS was created, and for me, the option that it is a fair copy by scribes who did not understand what they were copying, remains a possibility. In such a case, errors are likely to be abundant.

Even in normal manuscripts that are perfectly legible, and that were copied by professional scribes, there are some typical mistakes like copying a piece of text twice in a row, or jumping to the wrong source line while copying. The scribe should have noticed this as he could understand what he was copying, but in the end he is human and can make mistakes.
I could easily imagine that this is how the few words with two gallows (and other irregular words) came about, but I am not claiming this as being the right answer, and I can think of other explanations. I just don't know.

The bottom line is that we should not expect any rule to be hard and fast.

Now I have some ideas about the cases where Eva-q is not followed by Eva-o, but these are again guesses, without there being any way of knowing whether they are right or wrong.