+- 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: Sequential word repetitions in the VMS (/thread-61.html)
RE: Sequential word repetitions in the VMS - Emma May Smith - 07-02-2016
Hi Torsten, thanks for the reply, it clears up a few issues. But my main objection is still outstanding: there is a set of rules which governs the structure of most Voynich words which is not adequately explained.
We either agree there is an underlying structure—which you seem to hint at (2014, p5)—or there is none. Both leave things open to question. When you say that “we only see the changes he has made” or “he only repeated the same ideas most of the time”, you seem to be hedging yourself. You admit that a structure is apparent but that it is actually arbitrary. Can you understand why this is unsatisfying? Given his supposed freedom to alter words, why did he not take it?
RE: Sequential word repetitions in the VMS - Torsten - 07-02-2016
Hi Emma,
because he was free to do anything he was liking. This also implies that he was free not to do certain things.
If his intention was to write something similar to language he was limited to his concept of language. In the same way he was limited to his concept of aesthetics, to the limitations of the writing material and to the limitations of the copying process.
For instance because of his concept of language he preferred words of a certain length. Because of his concept of aesthetics he added line initial gallow glyphs and preferred certain glyph combination and avoided others. Because of the limitations of the writing material he had to shorten the words at the end of the lines. Because of the limitations of the copying process all the words in the VMS are more or less similar to the most frequent word 'daiin' (see footnote 11 on page 7 in You are not allowed to view links. Register or Login to view.). The length of 'daiin' by using the EVA transcription is 5. That the words are connected to each other also determines the distribution of the word length for token and types (see p. 21 - 23 in You are not allowed to view links. Register or Login to view.). Etc.
That he is not limited to a given set of rules also implies that he was able to learn during the writing process. This results in a shift from the language used at the beginning of the writing (Currier A) to the language used at the end (Currier B). For instance in Currier A he was using the bigram 'od' whereas in Currier B he preferred the bigram 'ed' for otherwise the same words (see p. 24 in You are not allowed to view links. Register or Login to view.). The shift from language A to language B therefore also implies that the VMS is the work of one scribe.
RE: Sequential word repetitions in the VMS - Wladimir D - 07-02-2016
Timm I do not know English, so I not familiar with your works. But I also consider the artificial generation of words. I hope that at least a part of the manuscript makes sense. Has sufficed me just of two symbols, to encompass most of the words in the algorithm on the principle of the addition (changes) one character. Each symbol has its own meaning and their combination creates a new within the meaning of the word.
For example gallows *H*, H* small, big. The word o(пH *)с89 = oskhedy, I translate - the big fresh (water-containing "o") pedicels. If the gallows will jump from the second floor, I did translate the other part of speech (akin to the word) adverb. оH*пс89 = okchedy many fresh pedicels. The "exclusion" о*H*H*пс89 = otkchedy I would translate "any amount" or "regardless of the number."
If the gallows are the whole word and when changing of the floor she translated of another part of speech, is such a language construct explains many features that Emma listed in the topic "The Impossibility of Double Gallows".
Let us return to the repetition of words. In the artificial generation of concepts is easier to use a single number. If you want to receive the plural, the word is double repetition. Triplicate means - a lot. Quadruple repeat - very much.
RE: Sequential word repetitions in the VMS - don of tallahassee - 08-02-2016
I believe there is a rigid structure to the VMS words.
I think I show it for the most common 505 VMS words at:
You are not allowed to view links. Register or Login to view.
This is a distilled version of my proposed solution that works for these words.
Thank you.
Don of Tallahassee
RE: Sequential word repetitions in the VMS - Diane - 08-02-2016
On the matter of repetition -
I'm afraid that I can't see it on the internet now, but I once fell over an interview, or discussion about this with Jorge Stolfi.
Faced with the classic "dain, daiin, quokedy dain" and a couple of similar examples, he reproduced the pattern instantly, off the top of his head. I think the language was Malayalam, or Indonesian - one from that region, anyway.
Sorry to be so vague - I'll keep hunting for it and hope the deus-ex-machina will be kinder.
D
RE: Sequential word repetitions in the VMS - Torsten - 08-02-2016
@Don
Typical for the VMS are strong rules about character combinations and weak rules for the word order. Both features are the result of the copying process. If the scribe was adding a new element to the text generation process one of two things could happen: A.) He could add the new element into his set of copying rules. In this case it was frequently used and we would see it as a rule. Or B.) he could forget about the new element. In this case the new element is rarely used and it would look like an exception to us.
An example for the first case is the 'x'-glyph. An example for the second case is the variant of the 'r'-glyph used in 'rotcho' and 'rodaiin' on page <f10r> (see You are not allowed to view links. Register or Login to view. as Timm 2014: p. 33). 'x' is accepted as new glyph whereas the strange 'r'-glyph is not. But the only difference between both glyphs is that 'x' occurs 35 times and the strange 'r'-glyph occurs only twice.
The same is true for the shift of the 'od' bigram in Currier A into 'ed' for Currier B. Since 'ed' is frequently used we accept it as bigram even it is rarely used in Currier A.
Another example is the switch between 'a' and 'y' as last glyph of a word (see Timm 2014: p. 5). Normally, 'a' would change into 'y' as last glyph for a word. But sometimes the 'a' was not changed into 'y' like in 'cheda' in <f111r.P.33> or the scribe has written 'ya' like in 'dyaiin' <f105r.P2.10>.
As last glyph 'a' occurs 63 times. Is the existence of 'cheda' therefore an error or is it a valid word?
The bigram 'ya' occurs 22 times. Is the existence of 'dyaiin' therefore an error or is it a valid word?
Whatever the scribe does, for us it always looks like a rule or like an error. But they are the same. There are only too many possibilities for changing a word. It is impossible to use all of them frequently.
@Diane: That something can happen within language doesn't mean that it is typical for language.
RE: Sequential word repetitions in the VMS - don of tallahassee - 08-02-2016
You didn't look at what I sent, did you?
I didn't write anything about bigrams or switching between glyphs or valid words. I think all the VMS words are valid. I try to decode/deconstruct them all the same way. How can you be sure some aren't valid? That's a strange idea to me that would need a lot of proof, I think.
My proposed solution works a lot easier than the one you seem to be chasing. At least the deconstruction process does. Others have trouble agreeing that the proposed solution has much meaning from there on (I agree, but the VMS is what it is, does what it does, and the words deconstruct like I say they do).
Thank you.
Don of Tallahassee
RE: Sequential word repetitions in the VMS - Torsten - 08-02-2016
@Don
Your wrote on your website "There are two VMS words that do not fit the deconstruction pattern." As none fitting or "invalid" words you name 'she' and 'shee'.
'she' occurs 25 times. On page <f103> a sequence of 4 words resulting in 'she' can be found ('shek' -> 'sheek' -> 'shek' -> 'she').
See You are not allowed to view links. Register or Login to view.
'shee' occurs only 14 times. But on page You are not allowed to view links. Register or Login to view. it occurs 4 times.
See You are not allowed to view links. Register or Login to view.
RE: Sequential word repetitions in the VMS - don of tallahassee - 09-02-2016
I am not sure what point you are making or question you are asking. Are two exceptions out of 505 too many exceptions? If my ideas are in the right ballpark, would 4 repetitions of an ingredient's measurement and use in four different recipes on a page be too many for some reason?
If I added one more code to Group V (EVA = e), it would take care of the two exceptions. I just have not decided if that is correct or if there are two exceptions for some reason that I do not understand, an idea I find possible.
Don't you find the fact that the other 503 do fit to be more interesting?
I invite you to look at Table X at my fumblydiddles.com site to see an old, outdated version of what those 505 words might mean if the codes are decoded as shown at my site at page 24 of:
You are not allowed to view links. Register or Login to view.
The first part (the first 23 pages) isn't much to look at, but the second part, where the words are more or less alphabetized and arranged seems to show an astounding possible group of readings for the words. Please understand that some of the attributions for some of the codes have changed since the time it was written. The herb abbreviations/attributions/codes of Group I/Table I are the ones usually changed (which leaves the overall structure of the codes in the words listed in Table X intact).
How can I do this with Voynich Lite and Table X for all these VMS words if not close to being correct? I am not doing anything untoward or underhanded to the words to make this happen. The pattern is there in the VMS words. All I'm doing with Table X is giving the pattern some coloration and organization so you can see it. The colors may be wrong, but the pattern is there.
For the 505 most common words, it takes 103 group I codes and 34 other codes (if I remember right) to successfully deconstruct 503 of them. 99% plus.
Voynich Lite uses 103 Group I codes and 20 others to successful deconstruct 465 of the Voynich most common words. Still about 92%.
What other system of understanding about how the words are put together gives these kind of stats for the easily understood version, much less the fully expanded version of the proof?
But, damn, the results are really, really strange.
I'm interested in reading the proposed meanings of the codes put forward by others. Maybe someone else will come forward with more acceptable meanings for the codes and thus the words.
Thank you.
Don of Tallahassee
RE: Sequential word repetitions in the VMS - -Job- - 10-02-2016
(05-02-2016, 11:54 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.I wonder whether it is possible to create a dictionary, by which each Voynich word can be matched to one plain text word in some language in such a way that the result is a meaningful text in that language. I.e.: can the Voynich MS be translated word for word?
This is really independent of whether the system behind the MS text includes encryption or not. Suppose the text was encrypted using many nulls, then several different Voynich words map to the same plain text word.
My tendency is to think that the answer is no. (but of course I don't know).
It's not out of the question. Unfortunately, it's difficult to settle this definitively, even for a single folio.
On the other hand, given a number of sample texts, it should be possible to determine the size of the largest sentence in a VM folio which can be matched to a sentence in one of the sample texts.
The result should reflect the probability that the VM is readable via a decoding process involving word-substitution, such as a codebook or synthetic language.
For example, an analysis of f27v (58 words) using five sample texts shows that the longest match is a 35 word sequence in Dante:
dshy, kchy, cheo, da, dy, dchy, kchey, kchy, dchokchy, dsho, dchir, chodchy, etcheody, shld, okcho, chy, kcheed, chl, chol, kod, o, oksho, doeeeesm, qoky, shkeeo, schodar, shkol, chotchy, cthodol, dsho, kchrrr, okeedy, dchschy, sotchdy, sho
stregghia, a, ragazzo, aspettato, dal, segnorso, ne, a, colui, che, mal, volontier, vegghia, come, ciascun, menava, spesso, il, morso, de, lunghie, sopra, se, per, la, gran, rabbia, del, pizzicor, che, non, ha, pie, soccorso, e
The samples contained sections from Dante (Italian), Pliny (Latin), Bible (Latin), Bible (Hebrew), Moby Dick (English). Punctuation and line breaks were ignored.
It's interesting that, for all 5 sample texts, the longest match is the sequence of 33 words starting with "cheo da dy" above - Dante just matched the initial "dshy kchy" as well. It suggests that f27v, as transcribed, is not really compatible with the sample texts.
For folio You are not allowed to view links. Register or Login to view. (57 words) the longest match consists of 21 words, and the matches differ more substantially between each sample. I have not looked at other folios.
If the sample size had been much larger, then we might be able to say something about the VM's compatibility with the sample languages, and not just the sample texts - though i'm not certain that is true.