The Voynich Ninja

As a noob I looked at the manuscript and found some words that are nearly only exclusive at the line start like dshedy, ycheor, ycheol, dchedy, ychain. I also found some nearly exclusive at the line end like chary, opam, orom, okam (interestingly some of these in front of line cuts by plants), but I need to look at that in more detail to say something. 

However, the line start words seem very convincing for me. Also this words do not include the paragraph words, so they are not at the start of a paragraph. Also other have found similar patterns bevor  You are not allowed to view links. Register or Login to view.

How can that be explained? Is every line the start of a new sentence? But how to fill the line that you have such a smooth ending? Or does the text maybe contain filler words at the start and the end? Or words that starts the coding of a line? 

What are your thoughts and ideas about that?

(05-11-2025, 01:56 AM)Kaybo Wrote: You are not allowed to view links. Register or Login to view.How can that be explained? Is every line the start of a new sentence? But how to fill the line that you have such a smooth ending? Or does the text maybe contain filler words at the start and the end? Or words that starts the coding of a line? 

What are your thoughts and ideas about that?

There are a few possible explanations:

1.) the text is meaningless and the author is using filler words, which would be easy if the entire text is filler words. However, the "meaningless" interpretation has its own logical inconsistencies that make it, in my opinion unlikely.

2.) the text is "partially meaningless", as in, there are SOME filler words they use to round out a sentence/line. Again, seems hard to believe the text is only partially meaningful. Would make sense with "glossolalia" interpretations. Which again I think are unlikely given a variety of inconsistencies with other things we know

3.) the text is meaningful, and each line really is a sentence/complete thought.

There are a few categories of this:
3a.) the text is copied from an earlier script, and that earlier script was designed/reworded specifically to have every line act as a functional unit/sentence. This is logically consistent with much of what we know about the manuscript, especially with those who posit it's copied from an earlier text

3b.) the text was written in a stream of consciousness manner, but some property of the script/language/constructed language, allows for the enhanced ability for the author to estimate the exact space they will need to compose a sentence, and for reasons unknown, they had a strong preference and ability to write 1 sentence per line. This one would be quite extraordinary if it were the case, but in theory possible with constructed language hypotheses, if the constructed language has characteristics which allow for this sort of "on the spot estimation" of the space they need to finish up a thought. If this were the case, it may imply the text/meaning is much denser than previously imagined, since longer thoughts/sentences are easier to estimate the exact space you need.

Besides these options, it's quite hard to imagine any explanation which fits what we see in the manuscript. If text flowed from one line to the next, without lines acting as natural sentence enders/break points, it's really quite hard to imagine how that would work. Would words that end sentences act as a "\n" newline character, but only sometimes? That really doesn't make sense. 

Of course there could be another explanation we haven't thought of, but most likely, given the abundance of evidence we have that the text is meaningful, it's probably 3a or 3b if I had to guess. My money being on 3b for a variety of reasons.

(05-11-2025, 10:09 AM)qoltedy Wrote: You are not allowed to view links. Register or Login to view.Besides these options, it's quite hard to imagine any explanation which fits what we see in the manuscript. If text flowed from one line to the next, without lines acting as natural sentence enders/break points, it's really quite hard to imagine how that would work. Would words that end sentences act as a "\n" newline character, but only sometimes? That really doesn't make sense. 

Of course there could be another explanation we haven't thought of, but most likely, given the abundance of evidence we have that the text is meaningful, it's probably 3a or 3b if I had to guess. My money being on 3b for a variety of reasons.

There are many other possible explanations:
- the sequences at the ends of the lines are abbreviations, to avoid splitting words (or maybe the manuscript is enciphered and the nature of the cipher doesn't allow splitting words at all)
- the words are actually split between lines, and special cipher symbols are used to indicate that the continuation of the encoding is on the next line, and other symbols at the beginning of a line indicate a continuation from the previous line
- lines often contain lists of examples and whenever there is little space left and it's not very important to continue expanding the list, the line is terminated with something like "etc.", "and so on". In which case the next line would often start with something like "on the other hand", "however", etc.
- there are several possible ways to encode a sequence (one-to-many cipher) and naturally when the space is constrained (near the end of a line or otherwise), shorter sequences are preferred, while at the beginning of a line the longer sequences are more frequent.

I'm not sure any of these explanations is very good, but over the years there have been proposed dozens of different plausible explanations for LAAFU. The question as I see it is not "how is it possible at all", but "which of many explanations is the right one".

Any explanation of how lines start/end should also take into account "vertical pair" preferences.

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

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

(05-11-2025, 10:09 AM)qoltedy Wrote: You are not allowed to view links. Register or Login to view.If the text is copied from an earlier script

Again, this was almost certainly the case.  Because of the care required, and the difficulty of correcting mistakes, it would be insane to write anything straight from brain to vellum.   We may assume that the Author wrote a draft on paper and then recruited a Scribe (or various Scribes) to clean-copy it to vellum.

Quote:for reasons unknown, they had a strong preference and ability to write 1 sentence per line

Indeed there are no convincing precedents or reasoning that would explain how every sentence of a paragraph, except the last one, could be fitted precisely between the left and right rails, with hardly any stretching or shrinking of all the characters.  

Maybe the lines were all short, and the rest of the space was completed with filler words.  But there seems to be no statistical evidence of such systematic fillers...

Quote:If text flowed from one line to the next, without lines acting as natural sentence enders/break points, it's really quite hard to imagine how that would work.

Indeed, the text layout everywhere in the VMS looks totally like that of paragraphs in most European manuscripts.  Where line breaks were chosen by the Scribe, based on the available space and ignoring the breaks on the source text, and generally are just equivalent to an inter-word space.   There is even good evidence that this was the case, such as pages You are not allowed to view links. Register or Login to view. and f112v, or all the pages where the text is bounded or interrupted by illustrations.

Thus the "line as a functional unit" (LAAFU) thory needs more evidence than just some statistical anomalies at word ends, which can have several other explanations.  Such as:

4. The natural line-breaking algorithm used by scribes has the side effect of making the first word of the line longer than average, and the last few words shorter than average.  Since the most common words tend to be shorter, this bias can significantly distort the word frequencies at line ends.  And since character and digraph frequencies are dominated by their occurrence in the most common words, this bias also affects the character and digraph frequencies near line ends.

5. The scribe's handwriting will naturally be more "relaxed" at the start of each line, and become more cramped as it gets closer to the right rail.  As a result, in the digital transcription files, words at the beginning of the line are more likely to be incorrectly split (e.g. sorChaiin -> sor Chaiin), whereas words near the end of the line are more likely to be incorrectly joined (e.g. ar ol dy -> aroldy). 

6. Someone already mentioned hyphenation marks that could have been placed at the start of the second line rather than at the end of the first line.

7.  Another feature of You are not allowed to view links. Register or Login to view. of the time is the start of each sentence or important topic was marked in some way within the text (in that example, with a period and capital letter) but in addition another marker was written in the margin of that same line.  Maybe there VMS scribe used a similar convention.  That is, the Author's draft has sentence starts markers, and when the Scribe copied one, he also added a letter at the start of that line.   In that example red ink was also used to emphasize those sentence starts.

And surely there are many more possible explanations, other than LAAFU, that we haven't thought of yet...

All the best, --stolfi

(05-11-2025, 02:21 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Thus the "line as a functional unit" (LAAFU) thory needs more evidence than just some statistical anomalies at word ends, which can have several other explanations.  Such as:

So we have not only words in the end, we have also words at line start. The problem with the natural line-breaking algorithm explanation for me is the lack word pairs for the line start words. If it would be a line-breaking algorithm than you would expect that there would be combinations of start and end words. For dshedy there is never the same end word in the row above. 

.......................................different words
dshedy...................................................

(05-11-2025, 04:16 PM)Kaybo Wrote: You are not allowed to view links. Register or Login to view.only words in the end, we have also words at line start

Sorry, where I wrote "line ends" I should have written "both ends of the lines."

(05-11-2025, 02:21 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.

(05-11-2025, 10:09 AM)qoltedy Wrote: You are not allowed to view links. Register or Login to view.If the text is copied from an earlier script

Again, this was almost certainly the case.  Because of the care required, and the difficulty of correcting mistakes, it would be insane to write anything straight from brain to vellum.   We may assume that the Author wrote a draft on paper and then recruited a Scribe (or various Scribes) to clean-copy it to vellum.

Quote:for reasons unknown, they had a strong preference and ability to write 1 sentence per line

Indeed there are no convincing precedents or reasoning that would explain how every sentence of a paragraph, except the last one, could be fitted precisely between the left and right rails, with hardly any stretching or shrinking of all the characters.  

...

And surely there are many more possible explanations, other than LAAFU, that we haven't thought of yet...

While I agree that the "copied from an earlier text" hypothesis is the easiest to imagine from our perspective, and ideas such as it being "insane" to write straight stream of consciousness to vellum, we are indeed dealing with quite an "insane" document historical artifact to begin with.

I agree that there are no convincing precedents for the "stream of consciousness, while perfectly writing every sentence  to fit the space" that we know of in historical writing. But again, the Voynich Manuscript has no precedent. It is completely unique in several ways. The script, the statistical patterns, the illustrations, all seem like something "probably impossible/no good reason for it". Very clearly the author was on a different wavelength than the rest of us, both their contemporaries at the time, and everyone since then.

I think there is "reasoning" for why the author may have been able to write like this, and why they decided to.

Going back to the constructed language hypothesis, if the author intentionally designed their system without punctuation, they may have naturally concluded that a line should act as a functional unit, making a sort of built-in sentence demarcation for each line. It would actually be quite an elegant solution to avoid creating sentence terminating characters, to make each line act as a functional sentence marker.

As for the "insanity" of going straight from thought to vellum, we have several strong pieces of evidence the scribe(s) drew the illustrations and wrote the text in the same sitting with the same pen. The drawings are generally quite crude and do quite frankly, for the most part, seem like they could have been made with no prior planning. This indicates it's still entirely possible, and supported by evidence, that they were writing the text quickly without much planning. While we may not entirely understand why they would be willing to draw/write in such a lax way on such an expensive medium, we can only speculate based on the evidence in front of us.

As for the "how", that is also a matter of speculation, but given all we know and don't know about the particular ways the text was generated, it may be possible that some aspect of the writing system allows for enhanced estimation of sentence length in real time writing. We wonder why someone would make up their own language/script instead of using any existing language or cipher. Perhaps because their system provided advantages, such as this enhanced line length estimation ability.

Obviously we're all speculating here, but I think this line of reasoning is both logically consistent and consistent with the evidence of what we observe in the manuscript. It's just very counterintuitive from the perspective of people who didn't write the manuscript.

(05-11-2025, 05:40 PM)qoltedy Wrote: You are not allowed to view links. Register or Login to view.we are indeed dealing with quite an "insane" document historical artifact to begin with. ... the Voynich Manuscript has no precedent. It is completely unique in several ways.

True.  But the insanity of writing directly on vellum would be independent of any insanities in the contents. 

That is, the hypothesis "the Author had those crazy ideas and wrote straight to vellum" is much less likely than "the Author had those crazy ideas and wrote a draft for a Scribe"

All the best, --stolfi

What if the manuscript are just tables of numbers. In a lot of astronomy books of that time we find tables that can predict things in the future. That are mostly numbers, star names, zodiac signs. In that case you would get similar words in every column. That would also explain why you would get word repeats. 

Also I think if you look for words that are nearly exclusive or very often at the line start you will find more that 20 words that match and most of the words that are not single used words are that kind of words in at the line start. So there is something going on at the line start.