So, for quite a while I've wondered why Torsten Timm's auto-copying hypothesis hasn't made a bigger splash in the Voynich community.
Perhaps it would give the community more confidence in the auto-copying hypothesis if it could be statistically (not just subjectively) shown that the average edit distance between words in the VMS is anomalously low compared to other texts.
In theory, it should be possible for someone with more computer programming experience than myself to calculate, using certain pre-determined rules about how to calculate edit-distances:
1. The edit-distance between any two successive words. Then, display both a color-scale-coded, zoomed-out map of the VMS according to successive word edit distances (where red=1, orange=2, etc), along with the overall average successive word edit-distance in the VMS, and compare these with standard texts from other languages.
2. The average edit distance between any particular VMS "word" and all other words in the VMS, weighted according to the frequency of those other words (so, a word that is used twice as often will have its edit-distance count for twice as much as that of a different word only used half as often).
3. The average of this average. Compare with other standard texts, etc.
4. The average edit-distance between all two-word combinations (not just successively) on any particular page, weighted by the word frequencies. Compare with other standard texts, etc.
I predict that, if someone were to do these sorts of computational attacks on the VMS and other standard texts, one would find stark contrasts. The VMS has no two-word combinations like "throughout" vs. "swimming," to pick two random English words. In the VMS, if you had the word "throughout," you'd find plenty of words like "thrughut" or "trough" or "rouge" or "hotrug" but no word like "swimming" that is based on an entirely different word stem (aside from the other two word stems one would find in the VMS, if you were starting with "chedy." Yes, there are words like "ol" and "aiin," but no words like "nemrie" that are unassociated with any of those stems).
Natural languages don't work like this. This computational attack would settle, once and for all, that the VMS must either be a constructed language, a cipher, or gibberish.
Many people have considered the rosettes image to represent a map. Some suggest that the entire collection of circles is a map, while others concentrate on one of the circles. Known propositions of this type are from Diane O'Donovan, Nick Pelling, Jürgen Wastl and D. Feger, J.K. Petersen and from Marco Ponzi, but there are many more similar suggestions that predate all of these.
There could be a tendency to believe, by the sheer number of these suggestions, that there is indeed most probably a map here, but I am of a rather different opinion. I do not believe it is a map. For me it is something more abstract.
D'Imperio already implied that it could be some kind of synthesis of the author's vision of the 'cosmos' and I find this idea attractive.
I always had the feeling that the circles in the rosettes diagram refer to the individual circles in the remainder of the 'cosmological' part of the MS. A complete one-to-one mapping has not yet been achieved but there are a number of clear parallels. I know others have had the same idea. I still think there is more to be said and discovered about this. Note that the set of cosmological drawings we have may not be complete.
But I also have a much more specific hypothesis. Of course I can't prove it, just like nobody can prove that the illustration is a map. If I could, I would not be writing it here, in this way.
Let me present the pros and cons of this hypothesis.
First of all, I believe that the central circle represents the sky. I already mentioned it before. The sky is like a plane or sheet with stars held up by six towers.
The upper right circle I believe represents the Earth. Most of the buildings are in or near this circle, and just outside it is a T-O map. There is more to be said, but let me keep it short.
That leaves 7 circles......
These 7 circles could represent the seven planets.
They could also be associated with the 7 week days and 7 metals.
(Nick Pelling suggests in his book - I believe - that the individual circles in the cosmological section might refer to the planets, so this is a close parallel, if the above-mentioned mapping between individual circles and the rosettes image is real).
Now if this is right, can we decide which circle is which planet / week day / metal? Let me add the mapping first. I also add the association with the humours, but I am not sure if these are uniquely identified.
Moon - Monday - Silver - Cold and moist Sun - Sunday - Gold - Hot and dry Mercury - Wednesday - Quicksilver - ? Venus - Friday - Copper - Cold and moist Mars - Tuesday - Iron - Hot and dry Jupiter - Thursday - Tin - Hot and moist Saturn - Saturday - Lead - Cold and dry
I know of one MS where the planets and weekdays are represented by circles (though much more simple ones), namely Augsburg MS Ludwig XII 8. It is available digitally ad You are not allowed to view links. Register or Login to view. and I add the links for Mars and Tuesday below.
.You are not allowed to view links. Register or Login to view.
.You are not allowed to view links. Register or Login to view.
Only if a convincing mapping between the 7 circles and the 7 planets / weekdays can be made, will this hypothesis gain considerable weight. How can we do that? It would be good to have more examples like MS Ludwig XII 8.
Furthermore, we still have three more small items in the corners of the rosettes diagram (beside the T-O map).
In the lower-left corner there is the symbol that has been compared to a clock. For me, this is a reference to f67v2.
It also looks like a symbol for a metal, possibly, but not necessarily Quicksilver or Mercury. This needs to be confirmed, of course, in particular if such symbols were already used in the early 15th Century.
Beside that we have two sun faces, and here we encounter our biggest problem. Had we had one sun and one moon, we would be 2 down, 5 to go, but we don't. There's a minor hope that one of the two could be a Moon after all, but a moon with rays like the sun is really very unusual.
Further help in assigning the 7 planets would be if they were ordered somehow according to their humours. The two sun faces could just both mean hot, and we have cold in the other two corners.
There is another important problem, of course.
In the illustration, all circles on the outside are connected to each other, and connected to the sky in the centre.
This implies that the Earth would have a similar 'role' as the other planets, which is quite an unusual concept for the 15th Century of course....
Plenty of work left before this hypothesis gains any weight.....
Posted by: R. Sale - 21-07-2016, 07:28 PM - Forum: Imagery
- No Replies
In reference to pearlwort as Sagina procumbens - can anyone comment as to whether it has been discussed in any relation to the VMs herbal illustrations?
Recently I devised a neat little method to encode messages using the Voynich script. See You are not allowed to view links. Register or Login to view. for that. And if you want to have a crack at solving it without being spoiled first, then you need to STOP READING THIS POST RIGHT NOW.
=================================
In my original post I said that I would reveal the message and the procedure 1 year from now if nobody could figure it out in the meantime. Well, I realized that I'm not that patient and there are important implications of my method that I wanted to discuss right away, so here we go:
With my method, I was able to auto-copy "words" from previous lines that I had already written while still encoding a meaningful message. How was I able to do this? Because I had many degrees of freedom with regards to the choice of each letter. Why was that? Because each letter was really not expressing 1 out of 26 different possibilities, but in my case 1 out of 3 possibilities (more on that below).
We already know that in the VMS certain letters can apparently "replace" or "stand in" for other letters. They appear interchangeable. Therefore, the smallest semantic unit that we should really be dealing with is not the letter, but the "interchangeable group" to which these letters belong.
In my encoding procedure, I divided the Voynich script into 3 groups based off of Brian Cham's "curve/line system":
*curve letters
*line letters
*all others
(Edit: by the way, "a" counts as a dual curve/line letter, as per Brian Cham's system).
I let:
*curve letters represent dots
*line letter represent dashes
*all others represent null values that can be ignored
(Edit: so, to clarify, "a", being both a curve and line letter, is a special letter that single-handedly encodes "dot-dash").
Each "word" in my message then encoded a letter in Morse Code plaintext.
Let's say, for a particular letter, that I needed to encode a Voynich "word" that would go "dot-dot-dot-dash." I had all sorts of words to choose from to accomplish that. If I desired, I could start with a root like "chedy" and simply edit it until I had the desired pattern of curves and lines. I could take something like "4olam" that I had written a line above and just tweak it slightly if I needed to represent a close relative to the preceding dot-dot-dot-dash pattern. I could add gallows and other null characters to obscure the underlying pattern. And I could adhere to certain rigid sequential aesthetic rules (such as having gallows or 4o only in certain places) and still have the freedom to encode the information that I needed. It was not difficult to hold to certain aesthetic conventions (such as 4o only at the beginnings of words) AND encode the message AND do it quickly, especially once I got a few lines down and I could just start copy-pasting and tweaking what I had written immediately above to suit my needs.
And I wasn't even that picky about adhering to ALL of the aesthetic/structural conventions of the VMS. That's why you'll notice that some of my words don't look like proper Voynichese. I basically slapped this page together in about an hour. If I had taken more time on it, I could have made it look reminiscent of Voynichese to an arbitrarily close degree while still encoding the meaning.
In effect, every part of a letter that was not a curve or a line became a "null sub-character" or "null character component." With that many nulls in the message, and with the nulls taking the peculiar form of sub-components of characters, it would become well-nigh impossible to decrypt it. Plus, there was all that freedom to re-arrange those nulls for aesthetic/structural reasons.
But you might think: with so many null components, wouldn't it become cumbersome to decrypt the script back into readable text? Not at all! By the end of composing my message, with just about an hour of practice, I could easily ignore all of the null components and read off dot-dot-dot-dash straight off the page quite fluently.
Now, if you believe that the VMS was written in the 1400s, then of course Morse Code would not have been the plaintext.
But consider that I could have just as easily done something like:
*let 4o represent 2^4
*let all complex curve letters (other than EVA "e") represent 2^3
*let all complex line letters (other than EVA "i") represent 2^2
*let all "e" represent 2^1
*let all "i" represent 1
*let all gallows and other characters be null characters
Each Voynich "word," by adding up its numerical components, would then have a numerical sum which could then be related to a sequential letter of the alphabet. That's another way I could have done it.
Now, I'm not saying that that's the specific way that the VMS was written either. I'm just saying...that's ONE MORE way. There could be tons of possible ways of encoding meaning in the Voynich script if we imagine that letters are encoding a lot less information than we assume they are, and that interchangeable letter groups are really the smallest semantic sub-component rather than individual letters themselves, which could then be varied around, copied, tweaked, etc. simply due to the whims of the author, or his interest in the ease of writing, and/or out of a desire for the appearance of certain rigid aesthetic patterns in word and line structure.
And yes, I'm sure the historians will jump in and say that such ciphers were not known in 1400s Europe. Well, they should know that I also happen to be a fan of Rich SantaColoma's modern forgery hypothesis, and I am by no means wedded to the idea that the Voynich cipher (if it is indeed enciphered) has to be an old cipher.
Most of all, I wish more people would engage with Torsten Timm's mindblowing paper, while at the same time keeping their minds open that the VMS could still have encoded meaning even with all of the evidence that attests to auto-copying, letter interchangeability, weird entropy, weird "word" and line structure, etc. I have shown that it is possible.
Edit: One more thing that I anticipate will be brought up is the labels in the VMS. If each Voynich "word" is only encoding something smaller like a syllable or a letter, then how to the standalone labels make any sense? I don't know. Maybe the intention was to label things not with names, but with numerical values. Maybe it's all misdirection. How do we KNOW that the labels are intended to function as what we would consider to be ordinary name-labels?
I know I've seen the "ma+" bit that shows up on the You are not allowed to view links. Register or Login to view. marginalia a few times. Here's an example I just came across:
The manuscript I got this from is You are not allowed to view links. Register or Login to view.You are not allowed to view links. Register or Login to view., fol. 119v. The record says Southern German, mid 15th century. It can be found here (and it has more examples of the "ma+"):
You are not allowed to view links. Register or Login to view.
It really seems like this aspect of the VMS marginalia ought to be a known thing. Can anyone look at this manuscript and determine what it means?
This is my first post here, although I have followed online Voynich research for some time.
I come to you all not with a theory...but with yet another puzzle (as if the original VMS wasn't enough of one!)
Recently at my blog I put upYou are not allowed to view links. Register or Login to view. that I made using the Voynich script. As you will see, with this encipherment, I was able to recreate some interesting features of the VMS, especially its repetitiveness, while also conveying my information using a relatively simple procedure for unambiguously constructing and deconstructing the message.
Now, I'm not claiming that I just happened to stumble upon the method that was actually used in the VMS. The actual method used in the VMS, whatever it is, is most likely quite different. But I thought some posters on here might have fun trying to crack a code for a change where they have some assurance that there actually is light (a meaningful message) at the end of the tunnel.
So, check it out! If nobody figures out the correct decipherment in a year, I will post the solution and the procedure on July 18, 2017. Have fun!
There's a certain similarity between the design found around the central roundel on f70v2 (Pisces/March) and a border design commonly found on classical Greek pottery. Here's a comparison:
It's not an exact match of course, but in both cases we see small segments with a "four way" division separated by segments of 2-3 times their width containing "zigzag" patterns. It's also not hard to imagine that the simpler zigzag patterns we see in the VMS could have arisen from the "stepped meander" patterns of the Greek design via several generations of imprecise copying.
Of course, the resemblance could be completely coincidental, so what I'd like to know is if anyone knows of other patterns similar to the one found in the VMS, especially if there are any in the medieval Latin manuscript tradition.