+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Voynich Talk (https://www.voynich.ninja/forum-6.html)
+--- Thread: Mark Knowles' theories of pharma and other labels (/thread-3140.html)
Mark Knowles' theories of pharma and other labels - Mark Knowles - 01-04-2020
Having spent quite a lot of time recently looking into this area of the Voynich botanical labels and having spent some time studying Voynich labels as a whole I would reiterate that if one looks at this question through the lens of someone familiar with diplomatic ciphers of the early 15th century then certain explanations present themselves. The key features of ciphers of that time was the use of nulls, homophones and a glossary. Now in general nulls correspond to nonsense/meaningless text. Likewise homophones correspond to the scope for identical words being spelled in completely different ways. A glossary could apply in the context of rare symbols corresponding to specific words. I described before elsewhere how a diplomatic cipher key could be modified in a minor way to use glyph strings as an example of this.
The idea of a preponderance of null words i.e. filler text is something I have also addressed elsewhere. I am not yet persuaded by the idea that all the text is meaningless, but I find the idea that there is a significant amount of filler text hard to avoid.
In the past when I have mentioned ciphers the response has tended to be that the Voynich doesn't fit exactly with standard ciphers of the time, something I agree with now and did then, therefore it is said that it cannot be a cipher. However it is perfectly possible that it is a cipher highly influenced by cipher techniques of the time, but with some differences.
RE: Matching “pharma” / “small plants” labels in context - -JKP- - 01-04-2020
(01-04-2020, 01:44 AM)Mark Knowles Wrote: You are not allowed to view links. Register or Login to view.... The key features of ciphers of that time was the use of nulls, homophones and a glossary...
I don't really agree with this, Mark.
The key feature of the diplomatic ciphers collected by Tranchedino is the one characteristic that you didn't mention—that they are one-to-many and many-to-one. This is how most of the text was constructed. The other characteristics were in there to slow down decryption, but I wouldn't call them key features. I would call them secondary features.
Yes, there were a handfuls of nulls available for use with each cipher, but they do not constitute the majority of the text and the nulls in these specific ciphers sometimes resembled each other, which would make them easier for a cryptanalyst to pick out as nulls.
Yes, there was a glossary for each cipher, but they were not extensive, and they were used as much for saving space (similar to medieval abbreviations) as they were for encipherment. They were mainly common names and words.
For example, there might be one or two characters to stand for the word "pope" or the name of the pope, but this character was generally different from the others, which means, to a cryptographer, it would stand out as different from common words constructed with individual letters. It would slow down decryption, but not stop it. A cryptanalyst intercepting a diplomatic cipher can see fairly quickly that it's one-to-many/many-to-one because of the overall patterns and would know beforehand that it is likely to contain names of important people and places and that these might be compressed (abbreviated) or glossary-enciphered.
Quote:Now in general nulls correspond to nonsense/meaningless text. Likewise homophones correspond to the scope for identical words being spelled in completely different ways.
The monotone quality of the VMS is quite different from ciphers that use homophones. In most cases, the VMS text shouts "less" rather than "more". Variety is lacking so it's less likely that homophones are a significant property of the text.
Quote:A glossary could apply in the context of rare symbols corresponding to specific words. I described before elsewhere how a diplomatic cipher key could be modified in a minor way to use glyph strings as an example of this.
Rare symbols in the VMS are very rare. If they were somewhat rare, appearing once every few pages, for example, then maybe they might constitute glossary words. As it is, however, they tend to be used so little, one can essentially ignore them until the other 200 pages are deciphered. It's a lot of text. If. you have only half a page of text, it's difficult to tell what is rare, and it's more important to try to work out what uncommon characters mean because they are proportionally greater, but the VMS is 200 pages, so rare characters can be dealt with after the rest of the text is deciphered.
RE: Matching “pharma” / “small plants” labels in context - MarcoP - 01-04-2020
We drifted away from discussing Matching “pharma” / “small plants” labels in context. I suggest moving posts from #26 onwards to a new Mark Knowles' theories thread.
(01-04-2020, 01:44 AM)Mark Knowles Wrote: You are not allowed to view links. Register or Login to view.The idea of a preponderance of null words i.e. filler text is something I have also addressed elsewhere. I am not yet persuaded by the idea that all the text is meaningless, but I find the idea that there is a significant amount of filler text hard to avoid.
You seem to be quoting You are not allowed to view links. Register or Login to view. 's conclusions:
Timm&Schinner' Wrote:The proposed text generation method is not only supported by many details of self-similarities uncovered in the VMS text, and is fully compatible with the historical background, but also even quantitatively reproduces the key statistical properties. In particular, we were able to demonstrate that our sample “facsimile” text fulfills both of Zipf’s laws. Following Occam’s principle, this theory provides the optimal hypothesis available to explain all facts currently known about the VMS. It, however, does not totally dismiss the steganography hypothesis
Timm and Schinner correctly describe the idea of some meaning hidden within largely meaningless text as "steganography", while you blur the argument with terms like "diplomatic ciphers" (which, as you say, are different) and "homophones" (which, as JKP wrote, are not compatible with the structure of Voynichese).
I suggest you amend your terminology, replacing "diplomatic cipher" with "steganography", and quote Timm and Schinner, so that people can read a well-defined example of the techniques you are interested in. Also, you should make clear if you have anything constructive to add to their work, e.g. a different system producing something even closer to Voynichese, or a simpler system resulting in comparable results, or the details of how steganography could work in this context.
(01-04-2020, 01:44 AM)Mark Knowles Wrote: You are not allowed to view links. Register or Login to view.it is said that [the Voynich] cannot be a cipher
I am not sure who you are talking about: when reporting the opinions of others it's better to be specific and include references. As far as I know, options are only excluded by delusional solvers, e.g. You are not allowed to view links. Register or Login to view.: "the manuscript is not written in code at all". Such statements can be safely ignored.
RE: [split] Mark Knowles' theories - Koen G - 01-04-2020
Here you go
RE: [split] Mark Knowles' theories - Mark Knowles - 01-04-2020
I actually wanted to keep this in the context of the Pharma labels as that was the context in which I was addressing the comments of others and in which my owns thoughts were focused. Anyway I don't object to the thread being moved, so that's OK.
There is a very literalist approach, which I have referred to, namely that Voynichese must be identical to ciphers of the period in order for there to be any parallels. For example, I am suggesting that there is a much higher proportion of null filler text in the Voynich than one sees in typical diplomatic ciphers, which actually makes logical sense for the author given the much greater length of the text of the Voynich when compared to diplomatic ciphers.
I was trying first of all to introduce broad concepts such as the same word being spelled more than one completely different ways, which fits neatly with the nature of homophones.
RE: [split] Mark Knowles' theories - Mark Knowles - 01-04-2020
I mentioned diplomatic ciphers as they were the most sophisticated ciphers of the time and the key features that we see are the use of nulls and homophones.
The parallels between the use of nulls and filler text are very clear. The parallels between the use of homophones and the idea that exact same word being spelled more than one completely different way is also relevant here.
In the pharma discussion many different explanations were put forward for the evidence we see. A high proportion of null words and the exact same word spelled more than one different way was not mentioned and yet this explanation I think is very worthy of consideration.
RE: [split] Mark Knowles' theories - Mark Knowles - 01-04-2020
I wasn't quoting Timm, the idea of the high proportion of filler text I have come to independently looking particularly, though not exclusively, at labels such as the pharma labels.
RE: [split] Mark Knowles' theories - ReneZ - 01-04-2020
There is probably some confusion, since the term 'homophonic substitution cipher' uses the word 'homophonic' in a different way than what it normally means:
You are not allowed to view links. Register or Login to view.
To talk about 'homophones' in the sense of 'equivalent letters in a homophonic substitution cipher' is probably not correct (and certainly even more misleading).
RE: [split] Mark Knowles' theories - -JKP- - 01-04-2020
I think you missed the point that Marco was trying to make. When you are talking about a high proportion of filler text, it frequently is an example of steganography, something that has been discussed quite frequently in the context of the VMS.
Whether you came to the idea independently or not, it's still a good idea to use (and acknolwedge) the terminology that others are using (and which are part of the cryptanalysis lexicon) for these forms of codes. And also to acknowledge that the diplomatic ciphers (as collected by Tranchedino) do not include large amounts of filler text (are not in any way steganographic).
RE: [split] Mark Knowles' theories - Mark Knowles - 01-04-2020
I brought this up in the context of pharma labels as it was observed and as I have also observed that there are some examples of different plants with the same labels; moreover it is the case that there are non-plants with the same labels as plants elsewhere in the manuscript. In addition the question of the uncommonality of the plant label in the accompanying text was referred to.
Suggestions were made to explain this, such as:
1) Maybe plant labels aren't really names.
2) Maybe Voynichese is a highly inflective language as suggested by Koen.
(Elsewhere the possibility of the use of abbreviations has been suggested.)
A highly inflective language would still mean that the same/related word would in general have to some degree the same spelling as would a word that was sometimes abbreviated or not in different ways. However a word such as we see with the use of homophones that can be spelled in completely different ways would be much harder to spot.
The concept of nulls and filler text to me seem to be intrinsically related.
I would suggest renaming the thread to "Mark Knowles theories' of pharma and other labels"