• Options
• Login
• Desktop Style

Has anyone ever "deciphered" a paragraph?

• Index

  • The Voynich Ninja
  • Voynich Research
  • Analysis of the text
  • Has anyone ever "deciphered" a paragraph?
• Has anyone ever "deciphered" a paragraph?

• RE: Has anyone ever "deciphered" a paragraph?

  Mark Knowles > 27-04-2021, 12:51 PM

  (22-04-2021, 08:21 PM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.What I can say is that the previous list is an extract (subset) of the nomenclator part of one diplomatic cipher.
  
  Both are contrary to what we see in the Voynich MS, as I have already said many times.
  
  That should be more useful than just philosophising about it.

  
  I said: "It is perfectly possible for a cipher to decrease entropy."
  
  You said: "It would be of great interest to have a historical example of this."
  
  I provide you with a historical example illustrating how it is perfectly possible for a cipher to decrease entropy.

• RE: Has anyone ever "deciphered" a paragraph?

  geoffreycaveney > 27-04-2021, 12:55 PM

  (27-04-2021, 12:51 PM)Mark Knowles Wrote: You are not allowed to view links. Register or Login to view.I said: "It is perfectly possible for a cipher to decrease entropy."
  
  You said: "It would be of great interest to have a historical example of this."
  
  I provide you with a historical example illustrating how it is perfectly possible for a cipher to decrease entropy.

  
  Mark, could you please post the historical example here, on a public forum, so that we can all follow along with the claims that you are making here, on a public forum. Thank you.

• RE: Has anyone ever "deciphered" a paragraph?

  Mark Knowles > 27-04-2021, 01:03 PM

  (27-04-2021, 12:55 PM)geoffreycaveney Wrote: You are not allowed to view links. Register or Login to view.Mark, could you please post the historical example here, on a public forum, so that we can all follow along with the claims that you are making here, on a public forum. Thank you.

  
  I have explained why that is difficult. I will contact the archive to ask them if I can share this one cipher key on a largely private forum(Ninja is not as public as PInterest), though I would have to ask people not to post it outside of the one time that I have. I may also share it on a temporary basis and remove it subsequently or provide a temporary link from which it can be downloaded.

• RE: Has anyone ever "deciphered" a paragraph?

  Koen G > 27-04-2021, 01:12 PM

  Mark, if it would appear impossible to share the document, surely it's not inappropriate for you to tell us how the method works?

• RE: Has anyone ever "deciphered" a paragraph?

  Anton > 27-04-2021, 01:29 PM

  Yes indeed, the cipher principle is long since in public domain anyway

• RE: Has anyone ever "deciphered" a paragraph?

  Mark Knowles > 27-04-2021, 01:36 PM

  (27-04-2021, 01:12 PM)Koen G Wrote: You are not allowed to view links. Register or Login to view.Mark, if it would appear impossible to share the document, surely it's not inappropriate for you to tell us how the method works?

  
  Fair enough, I will.
  
  My caution has been just that I don't want to antagonise the Florence State Archive or cause them any annoyance or nuisance. It is very likely that I am going to ask the archive for more photoreproductions in the future and sometimes locating the documents when the customer cannot visit the archive is a nuisance for them. I would be interested in seeing the original letters from which this specific cipher key was generated. Also the archivist who generated the cipher key wrote an published book in the archives called "Crittografia Fiorentina" and there may be some interesting comments there. I don't want to create extra work for the archivists especially as I am unlikely to visit the Florence State Archives any time soon.

• RE: Has anyone ever "deciphered" a paragraph?

  ReneZ > 27-04-2021, 03:15 PM

  I have not seen an example of a diplomatic cipher that would reduce entropy. But I am afraid that I am repeating myself.

• RE: Has anyone ever "deciphered" a paragraph?

  Mark Knowles > 27-04-2021, 03:49 PM

  (27-04-2021, 01:29 PM)Anton Wrote: You are not allowed to view links. Register or Login to view.Yes indeed, the cipher principle is long since in public domain anyway

  
  Ok.
  
  First I will give some historical background and context.
  
  My study of the Rosettes folio lead me to a hypothesis of authorship and a dating of the Rosettes folio to the mid-1430s and so a likely dating of the whole manuscript to the 1430s assuming the manuscript was written over a short period of time. I won't describe how I came to that conclusion here as this is a long argument.
  
  I, subsequently, learnt that that proposed author had very closely family ties, through his brothers, to the world of Milanese diplomatic ciphers.
  
  I have done a lot of research on early 15th century and late 14th century diplomatic ciphers, with a particular interest in those associated with the Duchy of Milan. In 1447 the Castello di Porta Giova was burned to the ground by Republicans who formed the very short lived Ambrosian Republic. The castle contained the Milanese chancellery where the cipher records amongst others were kept and so the State Archives of Milan it appears do not have any cipher records from before 1445(I think one or two escaped the flames probably as they weren't in the chancellery at that time).
  
  So I have endeavoured to reconstruct the Milanese cipher record on the basis of other archives. These can be enciphered letters sent from Milan to other city states or political actors. They can also be cipher keys used by states or other political actors for communication with Milan. Of course it does not mean that if a cipher key was used for communication with Milan that it was generated by the government of Milan, it could have been produced by the other communicating party, however Milan was such a major player that it is more likely that they produced the cipher key when the other party was the more minor player.
  
  Anyway I have found examples of Milanese ciphers from 1397, between 1395 and 1404, 1424 and 1445. I am still on the look for more. Given my mid-1430s dating the two examples that are most relevant are the 1424 and 1445 ciphers. The 1445 cipher key is in the Milan State Archives and appears to be the earliest Milanese cipher that they have. It is a cipher key that has been reconstructed from enciphered letters, but is described as partial or incomplete by Lydia Cerioni in her book. When I last enquired some time ago the archivist didn't seem to know where to look for the original letters. I have found some clues as to where they might be in the archive and I will follow these up with them.
  
  The 1424 cipher key which is the particular focus of this discussion was generated from enciphered letters by the 19th century Florentine Archivist, Abbot Pietro Gabrielli. Gabrielli had deciphered well over 1000 enciphered letters throughout history from the Florentine and other archives.
  
  When I first saw the photoreproduction of this cipher key I was struct by its relative complexity given its 1424 dating. First of all I noticed that it clearly had features that I had not seen before 1445, these being mapping of letter pairs and common word mappings. It had more substitutions than any of the ciphers that I had seen from around this date. And most interestingly it had features that I have not seen in earlier or later ciphers. However in some respects in was simpler than one would expect from that time, no use of nulls and only homophones on vowels. It is interesting to compare, the sophisticated for the time, but very different cipher key that Gabrielli generated for another letter(s) from 1424. This can be found in "Die Anfange..." by Meister. It is worth noting that in 1424 the 1st war in Lombardy is taking place between Milan and Florence. (It is the case that when the features of these two cipher keys are combined they encompass all the aspects and more of diplomatic ciphers that we see up to the end of the 15th century and beyond. All the evidence shows that the early 15th century was a period of rapid diplomatic cipher development and innovation, given the simplicity of the ciphers from 1395 compared to the complexity of ciphers from 1445. There are also features of other diplomatic ciphers from this period that cease to be used in later ciphers, showing the kind of experimentation at that time.)
  
  The 1424 cipher key I focused on is headed by Gabrielli "Zaninus and Conradinus". It am pretty confident that this refers to Zanino Riccio and Conradino Vicecomes. Zanino Riccio was secretary to the Duke of Milan and Conradino also served the state. (The other different 1424 cipher key is headed to a representative/envoy of the Florentine government). This reason and others made it seem mostly likely that the Zanino cipher was of Milanese construction and not Florentine.
  
  So what struck me as interesting about it was that it had what appeared to be mapping of 3 or 4 letter common prefixes and suffixes. This mapping of parts of words, i.e. word endings and word beginnings I haven't seen elsewhere before or since.(I don't count the 2 letter pair consonant vowel mappings as they are more general). This would mean mapping in English, for example, word endings like "ing" or "ent" or word beginnings like "ex" or "pro".
  
  Furthermore a number of these mappings are to Roman numerals, which I thought of particular interest.
  
  This structure decreases entropy for 2 reasons:
  
  1) by mapping regular structures within words. As there are patterns within words in a language such as common letter sequences. For example, in the English language we can see the sequence of letters "ing" often at the end of words, but letter sequences like "xx" not at all. So it is possible to take the seemingly complex variable letter patterns in English words and map them to more ordered patterns. The same applies to other languages including latin as in the cipher key we are referring to. This mapping process is a little bit similar to compression. 
  
  2) by mapping to Roman numerals one is mapping to a much less variable and much more ordered structure.
  
  Below I have repeated the structure that I referred to copied from the generated cipher key in an earlier post.
  
  se -> XI
  mi -> XIIII
  qua -> XX
  si -> XVIIII
  su -> XXIIII
  te -> LVII
  to -> LXII
  tu -> LXVI
  toc -> LXIIII
  sent -> XVIII
  tuc -> LXVIII
  mus -> XLVIII
  tam -> LII
  sen -> XIIII
  is -> LIII
  est -> XXIII
  set -> XVII
  sum -> XXVI
  cum -> X
  
  Of course Latin is the underlying language (not English), which seems to be the case for all diplomatic ciphers of the time.
  
  It should be noted that the cipher key also uses the more traditional mappings to invented characters for quite a few substitutions and a couple of homophones for vowels which is entropy increasing. There are also the standard mappings of individual letters which has no effect on entropy.
  
  I think it is very plausible that in some later Milanese ciphers there would be a move over to only employing the common word beginning/ending mapping to a structure akin to Roman numerals rather than some of the more traditional mappings. Given the huge jump in complexity between 1404 Milanese diplomatic ciphers and the 1424 example I think it very likely that there would be further significant advances over the next decade which could account for other peculiarities that we see in Voynichese. Though I believe the closest relative to Voynichese to be a diplomatic cipher of the 1430s I suspect there to be a further degree of complexity in Voynichese that a diplomatic cipher couldn't get away with in any kind practical non-specialist usage. (I would point out that from 1404 to 1412 the Duchy of Milan was in a state of collapse; for this reason and others I doubt that there would have been significant advances over this period and so the rate of advances more pronounced from 1412 to 1424.) I speculate that there would have actually been a period of simplication from the late 1430s onward as I imagine that the ciphers would have become too complicated to be used effectively by ordinary diplomats, this would explain certain more sophisticated features going out of use.
  
  As an example consider the following:
  
  If we look at common English word endings with possible mappings:
  
  ght -> bz
  ing -> cz
  ous-> dz
  ise -> fz
  tion -> gz
  ism-> hz
  ery-> jz
  ment -> bx
  age -> cx
  ate -> dx
  ous -> fx
  ive-> gx
  ness -> hx
  ship -> jx
  
  (english word ending->mapped representation)
  
  If we now look at common English word beginnings with possible mappings:
  
  the-> qlv
  dis-> qlw
  pre-> qly
  trans-> qmv
  mis-> qmw
  inter-> qmy
  pro-> qnv
  sub-> qnw
  str-> qny
  
  (english word beginning->mapped representation)
  
  So:
  
  string-> qnycz
  
  This is about the same number of letters, but with a more ordered representation.
• Next Oldest Next Newest