16-08-2016, 07:19 PM
This is in response to a post from Sam G. in the Leo thread, but maybe it's worth starting a new thread on this.
(I hope that this isn't actually making Anton's effort more difficult rather than less.)
Sam G. wrote:
I can't go into all of it....
I am fully convinced that the conclusions from Friedman and Tiltman (that it's not a cipher) are not to be challenged.
The key point is, though, what they mean precisely with 'cipher'.
Nowadays, Nick Pelling is clearly in the 'cipher' camp. However, he may have a wider definition of cipher than what Friedman and
Tiltman intended.
From the area of linguistics, there isn't nearly as strong an expert opinion as in cryptography.
One of the statements I clearly remember from Jacques Guy is that, for him, the Voynich MS demonstrated a huge weakness in linguistics, in the sense that he could not think of a test that decided whether the contents are meaningful or not.
For me, after well over 20 years, I am not even certain that there is a meaningful text behind it.
If not, the question between cipher or language disappears.
My tendency is, however, to expect meaningful contents, but I would not be in the least surprised if someone some day demonstrated that there isn't. I have myself found several clear indications that the text is not arbitrary, which suggests a meaning. However, there are also places in the MS where the text looks exactly like arbitrary 'filler' material.
Another prediction that may well never be verified:
if one day someone explains how the text was generated, and there is a meaning behind it, I expect that the method will be relatively straightforward, and one might even be arguing whether this should be called a cipher or not.
I agree with Helmut's point. Indeed, all comparison statistics (from plain texts) have been made based on printed texts that have usually been spell-checked and certainly have no abbreviations. This is of course convenient since this can be done fully automatically.
One would have to be able to make statistics based on handwritten texts, that may not have any orthograpy rules, and may be abbreviated.
This, of course, would require a very significant manual effort, and it can be done in many different ways.
(I hope that this isn't actually making Anton's effort more difficult rather than less.)
Sam G. wrote:
Quote:Anyway, as I'm sure you are aware, the expert opinions are very strongly against the view that the VMS text is a ciphertext of any kind. Most notably here you have William Friedman, John Tiltman, and Jim Gillogly, all three of whom studied the VMS for decades and concluded that it was not written in cipher. Maybe you can find a cryptographer somewhere who thinks that the VMS is a ciphertext, but you won't find many, and certainly you won't find one with the reputation or demonstrated codebreaking ability of any of these three.
As far as professional linguists go, you have Jacques Guy, James Child, and more recently Stephen Bax who, although differing in the particulars, have clearly stated that the VMS is written in some unusual natural language, and is not encrypted. I believe there are some other linguists who have expressed this view in the list archives, and I am aware of no linguist who has ever expressed any other view. So we have unanimity here among several experts.
So I think it's clear that if we are going to decide, based on the translation/summary of some casual remarks from an art historian posted on the web possibly even without her knowledge, that depicting Aries as a goat is nothing out of the ordinary, then surely the nearly unanimous opinion of many expert cryptographers and linguists who have put an enormous effort into researching the VMS text must carry an even far greater weight. The VMS is not encrypted, and it is written in an unusual, otherwise unknown natural language.
Now, I know I have seen you express the view that the VMS in fact is a ciphertext, although interestingly you do not state this on your own website. I'm sure I could dig up the references if you want. That means that you hold a view that is completely in contradiction to expert opinion, despite your stated view that expert opinion should be respected.
So, from the standpoint of your view that the VMS text is a ciphertext, perhaps the idea that there is nothing unusual about the illustrations and no need to posit any foreign/ancient influence in the VMS might make some sense.
But how do we possibly reconcile a 100% medieval Western European origin for the VMS with the fact that it is written in an otherwise unknown language?
Should the fact that the VMS is written in an otherwise unknown language at all influence our ideas about what is and what is not possible regarding the origins and meaning of the illustrations, and of the content more generally?
I can't go into all of it....
I am fully convinced that the conclusions from Friedman and Tiltman (that it's not a cipher) are not to be challenged.
The key point is, though, what they mean precisely with 'cipher'.
Nowadays, Nick Pelling is clearly in the 'cipher' camp. However, he may have a wider definition of cipher than what Friedman and
Tiltman intended.
From the area of linguistics, there isn't nearly as strong an expert opinion as in cryptography.
One of the statements I clearly remember from Jacques Guy is that, for him, the Voynich MS demonstrated a huge weakness in linguistics, in the sense that he could not think of a test that decided whether the contents are meaningful or not.
For me, after well over 20 years, I am not even certain that there is a meaningful text behind it.
If not, the question between cipher or language disappears.
My tendency is, however, to expect meaningful contents, but I would not be in the least surprised if someone some day demonstrated that there isn't. I have myself found several clear indications that the text is not arbitrary, which suggests a meaning. However, there are also places in the MS where the text looks exactly like arbitrary 'filler' material.
Another prediction that may well never be verified:
if one day someone explains how the text was generated, and there is a meaning behind it, I expect that the method will be relatively straightforward, and one might even be arguing whether this should be called a cipher or not.
I agree with Helmut's point. Indeed, all comparison statistics (from plain texts) have been made based on printed texts that have usually been spell-checked and certainly have no abbreviations. This is of course convenient since this can be done fully automatically.
One would have to be able to make statistics based on handwritten texts, that may not have any orthograpy rules, and may be abbreviated.
This, of course, would require a very significant manual effort, and it can be done in many different ways.