Re: VMs: Re: Re: Inks and retouching

[Nick Pelling <nickpelling@xxxxxxxxxxxxxxx>]

Hi GC,

At 01:00 12/08/2004 -0600, GC wrote:
> A search of the archives will reveal my posts on the origin and approach to
> this material.  Not a magic key at all, but derived from 'cribs' into the
> text found in the work of Feely and O'Neill.

Shame on me, but I haven't read Feely's (1943) "Roger Bacon's Cipher: The 
Right Key Found": all I've seen of his labels are the five on 
http://www.voynich.nu/solvers.html taken from f78r -
        <okeedldlo>     FEMMININO               femminino
        <okeedy>        FEMMIN          femmin
        <daraloCThy>    ISTSNF(UNDU)(NTR)       istis infunduntur
        <deeedaly>      IMMCISN(NTR)            immiscuntur / imcistinantur
        <okaral>        FESTSN          festivi sunt
...plus his "ouari" crib you once mentioned on-list.

Face it, these are rubbish. :-(

Let's upgrade his transcriptions (Feely never saw the VMs itself, IIRC) 
based on the sidfiles:-

  <f78r.X.1;A>       okChdldlo =
  <f78r.X.2;A>       okChdy =
  <f78r.X.3;A>       dar aloCFhy =
  <f78r.X.4;A>       dChedaly =
  <f78r.X.5;A>       otaro dlAry - orary =     // "dlAry" uses the 
joined-up "Ar" glyph!
  <f78r.X.6;A>       okaral =

Firstly, the very first word - okChdldlo - looks *far* more to me like a 
Neal key than a "normal" VMs word ("dldl" appears nowhere else, right?), 
and its position (the top left of the page) only serves to reinforce that 
opinion. If it has a meaning, I'd say it's as a Neal key, not as a label 
per se. FWIW, it's in a section where we do get free-standing l's, so the 
odds are high that "dl" should be read as "d-l-", rather than as a verbose 
pair.

Secondly, the fifth label "otaro dlary - orary" also has that 
suspicious-looking "dl" pair in, and is placed at the top right of the 
second paragraph. I think that this too seems far more likely to be a Neal 
key (ie "a cryptographic helper relating somehow to decoding the following 
section") than a label (ie "a crib referring to an adjacent picture"). 
[FWIW, I think it would be interesting to analyse all occurrences of "dl" 
throughout the VMs, but that's a job for another day...]  [BTW, it's 
possible that the right-justified "titles" we see throughout the VMs may 
also be another form of Neal keys].

Thirdly, precisely the point I was trying to make was that this kind of 
simple substitution (even over an abbreviated Latin plaintext) fails to 
please when you look at all the common pairs. Never mind his reading "Ch" 
as "ee" (I can forgive him that), but seeing all the occurrences of "ar", 
"al", "or", "dy" as pairs of single characters within these strings just 
isn't getting what Voynichese is all about - IMO, it's missing the point 
about how Voynichese is structured, how it works.

  <f78r.X.1;A>       ok-Ch-d-l-d-l-o =  // Neal key for first paragraph
  <f78r.X.2;A>       ok-Ch-dy =
  <f78r.X.3;A>       d-ar  al-oCFh-y =
  <f78r.X.4;A>       d-Che-d-al-y =
  <f78r.X.5;A>       ot-ar-o  d-l-Ar-y  -  or-ar-y =     // Neal key for 
second paragraph
  <f78r.X.6;A>       ok-ar-al =

If these are typical of Feely's "clews", and are furthermore typical of 
that upon which you have constructed your own theory, then I truly wonder 
if you have been building on sand here.

Having said that, I've also read that Strong derived his ideas more from 
O'Neill's suggested labels (rather than from Feely's), and so wonder 
whether anything remains of Feely in his or your work?

> Again, you miss the periodicity problem, though I know we've had private
> discussions on this and I've made posts to the list on the subject.  the
> Voynich is filled with two and three letter groups, but guess what?  The
> 'half-spaces' relate to these groups.  Strong thought the word was the unit,
> that the system changed after every word, and tried to conform.  With more
> pages at hand, I know that the word is the *longest* period, but many words
> are split into groups of two or three, so that pairs may be taken from the
> tables.  Look at the string itself - 1-3-5-7-9-7-5-3-1 are all spaced by
> two, while the string section 1-4-7-4-1 is spaced by three.  A little
> variation in a glyph, and you have an instant offset of two or three.

As a decoder/interpreter, note that also gives you the ability to "see" 
half-spaces when it looks like a word isn't "fitting" the apparent pattern, 
and to "overlook" them when it is.

> Go back to the archives again, and you'll see me mentioning that the choice
> of alphabets tends to go along with the 'numbers of things' on the page.

You mentioned "numbers of things" here as some kind of mnemonic device (of 
which there were many)...
        http://www.voynich.net/Arch/2004/05/msg00227.html
...but I don't recall you ever connecting such numbers with the coding 
system before, FWIW?

> This is where my question of the importance of 'color' came in, at a time
> when we had no such information available.  Now I know that the color does
> not apply, but the 'number' relation remains the same.  We know now that
> both the drawings and text were done in most instances before color, which
> doesn't take a thing away from my previous observations, but does answer the
> question of color quite nicely.

I think that the "heavy painter" colour is almost certainly junk (ie zero 
semantic content), but there are still coloured parts which seem to be 
original which we should handle with care.

> Strong didn't believe your representation, and that's why he called it a
> 'peculiar' use of a polyalphabetic.

Are you referring here to my conviction that many common pairs (like "qo", 
"dy", "ol" "or", etc) actually verbosely code for single letters in the 
plaintext (or in the preceding cipher stage)?

>   I've certainly provided enough
> information over the 3 years especially, that could or should have made one
> think for a moment, and I've even highlighted the main points here once
> again.  I've said that nobody looks for key lengths of 2 or 3, but that's
> primarily what you get by using his 12-long sequence, even if there weren't
> other manipulations going on simultaneously.

I think that Strong was a clever guy, and that you've proved yourself (by 
rationalising and tidying up his approach) to be even cleverer. But if you 
need interpretable half-spaces & key lengths of 2 or 3 to make it all work, 
then I suspect that you've ended up with a system that has more than enough 
capacity to see just what you want to see (just as Strong saw "paprika")

> You're obviously applying this statement to my own work, as well as to other
> ideas put forth.  If you remember back a couple of years ago, I made the
> private offer to give you all the information I had, so you could either
> prove or disprove Strong's (and my) approach, but you didn't take the offer,
> and now I find you trashing something you haven't taken the time to
> understand, something evidenced by your misconceptions that I'm responding
> to in this very post.  I don't need to quote any other idea than this to
> know that you have in this case at least, demonstrated "laissez err", not
> "laissez faire". :-o

What I'm trying to get you to see is that there is indeed a brilliant 
cryptographic heart beating inside the VMs, but that the brilliant Doctor 
Strong had his finger on his *own* pulse - and that you're now doing the 
same. There are places where we can catch a fleeting glimpse (as did Feely 
and O'Neill) of something resembling simple cipher, but these are simply 
hallucinations.

AFA analysing Strong's (and your) results, I honestly don't know what kind 
of counter-critique you would find persuasive. My counter-argument simply 
amounts to pointing out that Voynichese doesn't work that way, and that if 
Strong's magic key was based on Feely and O'Neill's cribs, then it was 
built on sand. If you require jumping between lots of tables to remap a 
highly structured ciphertext onto a differently structured plaintext, 
perhaps (like Gordon Rugg) you're implicitly embedding your desire to 
comprehend the text into those tables - and so are reading yourself instead.

> My respect for you is immense, you're a brilliant man and a good person to
> talk to off-list.  I even consider you a friend, the number of which I can
> count on 1 1/2 hands.

Ditto, ditto, ditto, ditto, ditto. :-)

Hmmm... perhaps "daiin" means "ditto"?  It even *looks* like a simple 
substitution cipher... and yes, it originally came from the Tuscan dialect 
(according to dictionary.com, anyway). Another one for Dennis' Just For Fun 
page, perhaps... or perhaps not! :-)

Cheers, .....Nick Pelling..... 


______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list