A key to understand the VMS

[Emma May Smith]

I pointed out the problem of word structure to to him in 2014 after his first paper. Over two years later he still hasn't adequately responded to it, saying that the writer was free to write as he pleased within his 'concept of language'. I don't think he has an answer because it would spoil his theory.

As I recall, you recently mentioned cheol vs chole vs echol, which reminded me of the eol permutation thing I'd noticed a while back but had forgotten. Voynichese has plenty of similar strongly asymmetrical permutation sets, but I'm pretty sure that eol was one of the most extreme.

It's also interesting that we see all four gallows inserted in the middle of ch, but never anything else in the middle of ch. That's a rule all of its own as well, etc etc.

Indeed, it is easy to point out that so many of the possible permutations that a writer could make either don't exist or occur so few times that they might be errors.

Let's take the word [chedy], which occurs about 501 times (I'm using voynichese.com for the numbers) and should be the centre of a big network of permutations. Let us agree that it has four characters [ch], [e], [d], and [y], which means there are five possible places to insert a new character. Let us further agree that the following 22 characters are part of the Voynich script: [o, y, a, e, ch, sh, k, t, f, p, ckh, cth, cfh, cph, d, s, l, r, m, n, i, q].

Here are the stats for attempting to insert each of these characters into the word [chedy] in the five different positions:

[o]: [ochedy] 8; [choedy] 0; [cheody] 89; [chedoy] 0; [chedyo] 0
[y]: [ychedy] 13; [chyedy] 0; [cheydy] 0; [chedyy] 0; [chedyy] 0
[a]: [achedy] 0; [chaedy] 0; [cheady] 1; [cheday] 0; [chedya] 0
[e]: [echedy] 1; [cheedy] 59; [cheedy] 59; [chedey] 1; [chedye] 0
[ch]: [chchedy] 0; [chchedy] 0; [chechdy] 0; [chedchy] 1; [chedych] 0
[sh]: [shchedy] 0; [chshedy] 0; [cheshdy] 0; [chedshy] 0; [chedysh] 0
[k]: [kchedy] 22; [chkedy] 5; [chekdy] 0; [chedky] 1; [chedyk] 0
[t]: [tchedy] 33; [chtedy] 2; [chetdy] 1; [chedty] 0; [chedyt] 0
[f]: [fchedy] 11; [chfedy] 0; [chefdy] 0; [chedfy] 0; [chedyf] 0
[p]: [pchedy] 34; [chpedy] 0; [chepdy] 0; [chedpy] 0; [chedyp] 0
[ckh]: [ckhchedy] 0; [chckhedy] 11; [checkhdy] 2; [chedckhy] 0; [chedyckh] 0
[cth]: [cthchedy] 0; [chcthedy] 7; [checthdy] 1; [chedcthy] 0; [chedycth] 0
[cfh]: [cfhchedy] 0; [chcfhedy] 0; [checfhdy] 0; [chedcfhy] 0; [chedycfh] 0
[cph]: [cfhchedy] 0; [chcphedy] 3; [checphdy] 0; [chedcphy] 0; [chedycph] 0
[d]: [dchedy] 27; [chdedy] 2; [cheddy] 0; [cheddy] 0; [chedyd] 0
[s]: [schedy] 7: [chsedy] 0; [chesdy] 0; [chedsy] 0; [chedys] 0
[l]: [lchedy] 119; [chledy] 0; [cheldy] 0; [chedly] 0; [chedyl] 1
[r]: [rchedy] 11; [chredy] 0; [cherdy] 0; [chedry] 0 [chedyr] 1
[m]: [mchedy] 0; [chmedy] 0; [chemdy] 0; [chedmy] 0; [chedym] 0
[n]: [nchedy] 0; [chnedy] 0; [chendy] 0; [chedny] 0; [chedyn] 0
[i]: [ichedy] 0; [chiedy] 0; [cheidy] 0; [chediy] 0; [chedyi] 0
[q]: [qchedy] 1; [chqedy] 0; [cheqdy] 0; [chedqy] 0; [chedyq] 0

Now, Torsten only considers words with at least 4 occurrences to be securely valid ("For words occurring less than four times transcription errors become important." Timm, 2016, p2), so we will use this to sort each permutation into one of three classes: does not occur, does not validly occur, and validly occurs.

Here are the stats: of the 110 possible permutations for inserting a character into [chedy] only 16 (14.5%) occur validly, a further 14 (12.7%) occur invalidly, and 80 (72.7%) do not occur at all. Nine of the characters, [a, ch, sh, cfh, cph, m, n, i, q], cannot be inserted anywhere in the word validly. Also, no character can be inserted validly into the slots after [d] or [y] of [chedy].

Remember, [chedy] is the third most common word in the text, so most words will show even fewer permutations!

Almost every theory I have ever comes across—linguistic, cryptological, and even Rugg—account for the very obvious word structure. Torsten's theory is the only one which handwaves away some of the best evidence we have.