Home > f16r, f33r, f39v, f55v, f57v, f58r, f71r, Repeating Sequences > Repeating Word Sequences

Repeating Word Sequences

There are a few multi word sequences that appear on more than one folio in the manuscript. Looking at those that occur on a single line of text and at least three times on different folios, the distribution is as follows:

(The repeating sequences on f57v do not appear, since they all occur on the same folio. There are 1006 two word sequences that appear on at least three folios.)

There no four or five or greater length sequences – why?

Why do the sequences often end with “am”?

Perhaps these three “word” sequences are dates?

Why are most of the sequences later on in the VMs? The earliest folio found is f16r, then f33r, f39v, f55v, f58r, f71r …

Now, loosening or simplifying the Voyn_101 transcription using the following table (top is the original character, bottom is the replacement):

Again, we infer that phrases of more than 3 words never repeat more than twice in the VMs:

Knox ran some comparison tests, using the EVA transcription, and found quite different results. These are available in detail here.

I am surprised at the difference in the transcriptions.

For the running text only, in EVA without respect for line wrapping.
I doubt any are wrapped but I'll check tomorrow and find the line descriptors.
Trying to match V-101 (in parenthesis). Might have missed one or more.
Only one trigram has "s" here. 

chedy qokeey qokeey   3 ()
chey qol chedy        4 (#12,3)
ol chedy qokain       3 ()
ol s aiin             4 (#3,4)
ol shedy qokedy       5 ()
ol shedy qokeey       3 ()
or aiin cheol         3 ()
or aiin okaiin        3 (#9,3)
or aiin ol            4 (#7,4)
or or aiin            3 (#11,3)
qokedy qokedy qokedy  3 ()
qol chedy qokain      3 ()
shedy qokedy qokeedy  3 ()*
shedy qokedy shedy    3 ()
sheedy qokedy chedy   3 ()

*Two of these:
ol shedy qokedy qokeedy
  1. Rene Zandbergen
    March 11, 2010 at 12:57 pm

    Hello Julian,

    just a small detail (which could be of more general interest):
    when loosening or symplifying the V-101 alphabet, for the 8th character in your
    2-row table (Eva-g) I would propose to take an Eva-y instead of Eva-m.
    Eva-g essentially occurs at line ends and label ends, and replacing it with
    Eva-y almost always results in a ‘valid’ Voynich word.

    • JB
      March 11, 2010 at 4:39 pm

      Hi Rene,

      Many thanks for this useful tip: I’ll change that substitution, and regenerate the tables.


    • JB
      March 12, 2010 at 3:33 am

      Hi again Rene,

      Looking at the EVA characters, EVA-y looks like “9”, whereas I am trying to simplify Eva-g, which looks more like an “8” with a tail. I think I’ll remove this one “simplification” completely.

  2. March 25, 2010 at 10:10 pm

    I’m a little surprised how many of these repeating sequences contain both a freestanding ‘s’ (or sometimes a free-standing ‘r’) and ‘aiin’. If you accept the possibility that ‘aiin’ groups encipher (say) I / II / III / IIII, might this imply that free-standing ‘s’/’r’ enciphers ‘V’?

    So, could it be that these match simply because they encipher Roman numerals? Sorry if this is unusually prosaic for me, I’ll try harder next time. 🙂

    • JB
      March 25, 2010 at 11:13 pm

      Yes, I think so too. The trouble is, I couldn’t formalise a rule based on the frequencies of appearance of those sequences, if they are numbers (and thus probably dates). I went searching for patterns like MCDxxxx (e.g. MCDXXXV for 1435) in the hope that if they were dates then there would be a lot of “MCD” “iiiN” type sequences. But I became confuddled and dropped it!

      If these are indeed enciphering Roman numerals, one might expect there to be fewer occurrences in the Herbal folios than in the rest of the VMs … one doesn’t normally need to use numbers when describing herbs …

      • March 25, 2010 at 11:41 pm

        I really wouldn’t advise pre-judging the likely relative contents of folios in that kind of way. The herbal pages could well contain recipes, times, dates, times, etc: while the astronomical pages might contain mythology, who knows?

        Worry about all that once we’ve deciphered it! 🙂

      • March 25, 2010 at 11:44 pm

        Also, I get a certain deceptive buzz off the whole thing that makes me suspect that many Roman numerals may get transposed around just to be awkward to would-be decipherers. Don’t be too trusting! 🙂

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: