A family of grammars for Voynichese

[Jorge_Stolfi]

Recall that my "OKOKO" partial paradigm says that a Voynichese word is a sequence of "K" elements with "O" elements inserted before, between, and after them, with at most two "O" per slot.  Where the "O" elements are {a} {o} {y}, and the "K" elements are all the others.

Here are the counts of tokens that satisfy this paradigm, condensed to the number of {O} elements:

COUNTING TOKENS BY NUMBER OF 'O'S

  15656.687500 0.48866 O
  13798.312500 0.43066 OO
   1437.625000 0.04487 OOO
   1052.250000 0.03284 -
     89.500000 0.00279 OOOO
      5.437500 0.00017 OOOOO
      0.187500 0.00001 OOOOOO

These counts consider the running paragraph text only, of all sections, from the same transcription file as before (80% U, 20% Z).  The counts are fractional in order to account for uncertain spaces ',' as I explained before.

Thus it seems that, in addition to the limit of two "O" elements per slot, there is also a limit of three "O" elements in total.  I will add this rule to the model.  The exceptions are only 0.3% of all tokens, and may well be cases of two words run together in the transcription.

If we delete the "O" elements and count only "K" elements, we get this:

COUNTING TOKENS BY NUMBER OF 'K'S

  14144.875000 0.44148 KK
   9226.875000 0.28798 KKK
   5183.687500 0.16179 K
   2717.125000 0.08480 KKKK
    401.875000 0.01254 KKKKK
    281.750000 0.00879 -
     68.062500 0.00212 KKKKKK
     13.187500 0.00041 KKKKKKK
      2.437500 0.00008 KKKKKKKK
      0.125000 0.00000 KKKKKKKKK

Here the cutoff is not as clear as for the "O"s, but tokens with more than six "K"s can be considered noise.  Maybe those with six "K"s are noise too. 

I have not yet decided whether I should put that limit on total "K"s as another constraint of the "OKOKO" model.  Anyway the number of "K"s will be limited by the next stage of the model, the "layer" (formerly "crust-mantle-core") model. 

All the best, --stolfi