During the last couple of months I have been playing with various attempts to match Voynich labels with the nearby text. My hope was to understand more of word morphology, searching for words that match the labels but for some minor changes. I haven't done anything smart (e.g. writing some kind of software to automatically explore the search-space of possible word alterations): I just played around with voynichese.com and regular expressions.
For a change, the results are not brilliant, but I thought I'd share them anyway. Maybe discussing with others will provide new ideas on how to proceed.
I have tried the approach on two sets of data:
- quire 13 (the “balneo” / “biological” section)
- the pharma / jars-and-small-plants paragraphs (discussed You are not allowed to view links. Register or Login to view.)
The “pharma” section provided a very limited number of matches.
Things were slightly better for Quire13. The quire is made of ten folios (20 pages) and contains about 130 labels. Labels that match the top 300 more common words in the manuscript have been ignored. A whole folio (both recto and verso) was considered at one time. I looked for matches both in the folio that contains the label and in the whole quire. Since there are 10 folios, in principle a match has a 10% probability of occurring in the same folio as the label.
I found 124 matches (most matching labels generate multiple matches), 23 of which are in the same folio as the label. This is a 18.5% of matches in the right “folio”. Not great, but still almost the double of the “random” 10%.
Here is the complete list of matches:
Code:
LABEL:saino MATCH:_sain_ol_ f75
LABEL:saino MATCH:_saino_ f75
LABEL:darol MATCH:_dar_ol_ f75
LABEL:daldy MATCH:_daldy_ f75
LABEL:qoted MATCH:_qoked_ f75
LABEL:olkchy MATCH:_qolkchy_ f75
LABEL:olol MATCH:_ol_ol_ f75
LABEL:olol MATCH:_qolol_ f75
LABEL:olol MATCH:_qol_ol_ f75
LABEL:olol MATCH:_olol_ f75
LABEL:otol.shedy MATCH:_qotol_shedy_ f77
LABEL:opor MATCH:_opor_ f80
LABEL:otalshedy MATCH:_qokal_shedy_ f80
LABEL:otan MATCH:_qokan_ f80
LABEL:darol MATCH:_dar_ol_ f82
LABEL:okoldy MATCH:_qokoldy_ f82
LABEL:okeeor MATCH:_okeeor_ f82
LABEL:otechdy MATCH:_qokechedy_ f82
LABEL:otedal MATCH:_qokedal_ f82
LABEL:olkol MATCH:_olkol_ f82
LABEL:okedor MATCH:_qotedor_ f82
LABEL:otoldy MATCH:_qokoldy_ f82
LABEL:olsaiin MATCH:_ols_aiin_ f83
Most of the matches occur in f75 and f82. Since the “otol” matches are dubious (I am not sure “olol” really is a label) I will only comment f82 matches.
Of the 8 matched labels in You are not allowed to view links.
Register or
Login to view. and You are not allowed to view links.
Register or
Login to view.:
2 match directly and perfectly (okeeor and olkol)
1 is split into two words (darol / dar ol)
5 required the addition of a q- prefix; 4 of these have the shift k->t or t->k; one case also has the additional insertion of an 'e' otechdy->qokechedy
These results could suggest that q- really is a prefix that can be added to a “root-form”; one problem is that the q- matches typically also require to switch a 't' for a 'k' or vice-versa, and I don't think these data are enough to conclude that the switch is really possible (i.e. otoldy and qokoldy might still be two unrelated words).
You are not allowed to view links.
Register or
Login to view. has the interesting feature of presenting similar labels for similar illustrations: these are the labels near the rainbows.
- otedal (top right)
- otedy (top left) is also similar but not associated with a rainbow
- okedor (the only one I could match with a word on the recto)
- otedol
- oteedy
I doubt these labels can be different spellings for the same word: it seems strange to me that the same word and the same object are repeated several times in the same page. I think it more likely that both the rainbows and the words correspond to different but closely related concepts.
I have also collected five groups of identical or very similar labels (again, only considering quire 13).
See them highlighted on You are not allowed to view links.
Register or
Login to view..
daldy (light blue). This label occurs twice in the same illustration (in one case, it is split into two words). The illustration is fairly complex: ten nymphs stand below ten streams. Each nymph/stream has two labels. I believe that the bottom label refers to the nymph (see the discussion of "otoly" below). Both occurrences of “daldy” are at the top: they could refer to the streams or maybe be adjectives applied to the nymph labels (but of course they can really be anything, this is quite speculative).
<f75v.L1.5;H> dal.dy-
<f75v.L1.15;H> daldy-
darol (green). The two occurrences are identical. The first one is in the same illustration as “daldy” but occurs as the bottom label: it could refer to the nymph. The You are not allowed to view links.
Register or
Login to view. occurrence refers to some kind of amorphous fluid or gas.
<f75v.L1.4;H> darol=
<f82r.L2.1;H> darol=
o[tk]oldy (salmon pink). These two labels differ because of the
k<->t switch, so this match is not very reliable. The fact that the two occurrences are on the two sides of the same folio might add some plausibility to the match. One of the labels refers to a nymph; the second one to a bulbous nebuly-line object.
<f82r.L2.5;H> okoldy=
<f82v.L3.14;H> otoldy=
olkol (dark blue). The first occurrence is part of the label for a large pool containing 16 nymphs. The second label names a nymph; next to this one, there's another nymph (in a tub) who is labeled 'okain'. One could be tempted to consider the identification of the two nymphs "okain.olkol" with the pool "otoin.olkol".
<f81v.X.1;H> otoin.olkol=
<f82v.L3.7;H> okain=
<f82v.L3.8;H> olkol=
o[tk]ol[y|or] (yellow) Two of the labels are perfect matches (otoly) in the “inner side” of the first bifolio. For once, identical labels correspond to very similar images: nymphs wearing bonnets, facing left, at the far right of long rows of nymphs in large pools, with some liquid flowing from above. I think this supports the idea that the bottom labels in the You are not allowed to view links.
Register or
Login to view. pool refer to the nymphs.
'otolor' is the label of the right tube in an illustration that seems to compare two different “states” of a same substance (see also You are not allowed to view links.
Register or
Login to view.). I think this label is likely unrelated or only loosely related with 'otoly'.
'okolo' is again a nymph facing left, but she wears a peculiar diadem. She might or might not be related with 'otoly'.
<f75v.L1.20;H> otoly=
<f77v.L.2;H> otolor=
<f80r.X.4;H> okolo=
<f84r.X.11;H> otoly=