Voynichese Phonotactics

[Emma May Smith]

Sam, first off, I'm happy we're having this conversation. I think it must be the most sustained discussion of Voynich phonology in a long time.

It's possible, but there are also words ending in es and ees so interpreting y as a consonant in this context also makes sense (assuming we regard s as a consonant).

Also there are words ending in oy, which are admittedly very rare but this would also point to y functioning as a consonant (in parallel with common o-endings ol and or).

Where do words ending eos and eeos fit in with this? My own hypothesis is that y disappears in many internal positions and transforms into a in others. There are plenty of cases in internal y, but maybe fewer than we would expect.

Thus:
ey is paired to eo
es is paired to eos
edy is paired to eody
al is paired to ol
ar is paired to or
chd is paired to chod
chk is paired to chok
oy is paired to oa

...and so on through various permutations. If we agree that o is a vowel, then by considering y as a vowel with three different ways of being expressed we can achieve a fairly good match. These two vowels don't occur at the same frequencies in every case, but they do appear to be functionally equivalent, fulfilling the same kinds of role in the text.

A word like lkaiin would seem to parallel ykaiin, would it not?  And again there's the fact that y and l seem to be related generally.

I would assume ykaiin parallels okaiin.

Some stats:
okaiin (212 tokens) is paired with ykaiin (45 tokens)
otaiin (154 tokens) is paired with ytaiin (43 tokens)
olkaiin (31 tokens) is paired with alkaiin (5 tokens)
oky (102 tokens) is paired with yky (18 tokens)
oty (115 tokens) is paired with yty (24 tokens)
olky (22 tokens) is paired with alky (3 tokens)

In each case the o form is more common than the y form.