[Spellyans] Normalization of words in -ak and -ek

[Michael Everson]

On 23 May 2009, at 20:41, nicholas williams wrote:

> perhennek is attested, perhenogyon isn't. I should prefer perhednek,  
> perhenegyon.

And according to our principles, that would be right.

> Notice incidentally that marregyon occurs 5 times
> marrogyon 4 times.
> There is an argument in favour of marhek, marhegyon

In this case the stronger of the arguments would be the preference in  
the Revival of marhogyon. (It's attested and in use, so no value in  
changing.)

> Lhuyd give skouarnak s.v. Lepus AB: 78b
> but he also gives Skouarnog bîan 'leveret' s.v. Lepusculus AB: 78b.

Well that doesn't speak to the question of the vowel in the derived  
form.

> You don't mention lagasek but that is lagajak x 2 in BK and  
> ydnlagadzhak in Lhuyd.
> lagasek is attested twice in BM.

Ditto. Your dictionary gives:

Scotch argus butterfly - lagasek Alban, lagasegyon Alban
Cyclops - unlagasek, unlagasogyon
starer - lagasek, lagasogyon
gazer - lagasek, lagasogyon
starer - lagasek, lagasogyon; lagater, lagatoryon
wary person - lagasek, lagasogyon

>> cronek 'toad', cronogow -- recte cronak (BK 3256)
>

> This occurs several times in AB, always as kranag, e.g. 9b, 11c,  
> 33a, 45a, 136b, etc.

So cronak, cronogow should be OK.

Michael Everson * http://www.evertype.com/