En réponse à Peter Clark <[log in to unmask]>:

>         First of all, Christophe, I didn't get your reply. I only knew
> about it
> because of bnathyuw's reply. Oddly enough, I got your subsequent
> replies, os
> I suspect my mail server is blitzing out again. Fortunately, there's
> Yahoo,
> so this is going to be copy/paste. Sorry for messing up the thread.

No problem!

>         Ok, that's good to know. /4/ could be in allophonic variation
> with /l/ in
> Proto-Enamyn, and once /G/ -> /R/ -> /R\/, the two could merge into
> /r/.

Indeed, I thought that would be rather plausible.

>         Yes, I like it. After I posted the message last night (it was
> late, so that's
> my excuse for some stupid spelling mistakes later on), I realized that
> the
> nasals would not devoice. Your idea for the "original" mutation
> works--as the
> mutation system becomes increasingly grammaticalized, the stops would
> drop
> out, leaving just /B z j\ G/.

Nice then! :))

>         Ah-ha! This is why I value peer review: creative alternatives! I
> hadn't
> considered approximants. To avoid having to discover how the
> approximants
> mutate, is it reasonable to say that they only occur within the
> mutation
> system? (So I guess that they would be considered allophones of the
> voiced
> fricatives.)

Yes indeed. But I thought mutations happened between *phonemes*. Of course, you
can consider that the approximants disappeared by the time the system became
grammaticalised, and thus really earning the name "mutation".

 That way, once the mutation system has been
> grammaticalized, the
> approximants can change thusly:
> w -> ua -> Ma -> a
> r\ -> l
> /j/ would stay, because it is found in Enamyn, and I like G -> 0 for the
> same
> reason you do. :)

Very plausible! I like the system :)) .

>         That way, after all the sound changes occur, Enamyn would have
> some of the
> following changes in its first mutation:
> v -> a
> ts -> l
> k (in some instances) -> j
> r -> 0
>         Oh, I like that! :)

Me too! Especially ts -> l. That's definitely Maggelity! :))

>         Oops, silly typo, that should be m -> b. And yes, it's a very
> strange system,
> one that I can't explain besides waving a magic wand over it. The
> aspirates
> becoming voiced isn't too difficult to imagine, nor the unvoiced
> fricativizing, but the voiced becoming unvoiced is a bit of a stretch.
> As I
> think about it, maybe I should have the voiced stops become nasals. But
> then
> this means that we have a system in which m -> b and b -> m under the
> same
> phonological conditions!

Well, it could be nasalisation. I can imagine a previous nasal voicing the
following consonant, or fricativising it, or nasalising it. I can even imagine
that it would denasalise a following nasal by dissimilation. So it *is*
possible, although maybe a little stretched... You know that I tend to be quite
liberal with linguistic possibilities ;))) .

>         Ok, I was aiming for lenition with system 1. The idea of merging
> two previous
> systems had not occured to me; what do you think would be the best
> explanation?

I'd say that at the time the two systems were fully grammaticalised, it
happened that they were identical for many consonants, and just different for a
few of them. If they happened for similar grammatical reasons (with as only
difference maybe the gender of the noun or something like that), the difference
could have been levelled out by the speakers. I stay rather imprecise because I
don't know the specifics of those mutations. Let's say that one of the mutation
was nasalisation, while the second could be called hardening. If they were
identical for a lot of consonants except a few, they could have merged.

 Say, for instance, a system that deals with stops and a
> system
> that deals with fricatives merge? Or maybe a three-way merger, in which
> the
> stops, nasals, and fricatives all come from different systems? That
> would
> permit b -> m and m -> b, but what a nightmare! :)

No, I meant two mutation systems which deal with all consonants, but are
identical for a lot of consonants and appear in similar contexts. Add a bit of
erosion, people not remembering through time which mutation must be used,
mistakes getting more and more common, and the two systems would end up as one.
With this history, your system could even end up a bit inconsistent, with for
instant b following one of the original systems, but t and g following the

>         That's the only likely explanation I can think of. It isn't very
> pretty,
> though, as I said. Maybe I will make it j\ -> Z -> j. Would make for
> some
> interesting changes, as well as for a lot of glides, since I already
> have g
> -> j!

Nothing's wrong with glides :)) . How knows, maybe you could even make them
evolve into palatalisation! :))

>         It happened because I was getting sleepy. :) It should be z ->
> dz -> ts.

Still, were did this d appear? "hardening" is rare without a reason...

>         Another mistake. p\ -> f.


>         Well, it looks as though the rest of the thread has already been
> discussed,
> so I'll leave it hear. I'm relieved to hear that the chain for G -> r
> is
> plausible. Tell me how system 2 (or its back history) can be improved.

Well, I think I already did. IMHO the history proposed could, if tweaked a
little to fit the data, explain even a very inconsistent mutation system, as
long as its inconsistencies find a good explanation in two earlier regular
mutation systems.


Take your life as a movie: do not let anybody else play the leading role.