Re: Costanice Phonology Sketch

It would appear that my Very Short History was so short that it caused
a lot of confusion. Let me clarify: The idea is that Koine, after
wiping out previous dialects and establishing itself as the lingua
franca across the Eastern Roman Empire, began to split up into its own
regional dialects after a short while, much like Vulgar Latin *here*.
The dialects in mainland Greece took more or less the same development
as Greek here, while those in Asia Minor went in a completely
different direction that no real-world dialects took. The fall of
Constantinople in 1453 was the event that sent the refugees packing
off to Spain, but by that time they already spoke a language that was
quite different from *here*s Medieval Greek, and already similar to
Spanish in several important ways. In fact, I know very little about
Byzantine Greek and only a little more about Modern Greek, but my
starting point doesn't lie there--it lies in the Koine of the New
Testament and Church Fathers.

In contemporary IB, Greece speaks a language more or less the same as
*here*s Greek, but with pockets of Costanice-like dialect. The
Costanico enclave in Spain has been assimilated, leaving Nea Illenicia
as the main outpost of the formerly Asian dialects.

Some aspects of Costanice development before the relocation to Spain
may seem unlikely because they too-coincidentally resemble Spanish
changes, but that's okay with me.

To give a clearer idea of how Costanice evolved, here are the major
changes (leaving out a lot of detail):

BY 500 AD
 * Early vowel changes: ai > e, oi > y, ei > e:,
 * Voiced stops spirantized
 * Probably also y > i
 * (*Here* we must add E: > i, but see discussion below)

BY 1000 AD
 * Tonal accent gives way to stress accent (unsure on timing, but must
precede the next change)
 * Syncope of unstressed syllables where syllable structure permits
 * In unstressed syllables, intervocalic voiced spirants lost and
intervocalic voiceless stops voiced
 * Final stress shifts to penult stress except in verb conjugation

BY 1453 AD
 * Palatalization of /k G x/ to [tS Z S] before front vowels
 * e, o > i, u in pre-tonic unstressed syllables; conversely i, u > e,
o in post-tonic unstressed syllables
 * Loss of long vowels in stressed syllables: e: > i, E: > ie, o > uo
(and later uo > ue)
 * Monophtongization of eu, au > o in closed syllables

TO PRESENT DAY
These changes mimic changes in Spanish
 * Voice contrast lost on coronal spirants, so z, Z > s, S
 * S > x
 * Voiced stops spirantized intervocalically
 * lj > j` (orthographic |ll|)

Specific responses:

>> and set up a
>> community in Barcelona. From this point on the language was heavily
>> influence by Spanish.
>
>But isn't this a _Catalonian_ speaking area?
Er, good point. Have to change that.

> > Pronunciation is basically as in Spanish. Voiced stops are spirantized
> > between vowels, /r/ is a trill, etc. /v/ is marginal--it only occurs
> > intervocalically, and for most speakers is [B], i.e. identical to
> > intervocalic /b/.
>
> How do the /b/, /d/, /g/ series fit in? What is their origin? From the
> example of _zruebo_ below it, it suggests they developed from the voicing
> earlier /p/, /t/ and /k/ in certain environments. I notice /v/ is there,
> but what has happened to Byzantine /D/ and /G/ (from ancient delta & gamma)
> ? And what has happened to /z/ which has been part of the Greek phonemic
> inventory for more two thousand years?
In environments were voiceless stops were voiced, the voiced spirants
were lost. /z/ was merged with /s/, as in Spanish.

> > /k/ and /g/ before front vowels become [tS] and [x] respectively.
>
> I understand the palatalization of /k/ before front vowels; it occurs in
> some modern dialects *here*. But why does /g/ apparently become a
> fricative instead of being palatalized? What happens *here* in those
> dialects that do palatalize is AFAIK:
> /k/ --> [tS]
> /x/ --> [S]
> /G/ --> [j]
> (The last is common to all dialects)
This is more or less the state of affairs in Costanice before the move
to Spain; after that [Z] (palatalized /G/) merges with /S/, and both
of them shift to /x/. So now /g/ alternates with [x], while /x/
doesn't alternate at all.

> > Unlike in Spanish, this actually creates alternations within a
> > paradigm: /igo/ > [iGo], /ige/ > [ixe]. (That's "house" in the
> > nominative and dative, respectively.)
>
> So, Constanice has revived the ancient _oikos_ for "house" which *here*
Retained, not revived. Although my Greek dictionary does still contain
_oikos_, though it may not be the most usual word.

> Sorry - I'm puzzled. Are you saying that in IB the ancient distinction
> between long and short vowels , which had disappeared *here* at least by
> the 4th cent CE, actually remained in Byzantine Greek till the 15th
> century? *Here* also |ei| had become a _monophthong_ before the 5th cen
> BCE, being, as far as we can tell, simple [e:] in Classical (Attic) Greek
Yes, the long vowel distinction does need to survive longer in
pre-Costanice, something that I consider plausible. It doesn't survive
to the 15th century, but it does survive a few centuries more... the
chart I have above says "by 1000 AD", which should be plausible.

> (where eta was [E:]) before giving way sometime between the 4th & 3rd
> cents BCE to [i:].
Now *this* is new information. I knew that /E:/ eventually turned into
/i/, but I believed this to be a later change. This requires me to
revise. Let me think... if eta was /i:/ by the time of Koine, it will
need to remain so. But i: > ie is still a plausible change, so that
can remain, and all /i/'s, short and long, were lowered to /e/ in
post-tonic syllables. So the only real change we need to make is to
say that pre-tonic eta becomes /i/, not /e/. Which is fine with
me--it's not all that common anyway. This can actually be explained,
then, as a chain shift. Before the change, the long vowel system in
Middle Costanice would be /i: e: o:/. /e:/ began to shift to /i:/,
causing /i:/ to become /ie/. /o:/ then became /uo/ by analogy.

So now we have:
    PRE  TONIC  POST
/i:/ i    ie     e   (eta)
/e:/ e    i      e   (epsilon-iota)
/e/  i    e      e   (epsilon)
/o/  u    o      o   (omicron)
/o:/ o    ue     o   (omega)

The text will need to be updated, but not by too much.

> _zreubo_ is presumbly from Byzantine (and modern) /'anTropos/ - but why
> the shift in stress from the initial syllable? The diphthongization of
> stressed _Vulgar Latin_ /O/ had happened quite a few centuries before the
> 15th.
I discuss this together with the question of stress below.

> > But some words add a different, lexically determined consonant, such
> > as _huesga(r)_ which adds an /r/:
>
> Why /r/?
Etymology. Huesga(r) < ho:s gar.

> > Other forms drop the vowel. The 2pl verbal conjugation is among these:
> >  poyide tudo   you (pl) do this
> >  poyíd arte?   you (pl) do what?
> > Note that the stress remains on the same syllable, so an accent mark
> > has to be written in the forms lacking final /e/.
>
> I see the ancient _touto_ has survived - but what is origin of _arte_?
arte < a:ra ti.

> I may be mistaken, but it does seem to me that Constanice is basically
> sound changes that happened to Vulgar Latin in Castilian Spanish to
> ancient (not Byzantine) Greek pronounced (largely) in the Erasmian manner
> (which would IMO give a very attractive result).
This is exactly what I set out to do. My starting point was Koine
Greek, not ancient Attic, but otherwise this accurate. With some of
this new information about Koine pronunciation I may have to revise
some things.

> There were Greek colonies in Spain at a very early date - certainly by the
> 6th cent BCE and possibly even earlier. Why not have Greeks moving from
> the coast and setting up an enclave somewhere in the interior of the
> Iberian peninsular (maybe, to flee from the growing power of the
> Carthaginians) who would then have become isolated from developments over
> in the Aegean area? This would account for a more conservative form of
> Greek, preserving ancient forms which disappeared elsewhere. It would also
> allow the changes similar to those that affected Vulgar Latin to affect
> the development of Greek as spoken by these people.
This is an interesting idea, but it causes some problems in
non-linguistic areas. Most importantly, if the Costanicos were in
Spain they would certainly be Catholic, but I very much want them to
be Greek Orthodox. Furthermore, the writing that I have done in IB
regarding the Costanicos almost all presupposes that they fled from
Byzantium. These could be adapted, but it risks upsetting the rule of
QSS (quid scripsit scripsit).

This is food for thought.

> However - one point I feel I must mention. The form _zruebo_ seems to
> derive fom the ghastly Henninian stress accentuation of ancient Greek (a
> system *never* used by Greeks, either ancient or modern). It is due to a
> 17th cent Dutch doctor of medicine, Heinrich Christian Henning (who
> Latinized himself as 'Henninius'), who put forward the remarkable theory
> that the accents printed on Greek texts had nothing to do with ancient
> pronunciation and that ancient Greek was pronounced with the same stress
> rules as Classical Latin.
Perish the thought! This is definitely *not* the motivation for
_zruebo_. The reasons are much saner: the nominative was lost in favor
of the accusative early on, and among the remaining members of the
paradigm of /'anTro:pos/ all would have penult stress except the
accusative singular. Thus, stress was leveled to the penult syllable,
and thereafter the initial syllable was lost. Similar things happened
with other nouns, which is why almost no nouns retain antepenult
stress in Costanice--syncope or analogy has made them all
penult-stressed. (More examples survive in the verbs, where stress was
and is morphologically important.)

Thanks, Ray, for your detailed comments. You have given me much to
think about, although a lot of it needs to be worked out with the IB
crowd. Still, it is much appreciated!

--
JS Bangs
jaspax@gmail.com
http://jaspax.com

"I could buy you a drink
I could tell you all about it
I could tell you why I doubted
And why I still believe."
 - Pedro the Lion