[YG Conlang Archives] > [romconlang group] > messages [Date Index] [Thread Index] >


[Date Prev] [Date Next] [Thread Prev] [Thread Next]

An orthographic monster



# An orthographic monster

Last night I suddenly realized what an orthographic
monster Rhodrese _gl'ighlisdg_ /LI'gliS/ 'the churches'
is: two _l_'s, two _i_'s and two digraphs and one
trigraph involving _g_! In a way you might call ghl a
trigraph too since it uses _gh_ before _l_ to
distinguish /gl/ from /L/, which latter is written with
the digraph _gl_.  The funny thing is that practically
all Rhodrese consonant spelling conventions build on
Romance natlang -- in particular Italian -- models as
detailed below. Yet these conventions conspire to
produce something which looks quite unnaturalistic.

Do others have such cases of orthographies
where each spelling convention by itself is
more 'reasonable' but a lining up of several
of them in the same word creates an
'monstrosity conspiracy'?

BTW Old Rhodrese scribes handled final devoicing and other
things differently and would have written our
problematic word _gliclix_, requiring more etymological
and/or morphophonemic analysis of the reader, but
that's pretty monstrous too!

# Overview of Rhodrese consonant graphies

(Leaving aside unproblematic things like _b_ = /b/!)

Those taken directly from Romance natlang conventions
are:

* _ch, gh_ for /k, g/ before _e, i, y_ directly from
  Italian (and Rumantsch).
* _gn_ for /J/ directly from Italian and French.
* _gl_ for /L/ directly from Rumantsch and similar
  to Italian.
* _sc_ for /S/ before _e, i, y_ is taken directly
  from Italian.
* _g_ for /dZ/ before _e, i, y_ is taken directly
  from Italian.
* _c_ for /s/ or /ts/ before _e, i, y_ agrees with
  several Romance natlangs.
* _ç_ for /s/~/ts/ agrees with French, Portuguese,
  Catalan and Old Spanish, though word-final _ç_
  seems to be a Rhodrese specialty.
* _tx_ for /tS/ is taken from Catalan.
* _j_ for /dZ/ agrees with the historical stages of
  several Romance natlangs, though most/all of the
  modern ones have changed /dZ/ > /Z/.  Uncontroversial
  I think.
* _s_ between vowels for /z/ and _ss_ between vowels
  for /s/ agrees with French.
* _z_ for /z/~/dz/ agrees with (Old) French, Portuguese
  and Old Spanish, and for /dz/ Italian.

Those derived by analogy from Romance natlang
conventions are:

* _ghl_ /gl/ and _ghn_ /gn/ build on the Italian use
  of _gh_ to denote absence of palatalization.
* _sç_ for /S/ is a pretty straightforward analogy
  on the above uses of _sc_ and _ç_.
* _dg_ for /dZ/ is a pretty straight variation on
  _tg_ for /dZ/ on Catalan and for /ts\/ in Rumantsch.
  I guess _tg_ was a possible spelling of /tS/ in older
  Rhodrese, though the truth is I think it looks like
  a mismatch.  Also cf. English _dg_ = /dZ/ in the
  Romance loan _judg-_.
* _sg_ for /Z/ before _e, i, y_ is a pretty decent
  analogy on the Italian-derived use of _sc_.
* _sdg_ for /Z/ is a not quite happy second order
  analogy on the pattern of _c ~ ç, sc ~ sç, g ~ dg,
  sc ~ sg_, but I don't have any better alternative
  unless I use things like s-cedilla or g-cedilla
  which lack historical naturalism for a western
  Romance language.  Luckily this spelling is rather
  infrequently needed, but it occurs in some other
  words like _basdg_ BASIU 'kiss' and even word
  internally at least in one word: _visdgáu_
  VISIONE 'sight'.

* _r_ for /4/ and _rr_ for /R/ < /r/ has decent
  Romance natlang analogies.

There is one consonant spelling which lacks a natlang
precedent: _ll_ /r`/, but (1) no Romance natlang has
a comparable system of five laterals/rhotics
/l L r` 4 R/, (2) /r`/ derives from Latin /l:/ and
(3) Spanish and Italian use _ll_ for the sounds
they derived from Latin /l:/ (/L/ and /l:/
respectively).

# A possible 'solution'

Perhaps historical /gl/ which didn't become
/L/ instead became /g4/, giving _?gl'igridsg_.

This would give rise to other monstrosities, however:
what of Latin initial GL-?  Did it become /L/ or /gr/?
And what, in the latter case happens to GLORIA?  Does
it become _?graugle_ with a further dissimilation of
*/r_j/ after the newly arisen /4/, rather than
_?greure_ or _gleur_ /L-/ or _?ghleure_?  Or
perhaps _?greuye_ with */r_j/ > /j/?

Perhaps ECCLESIA suffered aphaeresis like in
Italian, giving the slightly less monstrous
_li clisdg_.

In either case there will still be recent loans like
_hieroghlyf_ and the *name* _Ghlórie_ which will
probably need _ghl_, since it feels unnaturalistic to
suppose that _gl_ was /L/ in the Rhodrese pronunciation
of Latin.

## Further problems with /gr/.

And there is still _sacculus_, which by the current GMP
becomes _saghle_, though it may likely be replaced by
_sachel_ < SACCELLU or _sachet/sacot_ (_sagre_ is out,
since that is the first choice for SACER), not to
mention CINGULU, ANGULU, and in particular SINGUL(ARI)U,
which are currently _cenghle, anghle, singhlair_:
the first two may be _?cengre, ?angre_, but _?singrair,
?singriagl_ or even _?singr(i)ay_ are all kind of iffy
(though I'm softening to _?singriagl_!), or should I
perhaps have a special spelling/reading rule for _ngl_?

/BP 8^)>
--
Benct Philip Jonsson -- melroch atte melroch dotte se
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 "C'est en vain que nos Josués littéraires crient
 à la langue de s'arrêter; les langues ni le soleil
 ne s'arrêtent plus. Le jour où elles se *fixent*,
 c'est qu'elles meurent."           (Victor Hugo)