Re: Indic loans in Euir Twas (was:Re: [CONLANG] Attention Roger! Sanskrit loans in Javanese etc.)

Alex Fink <000024@...> wrote:
 > On Tue, 25 Nov 2008 11:40:11 +0100, Benct Philip
 > Jonsson <bpj@...> wrote:
 >
 >> Thing is I've decided that the traditional Indic-
 >> derived writing system (Lipiir) of my conlang
 >> Euir Twas has a set of rather original
 >> correspondences between Indic letters and
 >> Lipiir sound values:
 >
 > I like. The Welsh effect on the Indic vowels is
 > particularly nice.

Heh! The original idea came from the idea of an
orthography for a language with seven vowels
written <a e i o u y w> and four high semivowels
/j H M\ w/ where these semivowels are written
with the same letters as their homorganic vowels
/i y M u/. The Welshness of the mappings kind of
gave them selves, and gave rise to the idea of
a Welsh methodist missionary designing the Roman
transcription of the language.  Then to explain
the un-Welsh and un-English features of the
transcription I invented a convert who was the
son of a traditional Tuas bard (_kawy_ < Skt /k@vI/)
whose craft demanded more than
average phonolgical insight sinsce traditional Tuas
poetry (_kaui_ < Skt /kAvj@/) utilized alliteration,
assonance and rime all at the same time.  Naturally
the _kawy_ were also the traditional scribes!

 > I had to draw a table of all the consonants to
 > see what was going on, though. Were the
 > consonants not in your list meant to be not used
 > in Lipiir, or used with their canonical values?

Used with their expected closest correspondants
rather.  I should probably have listed /t`/ > /t/ too! :-)

 >> [s`] > [x] is well documented in various
 >> languages. Indoaryan itself has cases of
 >> | /s`/ > /k_h/ e.g. /s\Is`j@/ > Panjabi /sIk_h/
 >
 > What about [s\]? I'd be surprised if [s\]
 > stayed in place while [s`] jumped over it to
 > become [x]. IIRC they merge in modern Indic
 > (nearly?) everywhere, but you didn't mention
 > the fate of [s\].

Indic /s\/ corresponds to Euia /s\/.  What happened
in the Prakrits was that all three Vedic sibilants
merged as either /s/ or /S/ depending on language.
Generally the Western Prakrits had /s/ and the Eastern
had /S/.  When later Pandits made an effort to
restore the threefold distinction in Sanskrit they
usually ended up using [s] for /s/ but merged /s\/
and /s`/ as [S].  I don't yet know if /s`/ > /x/ is
a sound change which happened in pre-Euia (as it
did e.g. in proto-Slavic) or if those who adapted
Indic script to pre-Euia were faced with several
Indic letters pronounced in the vicinity of their
/x/ and /X_w/ and made their pick.  Probably it was
a bit of both, i.e. pre-Euia had /X_w x s`/ with
facultative merger of the last two, so they made their
pick and assigned Indic /k_h/ to their /X_w/, the Indic
letter which was sometimes /k_h/ and sometimes /kS/ to
their /x/ and the letter which was sometimes /k_h/ and
sometimes /S/ to their /s`/, with the Indic <kS> letter
later falling out of use when the Euia merger of /x/ and
/s`/ was complete.

 >> The natural follow-up was that Euir has an old
 >> layer of Indic loans with these correspondences
 >> and a younger layer with correspondences
 >> derived from the areally dominant languages
 >> Javanese and Malay.
 >
 > So were your letter-values meant to be
 > indicative of sound changes in pre-modern Euir?

Partly.

 > How much of it is just adapting the Indic sounds
 > to the closest native match?

Mostly that.  The application of the more strident
voiceless aspirates to their fricatives and the
smoother 'voiced aspirates' to their voiceless aspirates
probably took a degree of contrastive analysis.
Possibly Indians heard both Euia voiceless fricatives
and Euia voiceless aspirates as voiceless aspirates
while Euia speakers heard both voiced and voiceless
Indian aspirates as voiceless, and then someone had
a bright idea.

 > wedging in Euir
 > sounds with no particularly good Indic
 > correspondent wherever they may fit?

A little bit of that.  However Euia /r\/ _z_
probably was /T/ at the time, later changing
first to /D/ and then to /r\/ in some positions
and merged with /s/ in others.  Unincidentally
/r\/ and /M\/ are the only Euia consonants
which can't occur word initially or after another
consonant. (The Roman spelling _ny_ for /N/ creates
apparent cases of post-consonantal /M\/ however! :-)

 > (The /k_h/
 > letter with value [X_w] seemed especially like
 > an instance of the latter, and a strange one at
 > that -- /k_h/ is a logical spot for [x], and /h/
 > which you've given [x] seems a reasonable place
 > for [X_w], being extrasystematic.)

Probably /X_w/ was originally spelled with an Indic
<k_hv> conjunct character, but the <v> part was later
dropped since there was, from the Euia POV a surplus
of guttural fricative letters anyway.

Note: My realizing the full consequences of inter-word
sandhi and the phonological shallowness of Euia
spelling has caused the lang to change name from Euir
to Euia.  In word final position voiced stops change
to voiceless, aspirates to preaspirated, /l/ to /n/,
/r\/ to /s/, and /r\`/ to the otherwise non-existing
approximant [6_^`], which in spite of being an allophone
of /r\`/ is spelled _a_ in Roman.  Dictionaries and
grammars disambiguate by putting acute accents on syllabic
<i u y w a> when next to another vowel or a semivowel,
as well as to final syllabic <i y w> after <n>, since
/J N N\_w/ are spelled <ni ny nw>.

/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)