New Project

I've started a new secondary project, totally unrelated to W.  It
contains consonant mutations and vowel changes a` la Spanish
stem-changing verbs (but more thorough)

Phonology:
Vowels (stressed syllables):
i  u"      u
 e  o"     o
  e` oe    o`
       a
e` = /E/, o` = /O/, oe = rounded /E/, <a> is central.

In unstressed syllables, only six vowels are distinguished:
i          u
  e        o
       @
       a

Where, of course, @ represents the schwa.  An o followed by an e is
indicated by <o.e> to prevent confusion with <oe>.

Final, unstressed syllables, only distinguish three vowels: i, u, a (=
/@/)

Consonants:
mb     nd      n~g    (prenasalized stops), n~ = /N/
mp     nt      n~k    (prenasalized stops)
b      d       g
p      t       k    ' (' = /?/)
m      n    nh n~     (nh = palatal nasal)
v      z zh jh gh     (jh = palatal fricative, gh = velar fricative)
f      s sh ch x      (ch = palatal fricative, as in German _ich_)
          nj          (<nj> = /ndZ/)
          nc          (<nc> = /ntS/)
bv     dz j
pf     ts c
w      r  y
       l  lh      (<lh> = palatal lateral, Spanish <ll>, or Port. <lh>))
wh     rh yh          (voiceless)

Syllables:
(fric)(C)(apr)V(C)(fric) - consonant clusters must be all voiced or all
unvoiced, thus *"sglun" would be illegal, but "sklun" (pronounced with a
voiceless l) would be legal, as would "zglun".  Roots' voicing
predominates, so that a suffix -s becomes -z when added to a word ending
in a voiced consonant.

Due to the way that the phonology evolved, with the exception of a few
loan-words, words can only begin with voiced stops, voiceless stops,
<m>, <n>, <bv>, <dz>, fricatives other than <jh>, <ch> and <x>, and
voiced approximates

Mutations:
There are two mutations, lenition and fortition.
Lenition:
Stops become affricates, exception: <g> and <k> become <gh> and <x>, and
<t> becomes either <ts> or <n> (it is actually a collapse of two earlier
phonemes, hence the variation)
Nasals become prenasalized voiced stops
Affricates also become prenasalized voiced stops.
Fricatives become aproximates (v --> w, z --> r, zh --> y, f --> wh, s
--> rh, sh --> yh), exception: <gh> becomes <n~g> (<gh> was once <ggh>)
Approximates don't change.

Fortition:
Stops become prenasalized (p --> mp, etc.), note: <t> becomes either
<nt> or <n>/<m>)
Nasals don't change
Affricates become nasals
Fricatives become affricates (exception, <gh> becomes <n~>)
Glides become nasals (y --> nh)
<r> and <l> become <dr> and <dl>, while <lh> becomes <nh>.

Vowel alternations:
Vowels are typically given in their stressed form, since the unstressed
form is usually predictable from the stressed form, but rarely the other
direction.

i, e, and e` become e
e` becomes e OR a
oe becomes a
(y)e` becomes i (the <y> may be represented by palatization of previous
consonant, see below)
u", o", u, o`, and o become o
wa becomes u
a becomes (@)

The only possible confusion comes with (y)e` (either e, a, or i - the
(y) cannot be identified as part of the vowel or as part of the stem,
thus ye` might be i or ye or ya) and wa, which may be w + a or the
cluster wa.

Stress is indicated by diacritical marks, graves, diereses, and acutes
(e`, u", etc. can only exist in stressed syllables, therefore they need
no further mark, i, etc. add an acute)

In a dictionary, words are listed in a non-existent form, showing all
the vowels which are subject to change in their stressed form -
inflections cause the position of stress to move around, creating three
forms for nouns and four for verbs.

Inflections:
Nouns are inflected for gender, by means of a prefix (human, animate,
inanimate), and, by suffixes, for case and number (singular, dual,
plural) - case and number are fused.  There are 5 or 13 cases (I created
8 "secondary" cases which I am debating whether to add).  The five
"basic" cases are Nominative I, Nominative II, Dative, Genetive, and
Accusative.  Nominative I & II are stolen from Teonaht's split
nominative.  Nom. I indicates non-volitional, Nom. II indicates
volitional.  Originally, it was an ergative language, which later became
an active language as the ergative was extended.  There was also a
partitative case indicating partially affected objects.  The partitative
took over the patient role entirely, becoming accusative, and so the S
role was split, and eventually the old absolutive (now Nom. I) took a
few transitive verbs.  Several cases are distinguished only by the
position of the stress, for example, Nom. I singular and accusative
plural, or Nom. I plural and dative plural, or genetive singular and
genetive dual.  The eight secondary cases were formed in an earlier
period from the five basic cases and fused postpositions.  They are:
Benefactive (built on dative)
Causative (built on genetive) - indicates "because of", causee is
instrumental
Instrumental (built on nom. I)
Locative (built on nom. I)
Allative (built on dative)
Ablative (built on nom. II)
Perlative (built on accusative)
Exlocative (build on nom. I) - indicates "outside ot"

Verbs split their inflections between the verb itself and an auxillary.
On the verb itself are personal prefixes (combining absolutive and
ergative), and a mood suffix (which also combines formality).  There are
five "politeness levels": plain (close equals and near-equals), high
(exalts subject), highest, low (lowers subject), and lowest.  There are
five moods: affirmative, negative, question, supposition, doubt, and
hope.

The auxillaries indicate aspect (perfect, imperfect, one-time,
habitual).  Prefixes to indicate valency, they are: intransitive,
transitive, ditransitive, motion, and location, all of which have
causative forms as well (causative intransitive is a transitive verb,
causee is in accusative; causative of others place causee in
instrumental).  There are suffixes that indicate tense (past, present,
future) and voice (active, antipassive, passive, reflexive), tense and
voice are fused, again, there are a few that differ only by the position
of the stress - passive present and passive future, and reflexive
present and reflexive future.


I created a protolang (consisting mainly of phonology and inflections)
to ensure realistic mutations.  It was simpler, no mutations at all.
There were 8 vowels and four diphthongs.  The vowels were i, e, e`, a",
a, o`, o, u (a" = front, low, unrounded; a = back, low, unrounded), and
the diphthongs were a"i, ei, au, ou.  Both vowels and diphthongs were
long.  Stress was on the penultimate "mora" (long vowels and diphthongs
counted as 2 morae), thus 16 vowels and 8 diphthongs.  Consonants were
far simpler than the modern form:

b     d     g
p     t     k    '   (' = /?/)
ph    th    kh       (aspirated stops)
m     n     (n)      (n becomes [N] before or after velars)
(nh)  nh    (nh)     (voiceless nasal, assimilates to place)) - source
of the two T's
v     z   zh
f     s   sh
w     r    y
      l    lh

syllables: (C)V(C) - where the last C could not be a glide.

The normal mutation was for voiced stops to become voiced affricates,
unaspirated stops became voiced stops, and aspirated stops became
voiceless stops, the unvoiced nasal became <t>.

Between vowels, voiced stops became prenasalized stops, unaspirated
stops became voiced affricates, aspirated stops became unvoiced
affricates, nasals became prenasalized stops, the unvoiced nasal became
<n>, and fricatives became approximates.

After nasals, voiced stops became nasals, unaspirated stops became
prenasalized voiced stops, aspirated stops became prenasalized voiceless
stops, fricatives became affricates (thru an intermediate step of
prenasalization, e.g., nz --> ndz --> dz), glides became nasals, nr -->
ndr --> dr, nl --> ndl --> dl, ny --> nh, and nlh --> nh

In syllable-final position, stops became fricatives, and nasals were
lost.

When unstressed, short vowels were lowered, thus i --> e, e --> e`
(which later became <e>), e` --> a", u --> o, o --> o` (later <o>), o`
--> a, a and a" became @ (which was frequently lost).  Long vowels were
simply shortened.  Later e` and o` became e and o and a" and a collapsed
into a central vowel.
a"i --> ei --> e: --> e
a":i --> e:i --> e: --> e
au --> ou --> o: --> o
a:u --> o:u --> o: --> o

When in final (unstressed) position, i, e, e`, a"i and their long
varients all eventually became <i>, u, o, o`, au, and their long
varients, became <u>, while a, a", a:, and a": became <a>.

When stresseed, the back short vowels moved forward, thus u --> u", o
--> o", o` --> oe, a --> a", while the front short vowels did not
change, except for a", which later became central.  I: --> i@ --> ia"
--> ya" --> ye`, u: --> u@ --> ua --> wa, the other long vowels became
diphthongs as well, with an off-glide a little higher, and then these
falling diphthongs became rising diphthongs, and the original vowel was
lost, leaving a higher vowel, e.g., e: --> ei --> i.  The short
diphthongs did the same.  The long diphthongs (e:i, etc.) changed
differently, e:i and o:u became e: and o:, then e and o.  a:u and a":i
became o`:u and e`:i, then o` and e`.

--
"It has occured to me more than once that holy boredom is good and
sufficient reason for the invention of free will." - "Lord Leto II"
(Dune Chronicles, by Frank Herbert)
http://members.tripod.com/~Nik_Taylor/X-Files/
ICQ #: 18656696
AOL screen-name: NikTailor