Re: Unilang: the Phonology

Right, so now I get to the business of my own idea of a phonological scheme
for the theoretical unilang...

Vowels: /a e i o u/
Semi-vowels: /w j/
Stops: /p t k b d g/
Frics: /f s h/
Nasals: /m n/
Liquids: /l r/

If I know you CONLANG-bunch right, :) :), that's rather "boring". But since
this isn't an artlang, that's hardly a concern of mine.

So how are they defined? I use Chomsky & Halle's good old SPE distinctive
feature system to define the sounds; here goes:

[+syllabic], [+sonorant], [-consonantal], [+voice] = vowels:

/i/: [-back], [+high]
/u/: [+back], [+high]
/a/: [+low]
/e/: [-back], [-high], [-low]
/o/: [+back], [-high], [-low]

The decision to require all vowels to be [+voice] is based on acoustics;
unvoiced vowels would be inaudible to most not familiar with them.
Conversely, I considered leaving voice free because some langs devoice
vowels in certain environments (e.g. Japanese).

Note that rounding is free in this system.

[-syllabic], [+sonorant], [-consonantal], [+high], [+voice]
= semi-vowels/glides/approximants:

/w/: [+labial]
/j/: [+palatal]


[-syllabic], [+consonantal] = consonants:

(A little table will save me time, presenting all the consonants except for
the stops)

    cont  nasal  lat  son  cor  lab  stri

/f/  +     -      -    -   -    +    +
/s/  +     -      -    -   +    -    +
/h/  +     -      -    -   -    -    +
/m/  -     +      -    +   -    +    -
/n/  -     +      -    +   +    -    -
/l/  +     -      +    +   +    -    -
/r/  +     -      -    +  +/-   -    -

Key: cont = continuant, lat = lateral, son = sonorant, cor +/- means it is
irrelevant; = coronal, lab = labial, stri = strident

With no intention of insulting anyone's intelligence, :), I should perhaps
explain the terms above, briefly: continuants are sounds that don't
completely block the air flow (frics, liquids, vowels); coronals are
dentals, alveolars, retroflexes, and palatals (i.e. front-of-tongue-
sticking-to-something); strident sounds are basically just fricatives
(acoustically, "high random noise factor").

Now, the stops. They're slightly more complicated, because what needs to be
outlined is how the "fortis" row is made distinct from the "lenis". First,
I'll define the articulatory features of all the sounds, and each pair:

[-continuant], [-nasal] = stops

/p b/: [+labial]
/t d/: [+coronal], but not palatal
/k g/: palatal, velar or uvular

/p t k/ can be made distinct from /b d g/ in the following manners:

[+/-voice] - e.g. as in French or English
[+/-spread glottis] - (= aspiration) e.g. as in Mandarin
[+/-constricted glottis] - glottalic sounds; I haven't heard of a language
with only two rows of stops, one glottalic and one non-glottalic; perhaps
they don't exist. In that case, this should perhaps not be an option. I
admit to ignorance on the subject of glottalics.


The system detailed here is more flexible than it might seem; if you
consider the many ways in which one can pronounce /r/, for instance; French-
speakers, English-speakers, Spanish-speakers, and Mandarin-speakers, could
all pronounce it exactly as it is in their own language (in the non-
orthographical sense, of course; there's no such thing as a "silent r").

Phonotactics are not detailed here; I reserve another thread for that.

Óskar

Reply