Re: Some Boreanesian Phonological History

Kristian Jensen <kljensen@...> writes:

> Jörg Rhiemeier wrote:
> > Kristian Jensen <kljensen@...> writes:
> >
> > > [...]
> > >
> > > Proto-B had a phonemic system of consonants and vowels like the
> > > following:
> > >
> > >   *p  *t  *c  *k  *q
> > >   *b  *d  *j  *g  *G
> > >   *f  *L  *S  *x  *X
> > >   *B  *l  *y  *Y  *R
> > >   *u  *r  *i  *@  *a
> > >
> > > [annotations snipped]
> >
> > This looks neatly symmetrical, though somewhat unnatural.
>
> Naturalness is not at all necessary for reconstructive purposes. The
> reconstructed system is extremely theoretical and only helps to explain
> the relationships found between vastly divergent languages.
Well, some historical linguists at least try to get a glimpse of the
proto-language actually spoken some time ago, and those tend to reject
a proposal that doesn't look like a plausible natlang, or shows
patterns that would puzzle linguistic typologists if they were found
in a language actually spoken in the world.  Others are less stringent
on "naturalness" of protolanguages, seeing them rather as a "formula"
expressing relationships between languages without claiming to be a
more-or-less exact image of the actual ancestral language.  *Every*
linguist, however, with the possible exception of two or three
hard-line American structuralists, would rightly reject a proposal that
includes an outright impossibility such as a glottal nasal, but that is
not what you did.

>       There isn't a
> lot of material that linguists in the Boreanesian universe could use to
> reconstruct the proto-language. Boreanesian is effectively a language
> isolate, all other related languages became extinct years ago. The last
> speaker of a non-standard Boreanesian language died some time in the
> 1950s.
Which means that they have to rely on internal reconstruction,
a method which always tends to come up with systems that are more
regular than what really happened.  Internal reconstruction may turn
up a number of lost regularities behind an irregular pattern, but it
doesn't tell *when* those regularities got lost.  The result is a
protolanguage which looks as if it was designed on a conlanger's
drawing-board, while in the actual history there always were
irregularities because all those regular patterns never co-existed at
the same time.

> One unusual characterisic you perhaps noticed in the above inventory
> is the complete lack of nasals.
Oh yes, I noticed it!

>       But this in itself is not unusual nor unnatural in
> languages with nasal harmony. All languages in the Boreanesian
> family had nasal harmony. Although I should have of course added a
> series of nasal vowels in the Proto-B's vowel inventory:
>
>    *u~ *r~ *i~ *@~ *a~
>
> Sorry... kinda forgot that. ;)
Ah, now a number of things become clearer.  So there were rules such
as /bu~/ -> [mu], right?

> [...]
>
> > >       Nasality was suprasegmentalized so that the voiced
> > > segments became coresponding nasals or nasalized segments.
> >
> > By which rules?
>
> Sorry, I guess I phrased that in the wrong way. I meant to say that
> nasality was _already_ an autosegmentalized feature in Proto-B. Like
> I said, nasal harmony was certainly a part of Proto-B. In most
> languages, nasality is an integral part of phonemes. However, in a
> few languages like Boreanesian,
...or some languages of the North American Pacific coast...

>       this feature is extracted from the
> segmental tier and placed on the suprasegmental tier so that it
> characterizes entire syllables, entire morphemes, or even entire
> words. Nasal-harmony in Boreanesian is characterized by foot-level
> (or stress-group-level) specifications for [- nasal] or [+ nasal],
> and certain segments (lenis stops, approximants, and vowels) surface
> differently in oral and nasal feet.
I.e. /bu/ -> [bu] but /bu~/ -> [mu] (or something like that), as I
already guessed above.  Do voiceless stops yield voiceless nasals?
And what about continuants?

> > > Nasal harmony was certainly a part of Proto-B. But as the time went by,
> > > nasality was not the only thing that was suprasegmentalized.
> > > Vowel features of rounding and fronting became compatible with
> > > peripheral (labial, dorsal/radical) and laminal/palatal consonants
> > > respectively.
> >
> > While the apicals remained neutral, thus co-occuring with both sets.
> > And if I understand it correctly, *a and *@ evolved front unrounded
> > and back rounded allophones (something like [E]/[O] for /a/ and
> > [e]/[o] for /@/)?  Or am I completely misled here?
>
> No, you're not misled at all. Although in standard Boreanesian, its more
> like [a]/[Q] for *a, and [i]/[u] for *@. Proto-B *a was more than likely
> /A/, being radical rather than palatal. Proto-B *@ merged with *i and *u
> in front and round contexts respectively. Proto-B *r retained its neutral
> status, even after merging with *@.
I.e., it is reflected in modern B. always as [@].  Let me try to
figure out the result of these rules:

Proto-B   Front  Neutral  Round

 *u        [u]     [u]     [u]
 *r        [@]     [@]     [@]
 *i        [i]     [i]     [i]
 *@        [i]     [@]     [u]
 *a        [a]     [A]     [Q]

Just a wild guess, correct me if I'm wrong.  Did I miss something?

> > > [vowel harmony]
> >
> > I.e., they are all either front unrounded or back rounded, or what?
>
> Not all. Some! Remember (see below) some consonants are opaque to rounding
> but are hosts to fronting, while some are opaque to fronting but are hosts
> to rounding. Apicals, being as you said "neutral", are opaque to both.
You mean, transparent to both, as they don't affect vowels.

> > >       However, the vowel harmony
> > > in Boreanesian is complicated by the fact that peripheral consonants
> > > are hosts to roundness but blockers of fronting, while laminal/palatal
> > > consonants are hosts to fronting but blockers of rounding.
> >
> > Now I am completely lost.  Which factors govern the quality of vowels?
> > Are vowels phonemic at all?  For some reason, all this sounds as if
> > the quality of vowels was determined by the adjacent consonants
> > according to some kind of rules, but I might be entirely off the mark.
>
> Vowels are phonemic. But not all are used in different types of
> syllables. Only heavy syllables have the complete inventory.
What is the complete inventory?  I seem to remember /i @ u a/ from
previous posts, but I may be missing something.

>       Light
> syllables have only /i u @/, and some light syllables are even
> vowelless at the underlying level -- having only epenthetic [i u
> @]. It is these epenthetic vowels that are subject to vowel
> harmony. Now as for whether the quality of these vowels is determined
> by adjacent consonants or not is difficult even for me to say. The
> statement for spreading of front or rounding is quite complex and
> requires specifying both the vowels and consonants
> involved. Personally, I go for a more autosegmental or prosodical
> approach.
Yeah, I am also quite fond of autosegmental phonology.  Good for
interesting effects!  I have some weird high/low vowel harmony system
lying around which I concocted shortly after reading about
autosegmental phonology for the first time in a textbook.  In that
system, vowel height follows similar rules as tones do in some
African languages.  The number of vowel height autosegments is not
necessarily the same as the number of syllables in the word, and
certain suffixes don't have their own autosegments.  If the word has
more vowel heights than syllables, two heights are assigned to a
single vowel, which thus becomes a diphthong; if a suffix is added,
the diphthong is monophthongized because the second autosegment now
has a syllable of its own.

> > > All of these developments must have already occured some time before
> > > the Austronesian influx. Austronesian and Japanese loanwords are
> > > subject to the whims of nasal and front/rounding harmony -- at least
> > > in the standard language.
> > >
> > > If it sounds complicated, well... that's because it is.
> >
> > Indeed!  A few examples would help.
>
> Okey dokey! Here are some made-up examples just to keep things simple.
>
>    /su + diw/             > sudiw
>    /su + kay/             > sukway
I.e., /u/ labializes following velar.

>    /pkway/                > pukway
/p/ is labial, hence the epenthetic vowel is [u].

>    /si + pkway/           > sipukway
>    /T@ + pkway/           > T@pukway
What is /T/?  A voiceless dental fricative, or something else?
A laminal stop (or fricative)?

>    /d + pkway/            > dupukway
/d/ is apical and thus neutral, hence the epenthetic vowel
duplicates the vowel of the following syllable.

>    /d + Tway/             > d@Tuway
>    /d + pkway + Tway/     > dupukwa:cuway
>    /bkwi + T + pkway/     > bukwicupukway
I can't see what is going on here, mainly because I don't know what
/T/ is.

> Hope that helps.
A bit at least.  A lot of things have become clearer to me.

Jörg.

Reply