[Corpora-List] Comparing Phonetic Corpora of AmerIndian languages

[Mike Maxwell]

Yuri Tambovtsev wrote:
> The two texts of Guambiano (Part 3 and Part 4) 
 > also show  a greater degree of similarity (3.02%).

FYI, there were at one time two competing orthographies for Guambiano. 
Native Guambiano words lack a voicing distinction on obstruents; both 
voiced and voiceless obstruents exist, but they are allophonic. 
Therefore a phonemic orthography would use either ptk or bdg (I believe 
ptk was chosen for the phonemic orthography, or possibly ptc/qu).

However, Spanish has a phonemic voicing distinction, and it is 
represented orthographically (or over-represented, with b and v standing 
for a voiced bilabial obstruent, and c/qu in complementary distribution 
for the voiceless velar obstruent, not to mention g/gu...).  Many 
Guambianos first learned to read in Spanish, there are many Spanish 
loans in Guambiano.  The result is that many (perhaps most) Guambianos 
are aware of the voiced/ voiceless distinction, even in their native 
vocabulary.  Hence there was considerable pressure from community 
members to represent this distinction in the Guambiano orthography, so 
that a competing orthography was developed with both bdg and ptk.

The SIL linguists working in Guambiano in the 1980s wanted to leave the 
decision between these orthographies to the community.  In fact, they 
published some Guambiano literature in a sort of diglot, with the two 
orthographies on facing pages.  In the late 1980s, another group began 
working with the Guambianos, and pushed hard for a phonemic orthography. 
  As an SIL member at the time, I was asked to hold a workshop in which 
we brought together representatives of the two community groups, taught 
a bit of phonology and orthography design, and recommended that they 
work it out from there.  (My parting comment was something like, "The 
best orthography is an orthography that is widely used.")

The c/qu vs. k difference has also hit a number of the languages in 
Spanish-speaking Latin America; up until the 1980s, governments and 
educators often exerted considerable pressure to use the Spanish-based 
c/qu.  The pressure was later reversed, meaning that many native 
language orthographies underwent a shift around that time.

The preferred representation of nasalization on vowels (at least in 
Spanish-speaking countries, I'm not sure about Brazil), and of tone 
(often simply omitted) also underwent shifts, or at least were 
inconsistent among languages (this might have affected the Tucanoan 
languages).  Both nasalization and tone act as suprasegmentals in many 
the Amazonian languages, so that in orthographies these features are 
often shown on only one syllable, and assumed to spread (generally to 
the right).  Waorani takes this to extremes, so that in a phonemic 
orthography there is no distinction between nasal and non-nasal 
consonants; the vowels are instead marked for nasalization.  (More 
recently, a Spanish-like orthography has made in-roads in Waorani.)

Other fads included the representation of glottal stops or glottalized 
consonants (mostly with an apostrophe, but sometimes with the numeral 7, 
or the glottal stop is simply omitted).

All of this to say that with languages where the orthography is in flux, 
it's necessary to be very careful interpreting the data.

 > The Chibchan family (Cofan, Guambiano) is  much less similar – 43.90%.

I doubt that Cofan is really Chibchan; it's more likely an isolate. 
It's also doubtful whether Guambiano is Chibchan; it's been assigned to 
the Barbacoan family.  I'm not sure that even that is generally 
accepted, but in any case there's no good evidence (AFAIK) that the 
Barbacoan group is "in" Chibchan.
-- 
    Mike Maxwell
    What good is a universe without somebody around to look at it?
    --Robert Dicke, Princeton physicist

_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora