[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[lojban-beginners] Re: More phonology: voiced/unvoiced and fricative/fricative

On 6/23/07, Klaus Schmirler <KSchmir@online.de> wrote:

My main point against lojban is that it is letter-centric and not very
sophisticated in the phonetic department. For instance, I would argue
that each stop is inherently coupled with its release, and that t, ts,
tc, tl or p and pf are one sound each.


That's not a universal law, it varies from language to language.
For example in Polish (according to Wikipedia) there is a contrast
between czysta 'clean (f.)'  vs. trzysta 'three hundred', where
the first has an affricate and the second a stop-fricative sequence.

Lojban does not make that contrast, so it is allowed to pronounce
the stop-fricative sequence as an affricate, even though it is
technically a stop-fricative sequence.

See <http://en.wikipedia.org/wiki/Affricate_consonant> for a better
explanation of the difference.


But to get on topic:

Among the rules for consonant clusters, I find the following:

Voiced/unvoiced pairs are forbidden except with l, m, n, r. Their
distinction being that they can be syllabic, I guess. But so can v.


{v} cannot be syllabic in Lojban. Different languages have different
rules about what consonants can be syllabic. None of l, m, n, r
can be syllabic in Spanish, for example.


Combinations among c, j, s, z are forbidden. They are fricatives, but
so are f, v, x and voiced x, an allophone of r.


The rule really is that combination of sibilants are not allowed.
(xc and cx are not allowed either, although xs an sx are.)


Now there's a whole bunch of xr words:

xrabo Arabic
xrani injure
xriso Christian
xruba buckwheat
xruki turkey
xrula flower
xruti return

They taste best to my German palate when I pronounce them with an
initial ach sound and then just add voicing. And I think one of the
design principles (which admittedly I haven't found, or I'd have cited
it above) was that sounds should be maximally distinct. So does [xR]
count as maximally distinct - if so, I rest my case - or is the uvular
r forbidden in this context?


[xR] is allowed, but I prefer [x4].

mu'o mi'e xorxes