aspiration was: Phaleran: the Webpage.

On Mon, 14 Jun 1999, Joshua Shinavier wrote:

>
> In modern Hindi they really are aspirated affricates: the "ch" in chatri
> (not sure if that's the correct transcription; I just know the Devanagari),
> for instance, sounds something like ch-hatri.  It's really little different
> from the aspiration of stops: in English aspiration means a slight hiss of
> air, in Hindi (and there's probably a special word for this which I don't
> know) it's always a "ha" sound.  E.g. prabhu really sounds like prab-hu.
> Imagine running the -b and the h- together as in "drab house", said very fast.
>
> Josh
>
There has been a lot of argument about aspirated voiced consonants;
Ladefoged has used the term _breathy_ or _murmur_ for sounds like /bh/,
exactly because the phonation type seems very different from a plain
voiced consonants.

The difference between /t/ and /th/ is that in the vowel following
the consonent the onset of the voiced quality of the vowel is delayed
(Ladefoged 1972: fig. 3): the first part of the vowel is pronounced
voiceless. Since in a sequence of voiced consonant and voiceless vowel
there is no onset of the voiced phonation type, the difference between /b/
and /bh/ cannot be explained in the same manner as the difference between
/t/ and /th/, and Ladefoged explains the quality of the sound in /bh/
as opposed to /b/ by the different alignment of the vocal chords in /bh/
compared to /b/. Thus /bh/ is said to have a breathy or murmured phonation
type, while /b/ is voiced. According to Ladefoged (1997: 13), this holds
for most if not all Indo-Aryan languages: he mentions specifically Hindi,
Sindhi, Marathi, Bengali, Assamese, Gujarati, Bihari and Marwari. Lass
(1982: 90) seems to have taken Ladefogeds analysis as received wisdom,
and does not question it at all. (Of course, /b/ is not an affricate,
so this discussion is a bit beside the point).

But on the whole, your argument (and mine, now) has little to do with
the question whether the sound associated with the devanagari sign _ch_
is an affricate or a stop. In Sanskrit, and presumable earlier Prakrits
(although I don't have Masica's _The Indo-Aryan Languages_, so I can't
check) it will have been a stop; as it still is in Nepali.  Daniels and
Bright (1996: 386) still give [c] for _c_, but they are not clear whether
they're's talking about devanagari as used for Sanskrit or devanagari as
used for Hindi. For the Indo-Aryan languages with script of their own,
such as Bengali, he gives the value as [ts] (I really should learn
decent ASCII IPA).

If there is in modern Hindi (which I have not studied) a tendency to
change from /th/ to /taha/ and from /bh/ to /baha/, that would not be
really surprising, but I've never heard of it before. Neither would be
a tendency to affricate centro-palatal stops, and indeed, that seems
to have happened.

Burmese (zan ~ san ~ shan, zaung ~ saung ~ shaung, Ladefoged 1972;12)
shows a contrast between voiced, voiceless and voiceless aspirated
fricatives, and affricates (the signs transcribed by c, ch and j: Richter
1983:89). Mandarin _c_ is also an aspirated affricate; which might seem
strange when contrasted with Mandarin _ch_: the difference is that _c_
is dental, while _ch_ is retroflex. (Liang and Hagenaar, 1982: 12).
In summary, a contrast between aspirated non-aspirated affricates isn't
that strange - for that, try an interlabial stop, whether aspirated or
not ;-).

References

Daniels, Peter T. and William Bright. 1996. _the world's writing systems_,
Oxford.

Ladefoged, Peter. 1992. _Preliminaries to Linguistic Phonetics_, Chicago.
Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt

Lass, Roger. 1982. _Phonology_. Cambridge.

Liang James C.P and Elly Hagenaar. 1982. _Het Pinyin Transkriptiesysteem_,
Leiden.

Richter, Eberhardt. 1982. _Lehrbuch des modernen Burmesisch
(Umgangssprache)_ Leipzig.

Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt