Re: retroflex consonants

Julia "Schnecki" Simon skrev:
> Hello!
>
> On 6/6/05, Rob Haden <magwich78@...> wrote:
>
>>Hi all,
>>
>>I was wondering if anyone here knows how retroflex consonants can arise in
>>a language.  For example, can it arise through glottalization?
>
>
>>From what I remember about Indian languages (in India; not Native
> American languages! ;) ,
The disambiguation trick is to call those of India 'Indic'
and those of America 'Amerindian'

> retroflex consonants can come from other
> context/assimilation phenomena. For example, some of the rules for
> internal sandhi (assimilation/dissimilation within words) in Sanskrit
> transform a dental plosive, nasal, or sibilant into a retroflex one
> when it's preceded by a /k/ or a /u/ -- so I guess that in this
> particular case, the retracted tongue root is to blame.
In Dravidian languages retroflexes seem to go back to
Proto-Dravidian.  In the Indo-Aryan languages they
arose either through sandhi phenomena, through loans
from Dravidian or spontaneously(!).  It is true that Sanskrit
*s after *i/u/r/k (including the semivowels /j w/ |y v|)
becomes /s`/; also final /s\/ is realized as /s`/. Also
/n/ becomes /n`/ if preceded by /r/ and no dental consonant
or non-low vowel intervenes.

In Middle Indo-Aryan (aka Prakrit) many retroflex consonants
arose from Old Indo-Aryan (Sanskrit) /r=/ + dental sequences
which became /@/ + retroflex.

All in all assimilation of dentals to a neighboring /r/ is
the most common source of retroflexes.  It happens in
Swedish and Norwegian where |rt rd rn rs rl| are realized
as retroflexes (except in those dialects that have uvular
/R/), and something like that happened in Old Chinese.
In the history of Tibetan *any* stop + /r/ became a
retroflex stop, so that e.g. what is written |khrang| is
pronounced [t`ha:~]!

--

/BP 8^)>
--
Benct Philip Jonsson -- melroch at melroch dot se

         Solitudinem faciunt pacem appellant!
                                             (Tacitus)

Reply