Vedic l

[Peter M. Scharf]

Dear Colleagues,

	I'm describing the phonetic features of Sanskrit sounds for the  
appendix of LInguistic Issues in Encoding Sanskrit my colleague  
Malcolm Hyman and I are putting the finishing touches on.  One I'm a  
bit uncertain about it the so-called Vedic l, sometimes called  
Marathi l, which occurs intervocalically in Rgvedic dialect for  
retroflex d.  I'm unsure whether this is an approximant (like l) or a  
flap like .d (that is the Devanagari d with a dot below as in Modern  
Hindi la.dakaa).  Gujarati also has intervocalic flap developing  
from .d.  These flaps are not l's bur r's.  The question is, "Is (or  
was) the Vedic l a retroflex flap or a retroflex lateral approximant?

	Although Allen mentions flap-articulation, his references are not  
probative.

Allen (1953: 73) "they [our authorities] take note, however, of the  
especially brief and light flap-articulation of intervocalic voiced  
retroflex stops. Note 2
[Note 2:SESirIya-SikzA 22. duspfzwaS ceti vijYeyo qa-QayoH svara- 
maDyayoH (in the RP (see 1.110 above) duHspfzwa = Izatspfzwa and is  
used to describe the articulation of the semivowels).  Cf. also VP  
4.146 qa-QO La-LAv ekezAm; RP 1.52.  For a kymographic study of this  
feature in Gujarati see t. N. Dave BSOAS 6.673ff.

I checked the reference:
Dave, T. N. 1931. "Notes on Gujarati Phonology." Bulletin of the  
School of Oriental and African Studies, vol. 6, no. 3, pp. 673-678.   
Stable URL: http://www.jstor.org/stable/607202

Dave says only that
Gujarati M.I. .d > G. tapped r.

It seems to me Allen's references weigh against his conclusion that  
Vedic l is a flap, but are not absolutely conclusive.  It may not be  
that all the ancients identified duHspfzwa with Izatspfzwa and the  
former could indicate tap or flap versus approximant.  Alternatively,  
even if the two terms are identical, they need not distinguish tap  
and flap from approximant.

One bit of evidence that might settle the question concerns whether  
there are cases in Indic of the retroflex lateral approximant  
occurring particularly in the intervocalic environment.  Is the Tamil  
retroflex l exclusively or particularly intervocalic?

Your help would be appreciated.
Peter

*********************************************************
Peter M. Scharf                           (401) 863-2720 office
Department of Classics             (401) 863-2123 dept.
Brown University
PO Box 1856                               (401) 863-7484 fax
Providence, RI 02912                Scharf at brown.edu
http://www.brown.edu/Departments/Classics/people/facultypage.php? 
id=10044
http://sanskritlibrary.org/
*********************************************************