<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<TITLE>RE: Sila[ ]</TITLE>
<META content="MSHTML 5.00.3017.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=847182421-27042003>I
wasn't questioning the cognacy of your Oceanic forms. I was thinking rather of
the suggested extra-Oceanic cognates, which would imply POc *(qa)sol(ei) or some
such, right? </SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN
class=847182421-27042003></SPAN></FONT> </DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=847182421-27042003>Ross
Clark</SPAN></FONT></DIV>
<BLOCKQUOTE
style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<DIV align=left class=OutlookMessageHeader dir=ltr><FONT face=Tahoma
size=2>-----Original Message-----<BR><B>From:</B> John Lynch
[mailto:lynch_j@VANUATU.USP.AC.FJ]<BR><B>Sent:</B> Monday, 28 April 2003 8:46
a.m.<BR><B>To:</B> 'r.clark@auckland.ac.nz'; AUSTRONESIAN LANGUAGES AND
LINGUISTICS<BR><B>Subject:</B> RE: Sila[ ]<BR><BR></DIV></FONT>
<P><FONT size=2>The Fijian/Anejom cognates silaa and na/dej fairly clearly
point to *sila[ ] - possibly *silaqa or *silaRa. It is difficult to see PPn
*sana coming from this (but stranger things have happened).</FONT></P>
<P><FONT size=2>-----Original Message-----</FONT> <BR><FONT size=2>From: Ross
Clark (FOA DALSL) [<A
href="mailto:r.clark@auckland.ac.nz">mailto:r.clark@auckland.ac.nz</A>]
</FONT><BR><FONT size=2>Sent: Saturday, 26 April 2003 6:31 p.m.</FONT>
<BR><FONT size=2>To: AUSTRONESIAN LANGUAGES AND LINGUISTICS</FONT> <BR><FONT
size=2>Subject: RE: Sila[ ]</FONT> </P>
<P><FONT size=2>These all seem formally a bit loose, though not impossible.
Would there be a</FONT> <BR><FONT size=2>place for Proto-Polynesian *sana in
the big picture? In Samoan at least this</FONT> <BR><FONT size=2>now means
"corn" and Job's tears are "sanasana". </FONT></P>
<P><FONT size=2>Ross Clark</FONT> </P>
<P><FONT size=2>> -----Original Message-----</FONT> <BR><FONT size=2>>
From: Laurent Sagart [<A
href="mailto:Laurent.Sagart@ehess.fr">mailto:Laurent.Sagart@ehess.fr</A>]</FONT>
<BR><FONT size=2>> Sent: Wednesday, 16 April 2003 9:55 p.m.</FONT>
<BR><FONT size=2>> To: AUSTRONESIAN LANGUAGES AND LINGUISTICS</FONT>
<BR><FONT size=2>> Subject: Re: Sila[ ]</FONT> <BR><FONT size=2>>
</FONT><BR><FONT size=2>> </FONT><BR><FONT size=2>> Tsuchida, S. (1977)
Some plant names in Formosan languages. CAAAL 7,</FONT> <BR><FONT size=2>>
79-119.) reconstructed "Proto-Hesperonesian" *z@lay 'Coix </FONT><BR><FONT
size=2>> Lachryma-jobi',</FONT> <BR><FONT size=2>> with Formosan
reflexes. Verheijen (1984: 14, 48) identifies </FONT><BR><FONT size=2>>
this item as</FONT> <BR><FONT size=2>> Coix lacrima Jobi var. Ma-Yuen.,
citing further reflexes in </FONT><BR><FONT size=2>> Bisayan, Roti,</FONT>
<BR><FONT size=2>> Tana-Ai, Maranao and Sika.</FONT> <BR><FONT size=2>>
</FONT><BR><FONT size=2>> L. Sagart</FONT> <BR><FONT size=2>>
CRLAO</FONT> <BR><FONT size=2>> 54 Bd Raspail</FONT> <BR><FONT size=2>>
75270 Paris cedex 06</FONT> <BR><FONT size=2>> France</FONT> <BR><FONT
size=2>> Tel: +33 1 49 54 24 18</FONT> <BR><FONT size=2>> Fax: +33 1 49
54 26 71</FONT> <BR><FONT size=2>> ----- Original Message -----</FONT>
<BR><FONT size=2>> From: "Waruno Mahdi"
<mahdi@fhi-berlin.mpg.de></FONT> <BR><FONT size=2>> To: "
AUSTRONESIAN LANGUAGES AND LINGUISTICS" <AN-LANG@anu.edu.au></FONT>
<BR><FONT size=2>> Sent: Wednesday, April 16, 2003 11:29 AM</FONT>
<BR><FONT size=2>> Subject: Re: Sila[ ]</FONT> <BR><FONT size=2>>
</FONT><BR><FONT size=2>> </FONT><BR><FONT size=2>> > > Paul
Geraghty and I are looking for possible cognates of </FONT><BR><FONT
size=2>> Fijian /silaa/</FONT> <BR><FONT size=2>> > > 'Job's
tears, Coix lachryma-jobi', which has also taken </FONT><BR><FONT size=2>>
on the meaning</FONT> <BR><FONT size=2>> ></FONT> <BR><FONT size=2>>
> The formally PMP form, with numerous attested reflexes in
</FONT><BR><FONT size=2>> WMP and CMP</FONT> <BR><FONT size=2>> >
languages of Indonesia, seems to be *qa[n]zel[a]i, but the
</FONT><BR><FONT size=2>> *z calls for</FONT> <BR><FONT size=2>> >
caution, suggesting possible post-PMP horizontal </FONT><BR><FONT size=2>>
propagation. In other</FONT> <BR><FONT size=2>> > words, this is a
post-factum reconstruction on the base of what a</FONT> <BR><FONT size=2>>
> somewhat mixed process of inheriting and borrowing has left
behind,</FONT> <BR><FONT size=2>> > and must not necessarily be an
accurate representation of </FONT><BR><FONT size=2>> the original</FONT>
<BR><FONT size=2>> > form. So, Fiji silaa and Anejom nadej might perhaps
well prove to be</FONT> <BR><FONT size=2>> > ultimately related after
all.</FONT> <BR><FONT size=2>> ></FONT> <BR><FONT size=2>> > The
reflexes I have (Mahdi 1994:471 n. 113) are:</FONT> <BR><FONT size=2>> >
WMP:</FONT> <BR><FONT size=2>> > Malay h@nj@lay, Sundanese hanj@li, Old
Javanese jah@li,</FONT> <BR><FONT size=2>> > Madurese jhAnglE(h),
Balinese jali-jali, Ngaju j@lEy.</FONT> <BR><FONT size=2>> > CMP:</FONT>
<BR><FONT size=2>> > Roti dele, Haruku sale.</FONT> <BR><FONT
size=2>> > (@ = schwa, A = v-rotated, E = epsilon)</FONT> <BR><FONT
size=2>> ></FONT> <BR><FONT size=2>> > That's all I have.</FONT>
<BR><FONT size=2>> ></FONT> <BR><FONT size=2>> > Aloha,
Waruno</FONT> <BR><FONT size=2>> ></FONT> <BR><FONT size=2>>
></FONT> <BR><FONT size=2>> </FONT></P></BLOCKQUOTE></BODY></HTML>