<!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: cognate vocab as a measure of relatedness</TITLE>
<STYLE type=text/css>BLOCKQUOTE {
MARGIN-BOTTOM: 0px; MARGIN-TOP: 0px
}
DL {
MARGIN-BOTTOM: 0px; MARGIN-TOP: 0px
}
UL {
MARGIN-BOTTOM: 0px; MARGIN-TOP: 0px
}
OL {
MARGIN-BOTTOM: 0px; MARGIN-TOP: 0px
}
LI {
MARGIN-BOTTOM: 0px; MARGIN-TOP: 0px
}
</STYLE>
<META content="MSHTML 5.00.3017.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=328325204-12062002>In the
back of Samarin's book Field Linguistics, he gives parallel 100-item lists in
English, French, Spanish and German, which makes it fairly handy for a quick
computation. Using a generous definition of cognacy, I get</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN
class=328325204-12062002></SPAN></FONT> </DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN
class=328325204-12062002> English-German
75%</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN
class=328325204-12062002>
English-French 36%</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN
class=328325204-12062002></SPAN></FONT> </DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=328325204-12062002>Given
that 200-list figures are normally lower, this would be roughly consistent with
the numbers that Andy has quoted.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=328325204-12062002>Note
that the English-French cognates are almost all directly-inherited. The only
ones clearly resulting from ME borrowings are person/personne, grease/graisse,
mountain/montaigne and round/rond.</SPAN></FONT></DIV>
<DIV> </DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=328325204-12062002>Ross
Clark</SPAN></FONT></DIV>
<DIV> </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> Andy Pawley
[mailto:apawley@coombs.anu.edu.au]<BR><B>Sent:</B> Wednesday, 12 June 2002
3:15 p.m.<BR><B>To:</B> AUSTRONESIAN LANGUAGES AND
LINGUISTICS<BR><B>Subject:</B> Re: cognate vocab as a measure of
relatedness<BR><BR></DIV></FONT>
<DIV>Re Adrian Clynes' question:</DIV>
<DIV><BR></DIV>
<BLOCKQUOTE cite type="cite"><BR></BLOCKQUOTE>
<BLOCKQUOTE cite type="cite">1) What percentage of basic vocabulary is
cognate in English and German,<BR>using (say) a 200-item Swadesh list?<BR>2)
Ditto, for English and French?</BLOCKQUOTE>
<DIV><BR><BR></DIV>
<DIV>In his book<I> A Course in Modern Linguistics</I> Hockett (1958) says
contemporary English and German share approx. 59 percent cognates on the 200
meaning list.</DIV>
<DIV><BR></DIV>
<DIV>I have an idea the figure for English and French is around 20-25 percent.
As I recall, figures for most of the Indo-European families are given in a
book on the lexicostatistics of IE languages by Isidore Dyen and Paul Black. I
forget the exact details but I think it was published in the 1970s.</DIV>
<DIV><BR></DIV>
<DIV>Literary Dutch and German I believe have been scored at 84 per cent. That
reinforces David Mead's point about 60 to 90 percent being a grey area re
mutual intelligibility. Actually my experience is that mutual intellibility is
always pretty low if basic vocab. cognation is below 80 percent -- except
where speakers have had extensive exposure to each others' languages and are
therefore at least passively bilingual. But obviously phonology plays a
big part.</DIV>
<DIV><BR></DIV>
<DIV>Andy Pawley</DIV></BLOCKQUOTE></BODY></HTML>