<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2800.1126" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2>
<DIV>
<DIV><FONT face=Arial size=2>Dear info-childes members,</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>An interesting phonotactic feature, which seems to
hold in several languages, is known as OCP (the obligatory contour principle):
the preference for a certain level of contrast between consecutive vowels in a
word. Contour can be described in terms of differences in <FONT
face="Times New Roman" size=3>opening (or tongue height)</FONT>, <FONT
face="Times New Roman" size=3>place (or tongue position)<FONT face=Arial
size=2>, and </FONT></FONT>rounding between such vowels. Harmony, on the other
hand, is the absence of contrast between consecutive vowels.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Since I am interested in using connectionist
networks to model phonotactic preferences for contour or harmony (vis-a-vis
statistical biases that we have identified in a Brazilian Portuguese corpus), I
wonder if any of the members of this list has come across a marked degree of
preference for either harmony or contour in the language that is either produced
by or directly spoken to early language learners. </FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Thanks in advance for your answers.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Orlando Bisacchi Coelho</FONT></DIV>
<DIV><FONT face=Arial size=2>UMC/UNICAMP</FONT></DIV>
<DIV><FONT face=Arial size=2><A
href="mailto:orlandoc@terra.com.br">orlandoc@terra.com.br</A> </FONT></DIV>
<DIV><FONT face=Arial
size=2></FONT> </DIV></DIV></FONT></DIV></BODY></HTML>