<html><body>
<p><tt>>> Being rare and distinctive would be an advantage to preserving *W and *R<br>
>> and keeping them from merging sporadically with other sets. If, on the<br>
>> other hand, they were pre-nasalized stops, then what would consistently<br>
>> distinguish them in their reflex pattern from all the epenthetically<br>
>> pre-nasalized interior stops following nasal vowels? If the Dhegihan<br>
>> positional *aWa' was actualized as *ampa' or *amba', why wouldn't the<br>
>> interior consonant in 'day', *aN'pa or *aN'ba, have developed as *a'Wa ?<br>
><br>
> I wondered if this case was bothering you, since you mentioned it earlier.<br>
> This is pretty much the major case of "word internal" *W. Of course, it<br>
> is word internal mainly by the grace of how we now analyze and write the<br>
> elements involved. My take on this is that akha and ama < *akha and *aWa<br>
> are the result of lumping two elements following a noun into one:<br>
><br>
> nu=akha 'the man' < *pro-a=kha<br>
> nu=ama 'the men' < *pro-a=Wa<br>
[...]</tt><br>
<br>
<tt>It's not the Dhegihan *aWa case that's bothering me. I agree that *aWa is almost certainly recently derived from two separate morphemes, *a + *Wa, which would make *W originally initial in that case too, and I don't think I have any disagreements with any of the material you presented in support of that view. What I'm having trouble with is visualizing how *W could have been *mb and *R *nd without their respective reflexes getting mixed up across Siouan with the reflexes of internal stops preceded by nasal vowels. Epenthetically, I think we have many effective cases of mp/mb and nt/nd in words such as 'day', 'moccasin', or 'face'. My question is: Why were these sounds never reinterpreted as *W and *R to give us, e.g., OP a'ma/Os ha'pa/La. awe'tu for 'day', OP hime'/Os hape'/La. ha'wa for 'moccasin', or OP ine'/Os itse'/La. ile' for 'face', if mb = *W and nd = *R and the originals of these terms were actually pronounced something like *(h)aNmpa, *haNmpa, and *iNnte' ?</tt><br>
<br>
<tt>I see only three avenues to wriggle out of the problem here, and none of them is very satisfying to me:</tt><br>
<br>
<tt>1. There is a voicing difference in the stop that prevents any confusion. Hence, internal mp is never confused with mb, and internal nt is never confused with nd.</tt><br>
<br>
<tt>2. The nasality of the vowel was kept so strictly separate from the following stop that epenthetic mp and nt never occurred internally.</tt><br>
<br>
<tt>3. The speakers of all Siouan languages throughout history, so far as we can tell, systematically distinguished word-initial sounds from equivalent word-interior sounds, and for *mp and *nt regularly reduced the former rather than the latter.</tt><br>
<br>
<tt>I did originally consider the idea of pre-nasalized stops for *W and *R, but I found the issue of internal epenthetic developments, and consequent confusion between the two sets, to be too much of a problem. It was after that that the idea of nasally-released stops first occurred to me. These would still give us the basic nasal and stop features we would like, while staying completely distinct and distinguishable from the matter of internal nasal vowel followed by stop consonant.</tt><br>
<br>
<tt>Rory</tt><br>
</body></html>