<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Martin's latest message is helpful in clarifying the question. And
if I may add my two cents worth ...<br>
<br>
On 04/02/2018 19:06, Martin Haspelmath wrote:<br>
<span lang="EN-US"><o:p></o:p></span>
<blockquote cite="mid:5A76DB2F.2090105@shh.mpg.de" type="cite">
<p class="MsoNormal"><span lang="EN-US"><o:p> </o:p></span><span
lang="EN-US">I head-ache.<o:p></o:p></span> </p>
<p class="MsoNormal"><span lang="EN-US">You feet-washed me.<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US">Father knife-sharpened the
neighbor. (Alexey (“Lesha”) Vinyar’s Chukchi example)<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US">Now Alexey Vinyar’s
question is whether one gets this sort of pattern also with
pseudo-incorporation (PNI), where the possessed noun is still
strictly adjacent, but a complete noun phrase can occur in
this position (not just a noun stem). So this would be
something like <o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US">You [@dirty feet]-washed
me.<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US">Father sharpened-[@dull
knife] the neighbor.<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US">I don’t know of any cases,
but I’m not sure whether PNI is a particularly useful notion
to begin with. As seen nicely in Borik & Gehrke’s (2015)
book, there are a wide range of related constructions with
some family resemblances but apparently little overall
coherence.</span></p>
</blockquote>
Well in colloquial Malay/Indonesian, you can easily get structures
such as<br>
<br>
Aku jari kelingking sebelah kiri sakit<br>
I [ [ finger pinky side left ] ache ]<br>
<br>
and such structures are very common in western Indonesia and
Mainland Southeast Asia. The typical way of looking at them is as
instances of<br>
<br>
TOPIC [ SUBJECT PREDICATE ]<br>
<br>
where the relationship of possession is just one possible outcome of
a broader topic/comment relationship. Of course, in the cases that
I am most familiar with, at least, there's no evidence that the
SUBJECT (be it simple or complex) cliticizes to the predicate. But
Martin, at least, would surely share my reluctance to define a
construction type based on as elusive a property as wordhood.<br>
<br>
I suspect, however, that constructions such as the above are not
exactly what Alexey is looking for, which may be considered as
reinforcing Borik and Gehrke's point cited above by Martin. <br>
<br>
David<br>
<br>
<br>
<pre class="moz-signature" cols="72">--
David Gil
Department of Linguistic and Cultural Evolution
Max Planck Institute for the Science of Human History
Kahlaische Strasse 10, 07745 Jena, Germany
Email: <a class="moz-txt-link-abbreviated" href="mailto:gil@shh.mpg.de">gil@shh.mpg.de</a>
Office Phone (Germany): +49-3641686834
Mobile Phone (Indonesia): +62-81281162816
</pre>
</body>
</html>