Northwest IE attributes

[Sean Crist]

On Fri, 5 Nov 1999 X99Lynx at aol.com wrote:

> Satem is still a very good way to separate I-Ir from the western group.  (I
> believe the current stance is that satem may have been adopted by
> Balto-Slavic.)

>From the early days of comparative work in the Indo-European languages, it
has been universally observed and accepted that both Indo-Iranian and
Balto-Slavic underwent the satem consonant shift.  Saying that this is the
"current stance" suggests that there's been controversy on the matter.

> And Hittite separates itself from the northern group easily
> enough.  That leaves a distinct NW IE group.

If we exclude Anatolian, Indo-Iranian, and Balto-Slavic, this leaves
Celtic, Italic, Germanic, Greek, Armenian, Tocharian, and the poorly
attested branches such as Phrygian.  Did you really mean to include Greek,
Armenian, and Tocharian in your "NW IE" grouping?  This strikes me as a
rather odd grouping, and I doubt it's what you meant.

> ON THE OTHER HAND:  No one to my knowledge has posted a list of the "shared
> attributes" that the UPenn tree is based on.  The best I saw was the few you
> posted on the list and the few posted on the web.

The most I can say is to point you to the articles which have already been
published (I'm sure the references are on the web site), and to tell you
again that the team is working on a monograph which will give every bit of
information about their work.  This is work in progress, and getting
everything published takes time.

[...]
> It would be good to know - however - why *specifically* you think some group
> of IE speakers could not have branched off from PIE right from the start and
> moved Northwest.  That would be very interesting.

Suppose we construe "NW IE" to mean Italic, Celtic, and Germanic.  It is
quite true that these three branches share a number of lexical items not
found in the other IE languages, and on those grounds, we might be tempted
to group them.  However, there are morphological items which would lead us
to group the languages otherwise: for example, Italic and Celtic have an
unusual superlative suffix not found in Germanic or in the other IE
branches; other examples could be given.  We know that words are readily
borrowed, but inflectional morphology is almost never borrowed.  Probably,
the right answer is that the shared lexical items represent early loans
between languages which don't form a proper clade in the IE phylogeny.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]