The Pennsylvania Tree, constructed by statistical computer techniques
from a large amount of data by Donald Ringe (Ringe,D.A., Warnow,T.
and Taylor,A. 2002, 'Indo-European and Computational Cladistics')
removed Germanic, and only Germanic, from the family tree altogether
in order to produce 'a best fit'. This may according to James
Clackson (Indo-European Linguistics. An Introduction. New York
Cambridge University Press 2007 p.12ff.) "point to unusual changes in
the prehistory of Germanic, and possible convergence with other IE
varieties".
According to Clackson, Gray and Atkinson used a methodology based on
modern vocabulary, which was less likely to weed out later
influences. According to him "the Germanic languages have
vocabularies which have been influenced by and which have influenced
the modern Romance languages, and this is reflected in the closeness
of of the languages on the New Zealand tree".
Jouppe
--- In
cybalist@yahoogroups.com, "Richard Wordingham" <richard@...>
wrote:
>
> --- In cybalist@yahoogroups.com, "fournet.arnaud" <fournet.arnaud@>
> wrote:
>
> > What about the absence of isoglosses with Italic and Celtic ?
>
> E.g. Pokorny *teu 'listen to, observe' #2008, pp1079-1080
> Pokorny *tra:gH 'drag, move, run' #2024, p1089
> Pokorny *bHren 'to stck out, edge #293, p167
>
> You may recall that Gray and Atkinson had an Italo-Germano-Celtic
node.
>
> > and numerous isoglosses with eastern IE languages ?
>
> A centrally located language?
>
> Richard
>