IE "Urheimat" and evidence from Uralic linguistics

[Larry Trask]

Pete Gray writes:

>   I mean that creolisation / language mixing or whatever you call it provides
>  us with an example of a language which goes back to two ancestors, not one.

Creolization is not at all the same thing as language mixing.

>  Germanic is clearly and certainly related to other IE languages.   If the
>  theory about it being a creole were true (note the subjunctive, indicating
>  unreality), then it would be:
>     (a) related to other IE langs, and at the same time,
>     (b) also related to some other original language, which had no genetic
>  relationship to the other IE langs.

>  Does it make sense in a situation like that - whether it is Germanic or any
>  other  language which is involved - to insist that "all related languages
>  descend from a single common ancestor"?

OK; folks.  My two cents.

First, as others have pointed out, there are big problems with asserting that a
creole derives from two (or more) ancestors.  This is far from being obviously
true, and many of us would prefer to say that a creole has *no* ancestors at
all.  That is, a creole is *not* genetically related to any of the languages
which may have provided input to it.  It is connected to them in some way, but
it is not genetically related to them.

Second, Germanic is a simply terrible example.  Germanic looks not at all like
a creole, but it looks very much like a daughter of PIE.  If the Germanic
languages descend from a creole constructed in part from PIE, then why the hell
do they preserve PIE ablaut, PIE inflectional endings, PIE word-forming
morphemes, and other PIE complexities that should have vanished in any creole?
And why does Germanic phonology correspond so systematically to PIE phonology?
The idea of Germanic as a creole is a non-starter, so let's forget about it.

Now let's turn to a real and much better example: the North American language
Michif.  Michif is probably the finest example of a mixed language on the
planet.
 To oversimplify a bit, Michif consists of a French nominal system (with French
lexicon and phonology) bolted onto a Cree (Algonquian) verbal system (with Cree
lexicon and phonology).

OK.  What should we say about Michif?  Is it well described as a language
descended from two ancestors?  Or is it better described as a language
descended from no ancestor at all?

The floor is open.  I know what I think, but I'll leave that for later. ;-)

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk