Scandinavian languages

Larry Trask larryt at cogs.susx.ac.uk
Wed Apr 14 08:28:39 UTC 1999


Adam Hyllested writes:

[on my remark about the placement of Norwegian]

>  Then take a glance at my tree (upside-down).
>
>            NORDIC LANGUAGES
>  WEST NORDIC                  EAST NORDIC
>              Old Norse (+)                *Old East Nordic (+)
>    Icelandic                           Swedish
>      Faroese                Gutnish (+)
>        Norn (+)              Danish
>          Norwegian          Norwegian    (Nynorsk)        (Bokmål)

Well, I guess this cuts the Gordian knot, all right. ;-)

[snip]

[LT]

>> So, to put it crudely but picturesquely, Norwegian has migrated from one
>> branch of the tree to another.  And this is not the kind of phenomenon
>> that the family-tree model can accommodate at all well.
>> Some decades ago, either Trubetzkoy or Jakobson -- I forget which --
>> suggested that English had ceased to be a Germanic language and become a
>> Romance language.  Much more recently, C.-J. N. Bailey has likewise
>> asserted that English is no longer a Germanic language but may perhaps
>> be a Romance language.

> To me, that doesn't make sense. The genetic classification of languages is
> based on origins, not on linguistic similarities caused by later foreign
> influence.

Exactly.  So, based on origins, we'd expect Norwegian to form a branch with
Icelandic.  Yet all the published trees I've ever seen group Norwegian with
Danish and Swedish, in defiance of the original state of affairs, but in line
with modern realities.  I'm not defending any particular analysis here,
merely pointing to an inadequacy of our family-tree model.

> A Germanic language stays Germanic forever, no matter how
> unrecognizeable it may have become. The leaves of a family language tree
> simply cannot move from one branch to another.

According to the family-tree model, this is correct.  But my point was that
this model does not suffice to capture all aspects of language history:
sometimes reality is more complicated than this model can allow for.

I might add that the well-known Pennsylvania tree for the IE family sees the
entire Germanic branch as having done something similar: this view sees
Germanic as having started off as part of an eastern cluster of IE languages,
but then as having "migrated" (linguistically, I mean) into a western
cluster.  The authors' final decision is to put Germanic into a western
branch, but they explicitly acknowledge the inadequacy of this decision.

> Problems in family-tree
> classification occur, however, in cases of some pidgin languages with
> roots in two (or more) language groups, or in cases of "mixed languages"
> where the antecedents, including the direction of influence, remain
> unclear.

Yes, and this too is something I had in mind.  The assorted "mixed
languages", "portmanteau languages", "metatypic languages" and what not that
have been recognized or argued for just don't fit comfortably into any family
tree.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk



More information about the Indo-european mailing list