IE "Urheimat" and evidence from Uralic linguistics

[Robert Whiting]

On Tue, 15 Feb 2000, Hans Holm wrote:

<snip>
> You cite Anttila with
> RW>"'Related' is a technical term, exactly like the equivalent 'cognate',
> RW>meaning that the items were once identical."

> And then go on:
> RW>This is the criterion of genetic relatedness in historical linguistics.

> 1) Anttila speaks of lexemes/items.

No, Anttila is speaking of "Language Families and Family Trees."  Read
the heading of the section.  "Lexemes" are nowhere mentioned in the
section.  The "items" Anttila is talking about are languages and dialects.
Read the paragraph in which the statement occurs.

> 2) This was not the point I objected.

Good.  Then I presume that you agree that related languages are descended
from some common source and that therefore languages either start out
related or they will never be.

> It is no use to state things as true, which nobody can prove or disprove.
> So, changing your 'related' to 'relatable', in the sense of Anttila, will
> be okey.

It is equally useless to state that because there is no way to demonstrate
that any two languages are not related there is no such thing as unrelated
languages.  Since the non-relatedness of languages cannot be established
empirically, unrelated languages are simply those that cannot be shown to
be related.  And it not "my" 'related'; it is historical linguistics'
'related.' If you don't want to take Anttila 89 as the Bible (and there is
no reason why you should) then check out some other introductory textbooks
on historical linguistics (e.g., Hock 86, 8: "... these languages are
descended from a common source.  In such cases we speak of Related
Languages.").  But historical linguists know that when one says that
two languages are not related it means "not demonstrably related."

Bob Whiting
whiting at cc.helsinki.fi