Linguistic classification

[manaster at umich.edu]

----------------------------Original message----------------------------
I am pleased that Larry agrees with most of what I said, and so
I hasten to correct sthg I said where he justly takes me to task.
I do not assume that all lgs are related and I accept that one
can criticize successfully a given proposal for a linguistic
relationship (as Larry has done for most Basque-X ones), BUT
what I mean is that we stil do NOT know that Basque is not
related to Turkish or Hebrew or whatever, merely that it is
not so related IN THE PARTICULAR WAY proposed by some particular
authors.  It might still be distantly related to ALL or SOME of
them in some totally unsuspected way, as Larry himself points
out I think.  All I mean then is that if we know that X and Y are
relatively closely related, then we know that X is not related
any more clsoely to W, as in my Hungarian-Turkish-Finnish
example, so that the only way X could be related to W would
be if W is also related to Y.  That is all.
 
The crucial point, of course, is that there is a diffrence
between refuting a particular proposal for a given relationship
(which is often easy) and refuting the relationship itself
(which is generally impossible, excpet in the relative sense
noted).
 
In fact, maybe if we used the terms 'asbolute' and 'relative'
we could get things straight.  There is generally no accepted
way of refuting absolute relatedness but there IS an accepted
way of refuring claims of relative relatedness.  That is my
basic point.
 
AMR
 
On Mon, 16 Feb 1998, Larry Trask wrote:
>
> [snip]
>
> > Note III: It seems to me that it follows from this that the only way
> > to argue against a relationship at all would be to demonstrate a
> > DIFFERENT relationship.
>
> Absolutely not.  This is Merritt Ruhlen's position, and it is
> indefensible.  Such a conclusion follows *only* if we assume at the
> outset that *all* languages are related -- and who in his right mind
> wants to do that?
>
<snip>
>
> I must say that I consider Note III to be a fallacy, and a dangerous
> one at that.  If I cannot prove that I am related to anyone else on
> the planet, it does not follow that I must be related to you.
>
> > E.g., contrary to what many Hungarian scholars have often asserted,
> > Hungarian is not (closely) related to Turkic because it is more
> > closely related to Finnish and the rest of Uralic.  But of course
> > this only allows us to argue about relative degrees of relatedness.
> > We cannot in principle argue against the relatedness of Hungarian to
> > Turkic, becuse Uralic as a whole might be related to Turkic.
>
> Agreed, but not the same point.