reality of PIE

[Larry Trask]

Steve Long writes:

> I cannot think of stronger, more compelling evidence in favor of the actual
> existence of PIE than what Ante Aikio presented in his list of PIE borrowings
> in Uralic.

This external confirmation is indeed gratifying.  But it's not essential.
The IE-internal evidence is already sufficiently strong to justify the
reconstruction of PIE.

> The claim that PIE can be reconstructed accurately is always going to be
> confirmed by internal evidence, after all that is what the reconstruction is
> made of.  There's really no way to disprove it internally, because by
> definition it was created to be consistent with the evidence.

Oh, no.  Not so.  You're overlooking something vastly important.

We can't just choose some arbitrary languages and "reconstruct" a
common ancestor for them.  For example, we can't choose Norwegian, Basque
and Zulu and "reconstruct" a Proto-NBZ.  The data simply won't allow this.

But the IE languages *do* permit the reconstruction of PIE.

> But to see it confirmed in Uralic is very, very impressive.  That's
> predictability.

It is certainly helpful in persuading non-specialists that our methodology
really works.

> It is very hard to be cynical about the actual existence of *PIE when you
> have that kind of external evidence.  I wonder if the full impact of that
> work has been appreciated.

Linguists certainly appreciate it.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk