Re: Language reconstruction question

En réponse à Ian Maxwell <umlaut@...>:

> Hiya.
>
> So, I was just randomly wondering: How possible is it to reconstruct
> an
> ancestor language based on only one descendant?
Difficult but not completely impossible. If you have a lot of irregular forms
that seem to go together in schemes and/or a lot of alternations (basically
phenomena like gradation, lenition or mutation), you can try to explain those
forms by more regular ones which got mixed up through sound changes. The method
is about as precise as the comparative method, but usually works only on small
scales.

The PIE laryngeals are an example of internal reconstruction, and when Hittite
was discovered and discovered to have them, it was a proof the method worked.
The 1-vowel hypothesis also comes from internal reconstruction, but about this
one people are a bit more divided :)) .

 I was under the
> impression that reconstruction was basically done by comparing the
> grammars of various descendants.
Indeed, the comparative method is the main (and only in cases of languages with
few or no alternations, like the Chinese languages) method of reconstruction,
and the most successful. But it's wrong to think that internal reconstruction
gives less trustworthy results. It gives results as trustworthy as the ones
from comparison. It's just that it has a smaller scope (you can do comparison
on a large scale, but internal reconstruction works only on details).

 If it's not possible, theoretically
> there could be older versions of the reconstructed proto-languages
> that
> we don't (and can't) know about, ne?
>
Indeed. The further back in time you go, the less sure of it you get. Getting
anything older than PIE (and PIE is not that well known) is pure speculation.
The method just doesn't allow it.

Christophe.

http://rainbow.conlang.free.fr

Take your life as a movie: do not let anybody else play the leading role.

Reply