Re: long, flat, full

On 2008-10-06 15:37, Brian M. Scott wrote:

>
>
> At 6:04:20 AM on Monday, October 6, 2008, Arnaud Fournet
> wrote:

> > Why should everything change ?
>
> Doesn't really matter: the evidence clearly shows that
> languages *do* change over time.

Still, Arnaud's "why" deserves an answer. Languages change primarily
because of the nature of cultural transmission from generation to
generation. The kind of cultural replication which underlies language
acquisition is inherently imperfect, and copying errors inevitably
accumulate over time.

> > Do I have to prove that something can be stable ?
>
> Yes, if you want to claim that a significant number of words
> have remained stable for 7000 years at a stretch.

There are quite a few languages whose _documented_ history stretches
back two millennia or so, and not a single spoken language that has
resisted a significant amount of change for so long.

Piotr