Context-sensitive

From: tgpedersen@...
Message: 10572
Date: 2001-10-24

> --- In cybalist@..., tgpedersen@... wrote:
>
> In his zeal to create a Medusa's head to scare off abominable
> amateurs, Rosenfelder has created a monster that will turn any
> language comparison into stone, including IE (that is, excluding
> pairs of languages, the path between which consists of context-free
> rules only).
>
> And BTW, you may point out for the umpteenth time that no, no,
> Rosenfelder shows that you can't use "similarity" in linguistics.
But
> I agree, he does; unfortunately in the process he has shown
> that "proper" linguistics is impossible too. That's why I called
for
> a modification of his theory.

--- In cybalist@..., "Piotr Gasiorowski" <gpiotr@...> wrote:
> First, Rosenfelder is an amateur himself; not a professional, but a
> well-informed amateur with an admirably incisive mind and a talent
> for popularising good science -- rare gifts these days.
Yes, yes. I never claimed otherwise.
>

> Secondly, what I will repeat until you understand or I get tired is
> that Rosenfelder has no "theory" to sell.
I never claimed he had.
>He only evaluates claims
> based on mere "resemblances" and shows why they must be rejected.
Yes-yes, as I acknowledged, but in the process he shows that anything
using context-sensitive rules must be rejected too.

>If
> his model could be used to show that comparative analysis must be
> rejected as well, we'd have a reason to worry, but it can't.
All right, show me. Run an example of a set of related words of two
related languages through Rosenfelder's model.

>The
> comparative method offers controls that simultaneously minimise the
> dangers of accepting bogus positives and of overlooking _real_
> positives.
Yes, and I never claimed otherwise.

>Context-free rules, e.g. those that make up Grimm's Law,
> produce systematic correspondences just like context-sensitive
>ones,
> so I can't see how the resultant patterns could be rejected by
> Rosenfelder or anyone else.
Well, do the example then.

>The monster he has allegedly created must
> be something in your mind.
Do the example and it will also be in *your* mind.

> Piotr


Torsten