Re: [tied] l-ikW-

From: Piotr Gasiorowski
Message: 14366
Date: 2002-08-17

It should be PGmc. *laib-j-an- and *li:b-an- (with long *i:, as if from pre-Germanic *leip- via Grimm and Verner). The choice is really between PIE *leikW- 'leave, abandon' and *leip- 'adhere, be sticky'. Both solutions have their advocates. My own preference is for *leikW-, which is more plausible semantically (and "would be nice" indeed), and since the occurrence of *f for expected *xW in Proto-Germanic is well attested, albeit sporadic, I see no serious formal obstacles.
----- Original Message -----
From: tgpedersen
Sent: Saturday, August 17, 2002 12:13 PM
Subject: [tied] l-ikW-

What's the deal with "leave"?

My dictionary says Proto-Germanic *laib-ian, causative of
*liban "remain". But if the latter had been **leivan, it might have
come from **leiGWan, in Verner-relation to *leixWan, from PIE *leikW-
"leave", which would be nice. Has this been sorted out already?