Re: [tied] Re: An odd etymology

From: Piotr Gasiorowski
Message: 32597
Date: 2004-05-13

----- Original Message -----
From: "tgpedersen" <tgpedersen@...>
To: <cybalist@yahoogroups.com>
Sent: Thursday, May 13, 2004 11:02 AM
Subject: [tied] Re: An odd etymology

Just some loose comments:

> PIE *akWis-/*akWus- "axe"
> PIE *ak-/*ag^- "sharp"
> PIE *ak- "stone"

The labiovelar in Gmc. *akWizjo:/*ak(W)usjo: (cf. Myc.Gk. a-qi-ja) makes
comparison with *h2ak^- (no *h2ag^- variant!) impossible. The etymological
source is obviously something like *(h2)agWis-jah2.

> PIE *op-
> Latin opinor "believes"
> PGmc. *ho:p-/*hoff- "hope"
> No match outside Germanic; my Danish etymolgy book wants to connect
> it *hopp- "hop" as in "jump for joy". Hm!

The PGmc. reconstruction is faulty (actually the word seems to have
originated in the westernmost, Anglo/Frankish part of West Germanic, so
"PGmc." is an overstatement). There's no long *o: there, and the short /o/
of OE hopian and hopa represents lowered *u, so the hypothetical Germanic
prototypes would have been *xupo:jan- (verb), *xup-an- (weak noun).

> PIE *ap- "bind"

Isn't it *h1ep-?

> PGmc. *hap-/*hab- "grasp", "have"

But *kap-je/o- is surely PIE, not NWBlockish

> Latin odi "hate"
> Gmc hate
> but they have plenty of cognates

... and don't appear to be related. PGmc. *xadaz/*xadiz- was an *-es- stem
with /k/-initial cognates at least in Celtic (I have doubts about cognates
farther afield under Pokorny's *k^a:d- or EIEC *k^ah2des-)

> PIE *omb-/*ombh- "swell"
> English hump

But <hump> was first attested in 16th-c. Low German and Dutch; its
attestation in English is even more recent (the 17th c.). Not much to build
an ancient etymology on. It's accidentally (or phonosymbolically?) similar
to a number of other things, e.g. *kumbo- 'pot, bowl'. Again, the vowel of
<hump> doesn't match that of *ombH-

> PIE *op- "be abundant"
> Latin copia
> PGmc *haupa-/*hu:pon "heap"
> I suspect the -au- was posited to accomodate -o:- in low German. It
> might have -o:- to begin with.

English heap (OE he:ap) definitely requires *au. Latin co:pia < *co-op-
'plenty together'.

Piotr