non-IE/Germanic/k/q

[Larry Trask]

On Sun, 21 Feb 1999, Rick Mc Callister wrote:

> kalh- > callow [< ?Vasconic"] [tv97]

Why?

> kante [Celtic, Germanic, Italic non IE] > Kante "edge"
> [< ?Vasconic; see Basque kantu "slice, angle, rocks",
> common in Iberian toponymy] [tv97, tv95]

There are at least two different Basque words tangled up here, unrelated
and both of Romance origin.

The first is represented by Roncalese <kanto> `slice', Zuberoan
<k(h)antu> `side, edge, vicinity', `angle', `slice'.

An extended form of this is represented by Lapurdian and Low Navarrese
<kantoin> `slice', `piece', `corner, angle', Zuberoan <kantu~> (with
a nasal vowel) `slice', and by a hapax <kantoi> `slice', of uncertain
provenance.

The second is represented by Lapurdian <kantu> `rock, boulder'.

The first derives ultimately from a late Latin <cantu>, which yields
Castilian <canto> `extremity, side, point, corner', `crust, slice'.
This has an extended form <canto'n> `corner', which probably derives
from earlier *<cantone>.

Latin <cantu> or Romance <canto> suffices to explain the first Basque
word perfectly,  Its Romance extension <canto'n>, or an antecedent
*<cantone>, suffices to explain the Basque forms in <-oin>, <-oi> and
<-u~> perfectly, since these are the regular developments of Romance
<-one>.  (Compare <arrazoin> ~ <arrazoi> ~ Zuberoan <arrazu~> `right,
reason', from Old Castilian *<razone>, modern <razo'n>.)

The word meaning `rock' is straight from the common Ibero-Romance word
represented by Spanish <canto> `stone', of unknown but very likely
Celtic origin.

The Basque words cannot possibly be ancient in that language, because
plosives were categorically voiced after /n/ in early medieval Basque,
save only in Zuberoan and Roncalese.

> knife, knyft "pocketknife" [W Fries],
> kni:fr [Scandinavian],
> Kneif/Kneip "pocketknife",
> canivet [O Fr],
> canif "large pocketknife" [Fr],
> Basque ganibet, kanibet "pocket knife"
> [< ?Vasconic] [tv95]

Basque <ganibet> ~ <kanibet> just means `knife' in general, specifically
a knife which has neither a sheath nor a folding handle, such as a table
knife.  It is perfectly clear that this is borrowed from Old Gascon
<canivet> (modern Gascon <ganivet>), related to Old Aragonese <ganivet>,
Old Castilian (and still regional Spanish) <can~ivete>, and to the cited
Old French <canivet>.  All these are derived by adding a Romance
diminutive suffix to the word represented by modern French <canif>
`penknife, pocketknife', which in turn unquestionably is borrowed from
the Germanic word which is the ancestor of English `knife'.

The word cannot possibly be ancient in Basque.  First, a native word
does not exhibit fluctuation in the voicing of an initial plosive: this
is a typical characteristic of loan words.  Second, earlier intervocalic
/n/ was categorically lost in the early medieval period, and could not
have survived into the historical period.  Third, a native and
monomorphemic word does have a plosive in the onset of the third
syllable.  Fourth, a native word cannot end in /t/ -- or in any plosive
-- except where that plosive is secondary.

This is a "solution" to a problem which does not exist.

> k'rimp'an "tense, cramped" > krapfo "hook, claw" > cramp, Krampf; krapfen
> "fritters, donut"
> [< ?Vasconic;
> see pre-Basque *garba "broke, junk", *krapo "claw, clip, junk"]
> [acc. cw < ? IE *ger-] [cw, tv95, tv97]

Er -- what?  What on earth is meant by this "pre-Basque *<garba>"?
No such form can be reconstructed for Pre-Basque.  Moreover, if it could
be, it should appear unchanged in modern Basque as *<garba> -- but I
know of no evidence for any such word.

> ku:z/e [HG 15th c.] > Kauz [type of owl]
> [?< Vasconic *kuwonts/a > *k^u:nts > *k^u:ts,
> see Basque hutz, ontz] [tv97]

The Basque word is <ontza> ~ <untza> ~ <untz>; I don't know of a variant
*<hutz>.  The word is of unknown origin, but is suspected of being
imitative, like some of the neighboring Romance words for `owl'.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk