Gepids

[llama_nom]

Following on from these threads...

http://tech.groups.yahoo.com/group/gothic-l/message/8784
http://tech.groups.yahoo.com/group/gothic-l/message/8810

...I came across another hypothesis from Kemp Malone.

Kemp Malone (1933). `The suffix of appurtenance in "Widsith"' The
Modern Language Review 28:3, pp. 315-325.  [
http://links.jstor.org/sici?sici=0026-7937%28193307%2928%3A3%3C315%3ATSOAI%22%3E2.0.CO%3B2-P
].

He cites the following forms attested in Latin and Greek:

Gibedae, Gibedi, Gibites, Gibidi.
Gipidae, Gipides, Gipedes, Gippidos, Gippedi.
Gepidae (Gepidas), Gepidi, Gebidae (Gebidas), Gebidi, Gebedi, Gebeti.
Gebodi.
GHPAIDES (Greek)

(There's also Gypidas, Gyppidos and Gibidos manuscripts of the 'Origo
gentis Langobardorum'. Grimm notes that the Greek forms are
assimilated to the Greek word PAIS, thus: GHPAIDES, GHPAIDWN, GHPAISI,
GHPAIS. And regarding the possibility that eta stands for East
Germanic [i], compare Priscus's spelling ATTELA, and the discussion in
one of those earlier threads about the raising of eta from Classical
Greek [E:] to [i].)

The name of the people is attested twice in an early Germanic
language, namely Old English, each time in the dative plural: Gefðum
(Wídsíþ), Gifðum (Béowulf).

Malone's iggestion is that the name was formed from a Germanic root
'gaim', 'gîm', 'gim' found in Icelandic 'geimi' "sea" (Modern
Icelandic 'geimur' "big empty space; outer space") and 'gíma' "wide
opening; strip of clear sky". He points to place-names with the suffix
-and, such as Agand (whose inhabitants were the Old Norse 'Egðir') and
Verand (whence Old Norse 'Virðar'). He proseses that the 'm' of the
root was changed to 'f' to dissimilate it from the following nasal of
'and'--compare Old English 'heofon' : Old Norse 'himinn', Gothic
'himins'--and that the name of the people was formed from the root by
replacing the regional suffix 'and' with an ethnic suffix (suffix of
appurtenance) which consisted of a dental consonant-stem. This suffix
is attested in the names of several tribes and contains one of the
Proto-Germanic short unstressed vowels 'a', 'i', 'u' or no vowel at
all, the nil-grade.

Malone argues that the region and tribe took their name from a word
for "bay", a wide opening of water, this being a reference to their
old homeland: "No better example of a bay wide open on the seaward
side could be found that the 'Danziger Bucht', on the shores of which
the Gibids had their home."

The variation in the root vowel in the Old English forms could be
explained by paradogmatic levelling ('i' and 'e' each being
appropriate to different parts of the consonant-stem paradigm), or to
later (West Saxon) palatal mutation by the preceding 'g' = [j]. Malone
doubts that the variation between 'e' and 'i' in the Latin and Greek
forms is significant to the East Germanic pronunciation. He
reconstructs *Gibid- for the people and *Geband for the region. He
explains the form Gebodi as containing the 'u' grade of the suffixal
vowel. The Old English forms have an un-Vernerised /þ/ rather than
/d/, as is appropriate to a form with the nil-grade (with the stress
on the root).

Malone also discusses Jordanes etymology 'gepanta', which he takes to
be a punning term of abuse, a folk etymology: "We are told by Jordanes
himself that 'gepanta' "pigra" is a Gibidic word. And its 'e' for 'i'
and 't' for 'd' would have been enough to show that it has no place in
the Gothic of Wulfila." He suggests that the 't' could have come from
devoicing of 'd' before the 's' of the nominative masculine singular
in the strong form of the present participle, *gepand + s = *gepants,
the resulting 't' then having spread by analogy to the rest of the
paradigm. He keeps an open mind about the gender of the word, but
notes that if it had been a specifically feminine form, this would
have "added greatly to the 'convicium'" (slander) for Goths applying
the term to Gepid men.

LN

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/gothic-l/attachments/20080303/9e56c402/attachment.htm>