On 2007-03-28 13:40, Angela Johnson wrote:
> Esteemed Piotr and Mate,
>
> Thank you for your responses. My resources are limited, and I value your
> attention.
>
> Am I to assume that the Gothic and Lithuanian forms of the dual pronoun,
> for example, represent neologisms created subsequent to the earliest
> reconstructable PIE, or dialectal creations within late PIE, perhaps?
Goth. wit (also found in Old and Early Middle English, OFris. and OSax.,
and cf. OIc. vit) comes from *we plus a clipped form of the numeral
'two' (*we-d(wo:)). OE also had <git> (pronounced /jit/ > EME yit),
which seems to reflect *ju-t < *ju-d(wo:) influenced by the 1du. pronoun
(Cf. OIc. it, Bavarian es). Gothic probably had *jut, but the word
happens not to be attested in the Gothic Bible.
Lith. ve`du and ju`du are of the same origin as the Germanic forms.
The Germanic oblique forms are:
1du. Goth. ugkis, OIc. okkr, OE/OFris./OSax. unc
2du. Goth. igqis, OIc. ykr, OE/OSax. inc, Bavarian enk
The PGmc. forms were respectively *unk(w)- and *inkw- (the dat. ending
*-iz is from the dat.pl.). The former may come directly from PIE
*n.h3we, possibly with a change of *h3 > *g > k between a resonant and
*w ("Cowgill's Law"), or perhaps from reduplicated *n.h3w(e)-n.h3(we) >
*unwun > *ungun ("Seebold's Law") > *unk- (Grimm's Law). The origin of
*inkw- is... well... mysterious. It may have been influenced by the
pattern of 1pl. *unsiz, 2pl. *izwiz, but the 2pl. form is difficult to
explain too.
Piotr