Germanic weak preterites from yet another angle

Cayce: "Grammar of the Gothic language", p. 154
'2. gaggan (§ 313, note i) is properly a reduplicated verb, the pret.
of which, gaígagg, has been lost. The extant forms of iddja (§ 168)
are inflected like nasida (§ 317); in one instance a weak pret.
gaggida also occurs.'

If iddja "went" is inflected like a weak preterite, it must tell us
something about what the form is of the auxilliary verb *dhe:-.
Elsewhere Cayce says 'Goth. iddja, I went, cp. Skr. áya:m', so he
thinks it's an imperfect.
But as far as I know, it takes -jj- > -ddj- in Gothic so we are one
-j- short (Wikipedia: 'According to Lehmann (2007)
['Proto-Indo-European Phonology'?], the lengthening occurs as in the
contexts of PIE *-VwH-, *-iyH-, *-ayH-, *-aHy- (where V is any short
vowel, and H is any laryngeal).'
So, is it *j-éje/o "made go"? 3sg. injunctive *ej-j-o:m?


Something else:
The Gothic second weak conjugation, salbo:da, must be based a nominal
form salbo: (< PIE *solb-ax), plus a finite form of the *dhe:- verb.
Now -ax is a common case form, which means that if we assume the
unsuffixed verbal root could once be used as a participle vel sim.,
the <root>-ax is case form of that stem and can be used as such in a
periphrastic construction, cf. Latin 1st conj. ama:bam < *am-ax bhwom.
That means there is no need for haplolology in at least the Gothic
second conjugation, it's *salbo: dda, not *salbo:þ dda.

With that success, why not try the Goth. 3rd conj., with the stem
hab-a- with a < e: (cf. Latin). But it's 3sg pret. habaida, with -ida.

Is this -ida < iddja?


Torsten