Re: [tied] The Russian God Svarog

Svarog (*svarogU) is the only theonym securely attested both among the
Eastern and the Western Slavs. In a twelfth-century copy of a
translation of the work of Ioannes Malalas, a sixth-century Byzantine
historian (more precisely, in a section devoted to Egyptian deities),
and in the glosses and commentaries to the text, the Old Russian
commentator mentions Svarog, identifying him with Hephaestus, and
Svarog's son Dazhbog (< *dadjI-bogU 'giver of prosperity', a name
attested in some other East Slavic sources), equating him with Helios
(that is, the Sun as a deity). Other sources (apart from those that cite
the translation of Malalas) don't mention Svarog, but the name is
attested as a toponymic element, also in West Slavic lands (e.g. Pol.
Swaroz.yn, with a possessive suffix).

At the northwestern end of Slavia, Thietmar of Merseburg mentions
<Zuarasici> as a major god among "all the Slavs", and one of the most
important deities worshipped at "Riedagost" (a.k.a. Rethra), a famous
sanctuary in the country of the pagan Redarians, a tribe of the Polabian
Slavs. The description of the sanctuary is part of Thietmar's account of
Emperor Henry II's expedition against the Poles in 1005, in which the
Redarians took part as the Emperor's allies. <Zuarasici> is a rendering
of West Slavic Svaroz^ic (*svaroz^icI = *svarog- plus the patronymic
suffix -icI = East Slavic -ic^I, as in Russian patronymics, from
Proto-Slavic *-itjI). Svaroz^ic (Svarog's son) is probably the same
deity as Radogost ('hospitable'), widely worshipped among the Polabians,
or rather an alternative epithet of Dazhbog/*Dadzbog (or of his
brother!). One 14th-c. Old Russian source mentions <svaroz^ic^I> as a
name given to fire in pagan practices.

Therefore, Svarog seems to have been the father of the Slavic solar god
(possibly also the giver of wealth); if his functional identification
with Hephaestus has any real foundation, he may have been the Slavic god
of fire, and perhaps the artisan among the gods (but this is getting
speculative).

As for the origin of the name, *svarogU has no convincing Slavic
etymology (derivation from *svarU 'noisy argument' is folk-etymological,
as *-og- is not otherwise known as a Slavic suffix). Because of the
solar and fiery connections it's very tempting to compare it with
Indo-Iranian *s(u)war-/*su:r- 'sun, sunshine, heaven' (common descent as
opposed to borrowing is ruled out, since IIr. *r < *l in this case).

There are, however, problems with the comparison. First, the
hypothetical prototype of *svarogU ought to have been something like
*swa:raga-. It doesn't look like any known Indic or Iranian derivative
of *swar-. The *-ga- part could be the extremely common Iranian suffix
*-ka- with intervocalic voicing in Middle Iranian (which, however, is
essentially a West Iranian development); *swa:ra-, however, can't be
Iranian at all if it has anything to do with the 'sun' etymon, since
*sw- gives Iranian *xW- (--> Slavic *xv- in loanwords), cf. Av. xVar- 'sun'.

On the other hand, if we are prepared to accept the possibility of loans
into pre-Slavic from the non-Iranian language of the hypothetical
"Pontic Indo-Aryans" (or, to put it more cautiously, from a "basal"
Indo-Iranian language that shows no Iranian innovations), a derivative
in *-ga- makes little sense. Skt. s(u)var-ga- 'celestial; heaven' is
only superficially similar. It's a compound with <svar->; there are many
such in Sanskrit, but they have no "sva:ra-" variants (if such a form
existed, it'd be homophonous with <sva:ra-> 'sound, noise, tone' < IE
*sworo-s).

Jakobson hypothesised that *svarogU was a "mythological and lexical
borrowing" of *va:ragna- < *vr.þra-Gna- (see the etymology of Slavic
*rarogU and Lith. va~nagas 'falcon', discussed here a few months ago),
but his etymology doesn't account for the initial *sv-. Contamination
between Indo-Iranian (but non-Iranian) *swar- 'sun' and the Iranian name
of a divine serpent-slayer seems rather unlikely to me.

To sum up, I don't know where Svarog's name comes from. It does look
vaguely Indo-Iranian, but attempts to work out a plausible detailed
etymology haven't been particularly successful so far. Any ideas from
Cybalist?

Piotr