Re: [tied] Cymerians?

Somebody -- I think it may have been Eric Hamp, but I don't recall exactly where -- hypothesised that Cimmerian was a distinct branch of IE, characterised by a peculiar sound shift whereby the voiced aspirated series underwent devoicing and deaspiration (PIE *dH > *t) while the plain voiceless stops became voiced (PIE *t > *d; I forget what happened to the voiced unaspirated series). It was argued that a substrate with such a shift was detectable in Slavic (as unexpected voicing), and the ethnonym Kimmerio- was explained as something like *g^H[m]m-er-jo- (or maybe *-el-jo- if the Cimmerians rhotacised their l's) from *g^Hom- 'earth'. Pretty far-fetched, and to be taken with a very large pinch of salt, given that we have absolutely no historical documentation of _any_ Cimmerian words (except for a couple of Iranian-looking personal names).