Reconstruction talk:Proto-Indo-European/bʰréh₁wr̥

Root
How come the h₁ and w are swapped relative to their position in the root? —CodeCat 15:47, 6 July 2016 (UTC)
 * I've been trying to figure this out. Beekes and Kroonen each relate it to *bʰer-w-. This entire set of forms (*bʰer-w-, *bʰrew-, *bʰrewh₁-) seem to represent a constellation of related roots. You could, I suppose, get from *bʰréwh₁-r̥ > *bʰréh₁wr̥ with metathesis, but Beekes prefers *-wr̥ as the suffix as in and . LIV, under *bʰrewH-, seems to derive *bʰréh₁wr̥ hesitatingly, also suggesting a lengthened grade *bʰrḗw-r̥ in passing, but I'm having a little difficultly understanding their precise meaning (If you could double check, that would be very helpful). Given the existence of *-wr̥ and *bʰrewh₁-, I would propose metathesis of *bʰréwh₁-r̥ by analogy with *-wr̥ (in particular with ). Regardless, we will need several different roots (*bʰer-w-, *bʰrew-, *bʰrewh₁-) because *bʰrewh₁- alone will not cut it for . — JohnC5 17:57, 6 July 2016 (UTC)
 * is rather strange as far as roots go, as it ends in two sonorants. So I'd guess that it may be not a root but a root plus suffix. —CodeCat 18:20, 6 July 2016 (UTC)
 * That's true enough, and I often see it represented as *bʰer-u-. Are we imagining + *-u- gives “to move violently; to seethe, to bubble”? That's a somewhat odd of a semantic shift—plausible but odd. Could we propose *bʰer-w- with a vṛddhi-derivation *bʰrew- which was then extended to *bʰrewh₁-? — JohnC5 18:30, 6 July 2016 (UTC)