Native K, G sounds in Satem languages

[arvistro]

It appears there is some general sort of misunderstanding with K and G in Satem being treated as irregularity or a borrowing.

However the theory is (quotes from wiki):
"The centum group includes the "western" branches of Italic, Celtic, Germanic and Greek. This group merged Proto-Indo-European palatovelars and plain velars yielding plain velars only, but retained the labiovelars as a distinct set.[1]

The satem languages include the "eastern" branches of Balto-Slavic and Indo-Iranian. This group lost the labial element of Proto-Indo-European labiovelars and thus merged them with plain velars, while the palatovelars remain distinct, realized as a sibilant phoneme.[3]"

So PIE had kw, k, ḱ sounds (labiovelars, plain velars, palatovelars). Same for G.
- In Centum k, ḱ were merged into single k. Kw had different destiny based on branch.
- In Satem kw, k were merged into single k (!! Not S!!!). Palatized ḱ became s.

So, for example, Baltic words for body parts - kāja, galva, akis, etc, are considered native, coming from kw or k (galva of course from gw not kw).
Just as sirds is considered native coming from ḱ.

Hope it settles the matter.

 

[Taranis]

It appears there is some general sort of misunderstanding with K and G in Satem being treated as irregularity or a borrowing.

However the theory is (quotes from wiki):
"The centum group includes the "western" branches of Italic, Celtic, Germanic and Greek. This group merged Proto-Indo-European palatovelars and plain velars yielding plain velars only, but retained the labiovelars as a distinct set.[1]

The satem languages include the "eastern" branches of Balto-Slavic and Indo-Iranian. This group lost the labial element of Proto-Indo-European labiovelars and thus merged them with plain velars, while the palatovelars remain distinct, realized as a sibilant phoneme.[3]"

So PIE had kw, k, ḱ sounds (labiovelars, plain velars, palatovelars). Same for G.
- In Centum k, ḱ were merged into single k. Kw had different destiny based on branch.
- In Satem kw, k were merged into single k (!! Not S!!!). Palatized ḱ became s.

So, for example, Baltic words for body parts - kāja, galva, akis, etc, are considered native, coming from kw or k (galva of course from gw not kw).
Just as sirds is considered native coming from ḱ.

Hope it settles the matter.

This is certainly true. One issue you should consider is that you do have words in Balto-Slavic in particular that must be assumed to be loanwords because the cognate in other branches of IE (e.g. Indo-Iranic), or sometimes different words from the same root in Balto-Slavic have a *g or a *k despite the fact that they should have a fricative.

 

[arvistro]

This is certainly true. One issue you should consider is that you do have words in Balto-Slavic in particular that must be assumed to be loanwords because the cognate in other branches of IE (e.g. Indo-Iranic), or sometimes different words from the same root in Balto-Slavic have a *g or a *k despite the fact that they should have a fricative.

I dont argue that. In most cases seeing k sound where PIE must have had palatovelar means we are dealing with loan... or irregularity as in asmens/akmens duo.

What I say, is if PIE form had k or kw, then Balto-Slavic or Indo-Iranian k (not s) is expected. And there is ton of such words.

 

[MOESAN]

OK for words in Kw-, but I doubt for words in simple K- except before U, O); doubt, not certitude - I have french oil ialects and official language at hand -

 

[arvistro]

OK for words in Kw-, but I doubt for words in simple K- except before U, O); doubt, not certitude - I have french oil ialects and official language at hand -

I know what you mean
Before I, E indeed k might become soft(er) and become s. There are some words who survived into proto Baltic as geXXX, but they indeed seem to come from gw or gh initially.

I guess they get over it by simply assuming before i or e it was already ḱ in PIE.