zompist bboard

Is there anywhere where /kw/ (i.e. /k/ + /w/) develops differently from /k_w/ (i.e. labialised /k/)?

yes, *kwon "dog" turns palatal in the satem languages and stays /k/ in the centum languages. That's the only root I know that has /kw/, but there are probably others.

Wasn't it *ḱwon though? And I think the Satem languages tended to get rid of their labiovelars early on by merging them with their corresponding velars (correct me if I'm wrong.) It makes sense that if *kw>km, then *ḱw>ḱ.

Not an easy question. What dhokarena quotes as *ḱwon seems to be actually *ḱuwon-, and 'horse' *can* be actually an early loan (that spread among dialects when they were already slightly different). There are similar difficulties with most other examples IIRC.

I figured bricka was using a two-series model of PIE. If not, then the question might not be answerable, since plain velars in three-series models are very rare.

What dhokarena quotes as *ḱwon seems to be actually *ḱuwon-

Does PIE have the sequence -uv- anywhere else? It might just be that the two were allophonic variations.

Soap wrote:I figured bricka was using a two-series model of PIE. If not, then the question might not be answerable, since plain velars in three-series models are very rare.

Not really. *Medially* they were rather common. Although it's true that *initial* plain velars are rare.

EDIT: or rather, restricted to certain environments, like *Kr-

Pi= 3.141592653589793238462643383279502884197169399375105820974944592307816406286208998628034825342117067982148086513282306647093844609550582231725359408128481117450284102701938521105559644622948954930381964428810975665933446128475648233786783165271201909145648566923460348610454326648213393607260249141273724587006606315588174881520920962829254091715364367892590360011330530548820466521384146951941511609...

My favorite pies are apple, pumpkin and banana cream. Oh, you don't mean that kind of pie? Oh, bye!!!

mapking27 wrote:Pi= 3.14159

My favorite pies are apple, pumpkin and banana cream. Oh, you don't mean that kind of pie? Oh, bye!!!

I bet no-one's thought of that one before.

Nancy Blackett wrote:Is there anywhere where /kw/ (i.e. /k/ + /w/) develops differently from /k_w/ (i.e. labialised /k/)?

Yes, but they are few and far between.

PIE *kwe:p, *ke:wp "to boil, smoke"
Baltic kwapa (*k_w would have given **kapa)

PIE *kwotH "to boil, foam"
Sansk. kvathate (*k_w would have given **kathate)

PIE *kweit "to want, invite"
Baltic kwai

There are probably a handful of others.

Not to hijack the thread, but to hijack the thread: how did ablaut work in nouns of more than two syllables?