« Linguistics, Archaeology and Indiana Jones | Main | Kortlandt's Consonant Gradation Part II »



Glen Gordon

I don't think you're being clear about how this process caused voicing of *all* stops in a root. This is evidently then not the same as what we find in Uralic (or specifically BFS).


[this is good] I believe Kortlandt actually assumes no voiced consonants at all in his Uralic reconstruction. But I must say I know little about it.

But couldn't the 'plain voiced' stops come from the Uralic then?

As far as I know Uralic doesn't have a three way consonant opposition does it?

And I am actually not yet clear how this caused voicing in all of the stops. There seems to be a process of voicing in relation to stress, and I'm not sure whether you can be as specific about it as Kortlandt claims to be, or whether his reconstruction is even correct.

For now I will just be looking for identifiable possible pairs of voiced and unvoiced roots that seem to be related in accent.

After that I'll see if it is possible for me to write something meaningful on the actual relation between stress and voicing.

Glen Gordon

Phoenix: "As far as I know Uralic doesn't have a three way consonant opposition does it?"

True. The theorized "Indo-Uralic" 3-way stop contrast would have collapsed in Uralic. However if the existence of *-ð- is any indication, it would appear that voiced stops may have been preserved in word-medial positions as fricatives. Just a thought to chew on, at least.

"There seems to be a process of voicing in relation to stress, [...]"

Exactly, but if it's related to stress, the phenomenon can't explain an alleged alternation between *kerp- and *gʰrebʰ-. Afterall, if we push this reasoning to its conclusion, Syncope would predict earlier *kérVp- and *gʰVrébʰ-. Thus, the purported voicing of these stops is entirely independent of stress.


Yes I've been struggling with that too. And that's exactly the reason why I think it's worth having a thorough look at such instances, just to see what is going on.

It is of course weird that both stops alternate due to the jumping stress. While with the Sanskrit hrd- it makes more sense, since the root is actually in zero grade, That's what you'd want to see with the schwebeablaut as well. All in all, I'll be reading up on both schwebeablaut and trying to find instances of these 'voicing pairs' and see whether there is a relation between the two.

And if there is a relation, I'll go see how the hell the relation works. Because 'The root undergoes lenition when the accent is in the second syllable of the root' might be describing the process correctly but makes absolutely no sense from a phonetic point of view.

I think it's enough to write about for a scripture anyway. It's pre-indo-european, it's related to syncope and accent shifts, but doesn't directly ask me to completely redescribe the Indo-European system as we currently know it ;)

Oliver S

[this is good] Perhaps you mentioned this already in the previous post you refer to, but *ghrebh- and *kerp are reminiscent of the pair *kap- and *ghabh- (if one accepts these root-forms - LIV has *keh2p- and *gheh2b-). The h- of Skt. hrd is usually explained as contamination with gherd- 'intestine', isn't it? But deriving *kerd- and gherd- from a single ancestor seems a step too far! BTW as for schwebeablaut, why can't we explain this merely as the result of original CreC CrC roots, with secondary CerC forms created as back-formations from the zero-grade? Anyway, congratulations on the blog (and to you too Glen!)


Yes they are reminiscent of the same effect, though I am indeed inclined to reconstruct laryngeals rather than *a. ;-)

and yesh I believe they indeed interpet hrd as a contamination gherd. But personally I'm not convinced. Not sure if we should call them the same root originally. A lot of odd stuff goes on with The word for Hard in indo-iranian. Having, hrd, śrad- and Avestan having both zrad- and zərəd- I'm not sure what to reconstruct.

I'm a little sceptical at the moment of reducing all voiced aspirates as variants of voiceless aspirates (as Kortlandt does) because there's pairs like kerp- ~ ghrebh- which noone in their right mind will try to link semantically. It will all become clearer once I have my database done, and then you'll hear about my findings.

I suggest you read Kortlandts article on consonant gradation, it's very interesting, though difficult to follow.

Bottomline is, I'm not sure what I should conclude from these pairs, but one thing is sure, it's worth surveying ;-)

Now the other subject, the Schwebeablaut explanation of back-formations fro zero-grade does not appeal to me. Especially since these words have retained their full grade in some forms and then a schwebeablaut-variant full grade in others. Why would they come up with a new full grade if they already have a fullgrade?

I'm more inclined to reconstruct disyllabic roots in an earlier stage which due to accent shifts and syncope yielded several variants. This is also in the light of soom root nouns that seem to have schwebeablaut like *ḱerd-, but also dʰ(e)gʰ(e)m 'earth' (Hittite has tekan which points to dʰegʰm, while greek has χθων which points to dʰgʰoms (Is that a labio-velar rather than a velar btw? I can't remember from the top of my head)

Besides that, thanks for the congratulations on the blog and thanks for the comment.

Glen Gordon

Oliver S: "I guess Kortlandt, Glen and Rasmussen would each say that they’ve already succeeded in doing this ;-)"

Now, now. It's illogical to put words in other people's mouths. Speaking for myself, my work is inspired on the concepts of logic, extrapolation and evolution. So as far as I'm concerned, my theory (or any theory for that matter) is forever a work in progress that attempts to optimally solve linguistic riddles in a manner similar to approximating a circle with polygons as accurately as possible. If I ever felt that I had it all figured it out, I would have no reason to continually updating my pdfs and such on my blog, now would I? As I always say, if you can come up with a better theory, by all means! :-) If not, I guess these theories created by others are the best you've got so far.

Glen Gordon

Oliver S: "That’s great! I don’t know where to start with the Kortlandt article, it’s hard to assess such things..."

Perhaps I should comment on this as well. It's not really that "hard" to assess the merit of theories if you use Logic and the principle of Occam's Razor as a guide. If one uses no method at all except perhaps one's feelings as a guide, then I suppose it is rather difficult indeed.

The comments to this entry are closed.