Page 1 of 4

h4 and h5

Posted: Sat Jul 03, 2010 10:17 am
by dhok
I know it's generally accepted that PIE had only three laryngeals, but I have heard that somebody has postulated h4 and h5. What's his justification, and what happened to them?

Re: h4 and h5

Posted: Sat Jul 03, 2010 10:40 am
by WeepingElf
Daquarious P. McFizzle wrote:I know it's generally accepted that PIE had only three laryngeals, but I have heard that somebody has postulated h4 and h5. What's his justification, and what happened to them?
There are seveal theories which add more laryngeals to the usual set of three. What such designations as "h4" or "h5" mean, depends on the author. AFAIK, there really is no good evidence for such extra laryngeals.

Re: h4 and h5

Posted: Sat Jul 03, 2010 10:41 am
by Morrígan
Daquarious P. McFizzle wrote:I know it's generally accepted that PIE had only three laryngeals, but I have heard that somebody has postulated h4 and h5. What's his justification, and what happened to them?
I don't know about *h5, but *h4 is intended to explain places where PIE appears to have *h2, but Hittite lacks an [h], while *h2 should leave [h]. *h5 may be the same, but for o-coloring, since *h3 also left [h] in Hittite, though under slightly restricted conditions, IIRC.

Re: h4 and h5

Posted: Sat Jul 03, 2010 4:47 pm
by Etherman
TheGoatMan wrote: I don't know about *h5, but *h4 is intended to explain places where PIE appears to have *h2, but Hittite lacks an [h], while *h2 should leave [h]. *h5 may be the same, but for o-coloring, since *h3 also left [h] in Hittite, though under slightly restricted conditions, IIRC.
That's exactly why *h5 is proposed. There's also been *h6 proposed which has the same coloring properties as *h1 but appears in HIttite as <h>.

Posted: Sat Jul 03, 2010 5:49 pm
by Travis B.
From the sound of all this, would it not be simpler to just propose that there had been some kind of irregular shifting or elision of such somewhere in what could be called pre-Anatolian, rather than assuming, in a quasi-Neogrammarian fashion, completely regular sound changes between Anatolian and PIE proper? Having six laryngeals sounds like an overly complex phonological system just to explain the differences between the rest of IE and Anatolian, and that these extra three laryngeals are paired with the other three laryngeals but differ only with regard to their fate in Anatolian make them seem like only a contrivance to avoid positing irregular phonological changes which likely had no actual phonological reality.

Posted: Sat Jul 03, 2010 7:36 pm
by TaylorS
Travis B. wrote:From the sound of all this, would it not be simpler to just propose that there had been some kind of irregular shifting or elision of such somewhere in what could be called pre-Anatolian, rather than assuming, in a quasi-Neogrammarian fashion, completely regular sound changes between Anatolian and PIE proper? Having six laryngeals sounds like an overly complex phonological system just to explain the differences between the rest of IE and Anatolian, and that these extra three laryngeals are paired with the other three laryngeals but differ only with regard to their fate in Anatolian make them seem like only a contrivance to avoid positing irregular phonological changes which likely had no actual phonological reality.
It reminds me of the ridiculously complex phoneme inventories Russian Nostraticists propose for Proto-Nostratic, mostly, IMO, because they are insistent that Afro-Asiatic and Dravidian are Nostratic languages, which I think is a load of bull. Most of those reconstructed phonemes are a form of handwavium.

Posted: Sat Jul 03, 2010 7:46 pm
by Shm Jay
Why not postulate h6, h7, and h8, while we’re at it, and then we can put the words on a chessboard.

Posted: Sat Jul 03, 2010 8:24 pm
by Morrígan
Shm Jay wrote:Why not postulate h6, h7, and h8, while we’re at it, and then we can put the words on a chessboard.
Unfortunately, this seemed to be part Szemerényi's reasoning for saying there was only one laryngeal. To my knowledge, very few authors suggest that there were more than the three Laryngeals, as each extra one is invoked to explain progressively smaller numbers of irregularities.

I concede that there may be something of a case for *h4, but am unconvinced. The rest are dubious at best.

Re: h4 and h5

Posted: Sat Jul 03, 2010 8:30 pm
by Xephyr
TheGoatMan wrote:
Daquarious P. McFizzle wrote:I know it's generally accepted that PIE had only three laryngeals, but I have heard that somebody has postulated h4 and h5. What's his justification, and what happened to them?
I don't know about *h5, but *h4 is intended to explain places where PIE appears to have *h2, but Hittite lacks an [h], while *h2 should leave [h]. *h5 may be the same, but for o-coloring, since *h3 also left [h] in Hittite, though under slightly restricted conditions, IIRC.
I believe Gsandi's personal version of PIE has h4 and h5, maybe even h6 as well. If so, his site has justification. However, I'm not 100% sure on that, even though it would be utterly trivial for me to go to his website and check. Too lazy.

Posted: Sat Jul 03, 2010 9:14 pm
by Etherman
Travis B. wrote:From the sound of all this, would it not be simpler to just propose that there had been some kind of irregular shifting or elision of such somewhere in what could be called pre-Anatolian, rather than assuming, in a quasi-Neogrammarian fashion, completely regular sound changes between Anatolian and PIE proper? Having six laryngeals sounds like an overly complex phonological system just to explain the differences between the rest of IE and Anatolian, and that these extra three laryngeals are paired with the other three laryngeals but differ only with regard to their fate in Anatolian make them seem like only a contrivance to avoid positing irregular phonological changes which likely had no actual phonological reality.
I think this is why *h5 and *h6 never caught on. There were very few examples of them. As far as I know they can be explained without recourse to irregularities. For example one of them is the word *me:, "to meaure". Hittite has a laryngeal, so this would point to *h6. The explanation that I've seen is that PIE had original long vowels, *e: and *o:. So the proper reconstruction would be *me:h2 (I'm pretty sure it's *h2, but I'm going from memory here). Long vowels aren't colored by laryngeal which is why this isn't *a:.

Posted: Sat Jul 03, 2010 9:19 pm
by Etherman
TaylorS wrote:It reminds me of the ridiculously complex phoneme inventories Russian Nostraticists propose for Proto-Nostratic, mostly, IMO, because they are insistent that Afro-Asiatic and Dravidian are Nostratic languages, which I think is a load of bull. Most of those reconstructed phonemes are a form of handwavium.
A few weeks ago I would have agreed about Afro-Asiatic and Dravidian not being part of Nostratic, but I've made a discovery that's caused me to rethink my objections. But that's a subject for another thread. I absolutely agree that their phoneme inventories are ridiculous. I think a lot of those phonemes disappear once you consider conditioning environments and consonant clusters.

Posted: Sat Jul 03, 2010 9:21 pm
by Etherman
Shm Jay wrote:Why not postulate h6, h7, and h8, while we’re at it, and then we can put the words on a chessboard.
Szemerenyi mentions one hypothesis that has about a dozen laryngeals. One can hardly take such a hypothesis seriously.

Re: h4 and h5

Posted: Sat Jul 03, 2010 9:36 pm
by Etherman
Xephyr wrote: I believe Gsandi's personal version of PIE has h4 and h5, maybe even h6 as well. If so, his site has justification. However, I'm not 100% sure on that, even though it would be utterly trivial for me to go to his website and check. Too lazy.
He has six, but two of these are labialized versions of *h2 and *h3. As near as I can tell he posits these to explain instances of Hittite <hu> where other languages point to *w. So they are not the same as traditional *h4 and *h5. He does include an *h6 which is the same as *h1 except it appears in Hittite as <h>. I know of only 2 or 3 examples of this so I'm not convinced.

Posted: Sat Jul 03, 2010 10:07 pm
by The Unseen
Is the idea to project the laryngeals as separate phonemes? If so, then the more there is the more problematic it gets, but if it is phones or just theoretical gaps then I don't see a problem. I'm not familiar with PIE, but when I say theoretical gap, I mean positing a sound to fill something that would not make sense according to regular sound change. That does not mean that that sound is necessarily separate, but it is more conservative to suggest that, according to the method of finding regular sound changes, the method leads us to require to posit a new sound, even if probabilistically it was not actually a separate sound.

Posted: Sat Jul 03, 2010 11:00 pm
by TaylorS
Etherman wrote:
TaylorS wrote:It reminds me of the ridiculously complex phoneme inventories Russian Nostraticists propose for Proto-Nostratic, mostly, IMO, because they are insistent that Afro-Asiatic and Dravidian are Nostratic languages, which I think is a load of bull. Most of those reconstructed phonemes are a form of handwavium.
A few weeks ago I would have agreed about Afro-Asiatic and Dravidian not being part of Nostratic, but I've made a discovery that's caused me to rethink my objections. But that's a subject for another thread. I absolutely agree that their phoneme inventories are ridiculous. I think a lot of those phonemes disappear once you consider conditioning environments and consonant clusters.
A-A seems to have originated in NE Africa, which is why attempts to connect it to IE within the past 15,000 years seem foolish. That said, I do think PIE may have borrowed a lot of vocab from Proto-Semitic, but I don't know enough about A-A languages to assess Glenn Gordon's ideas about Proto-Semitic borrowings into PIE (though *septm has to come from PS *sabbatum, it just has to!).

Posted: Sun Jul 04, 2010 6:04 am
by sangi39
I remember seeing one reconstruction online somewhere which included a bilabial or labiodental fricative and shifted the plain velars back to uvular and the palatalised velars back to velar leaving the labiovelars as they are with corresponding voiceless fricatives for each as well as a glottal plosive:

/p t k kʷ q/
/b d g gʷ ɢ/
/p' t' k' k'ʷ q' ʔ/
/ɸ s x xʷ χ h/
/m n/
/r l/
/w j/

/ɸ/ and /xʷ/ were proposed as two varients of *h3 with /χ/ and /x/ as two varients of *h2 and /ʔ/ and /h/ as two varients of *h1 although I can't remember whether they were suggesting they were allophones of a single phoneme which underwent different sound changes or six distinct phonemes which later merged into three.

IIRC the main point was that within the Anatolian branch, the six sounds "coloured" the vowels as indicated by their current labelling but one was deleted while the other was retained. I can't remember exactly what was said but it was something along the lines of, with h3 for example, that either /ɸ/ or /xʷ/ was deleted after vowel-colouring occurred while the non-deleted one was retained simply as ḫ. The same would then be true of /χ/ and /x/ where one is deleted and the other retained and so on for /h/ and /ʔ/. In other branches, however, the two sounds "merged" and were both subsequently deleted after vowel-colouring.

Personally I'm still a bit iffy on exactly what the laryngeals were and how many of them there might have been but I have a hard time imagining that if there were six of them that they'd all be between the velar and glottal POAs or that they'd all be fricatives.

Posted: Sun Jul 04, 2010 12:35 pm
by Etherman
sangi39 wrote: Personally I'm still a bit iffy on exactly what the laryngeals were and how many of them there might have been but I have a hard time imagining that if there were six of them that they'd all be between the velar and glottal POAs or that they'd all be fricatives.
Nobody really knows how they were pronounced. I've seen them variously described as glottals, velars, uvulars, and pharyngeals, with or without voice and/or labialized distinctions. Usually they're assumed to be fricatives and/or stops, but Bomhard reconstructs two of them as affricates.

There are certain indications that *h3 was voiced. Unlike *h1 and *h2, *h3 did not cause aspiration in Indo-Iranian so *h3 was probably a stop while *h1 and *h2 were probably voiceless fricatives. If the Glottalic Theory is correct (but I don't think it is) then maybe *h3 is /?/ because it would glottalize preceding voiceless stops which correspond to traditional plain voiced stops.

Posted: Sun Jul 04, 2010 12:58 pm
by alice
At one time there were supposed to have been no less than ten laryngeals.

Posted: Sun Jul 04, 2010 7:02 pm
by TaylorS
sangi39 wrote:I remember seeing one reconstruction online somewhere which included a bilabial or labiodental fricative and shifted the plain velars back to uvular and the palatalised velars back to velar leaving the labiovelars as they are with corresponding voiceless fricatives for each as well as a glottal plosive:

/p t k kʷ q/
/b d g gʷ ɢ/
/p' t' k' k'ʷ q' ʔ/
/ɸ s x xʷ χ h/
/m n/
/r l/
/w j/

/ɸ/ and /xʷ/ were proposed as two varients of *h3 with /χ/ and /x/ as two varients of *h2 and /ʔ/ and /h/ as two varients of *h1 although I can't remember whether they were suggesting they were allophones of a single phoneme which underwent different sound changes or six distinct phonemes which later merged into three.

IIRC the main point was that within the Anatolian branch, the six sounds "coloured" the vowels as indicated by their current labelling but one was deleted while the other was retained. I can't remember exactly what was said but it was something along the lines of, with h3 for example, that either /ɸ/ or /xʷ/ was deleted after vowel-colouring occurred while the non-deleted one was retained simply as ḫ. The same would then be true of /χ/ and /x/ where one is deleted and the other retained and so on for /h/ and /ʔ/. In other branches, however, the two sounds "merged" and were both subsequently deleted after vowel-colouring.

Personally I'm still a bit iffy on exactly what the laryngeals were and how many of them there might have been but I have a hard time imagining that if there were six of them that they'd all be between the velar and glottal POAs or that they'd all be fricatives.
This is interesting, thanks! I also think that the "plain velars" were uvulars and the "palatovelars" were velar, and this originated out of a merger of Proto-Europic /a/ and /@/ in early PIE (which is why I think Proto-Europic had a 4-vowel system, not a 3-vowel one).

Posted: Mon Jul 05, 2010 12:20 am
by Basilius
TaylorS wrote:It reminds me of the ridiculously complex phoneme inventories Russian Nostraticists propose for Proto-Nostratic, mostly, IMO, because they are insistent that Afro-Asiatic and Dravidian are Nostratic languages, which I think is a load of bull. Most of those reconstructed phonemes are a form of handwavium.
Etherman wrote:I absolutely agree that their phoneme inventories are ridiculous.
Interesting. "Russian Nostraticists" who included Afro-Asiatic and Dravidian in Nostratic might refer to Illich-Svitych and Dolgopolsky. Whose reconstructed phoneme inventory is "ridiculously complex", I wonder? Or is this an example of how North American universities teach people to quote their sources?

Also, will it make things too complex for some racists' brains learning that Illich-Svitych was an ethnic Ukrainian, and Dolgopolsky is a pureblooded Jew and Israeli?
TaylorS wrote:A-A seems to have originated in NE Africa, which is why attempts to connect it to IE within the past 15,000 years seem foolish.
Before calling something foolish, you might consider the idea that when you measure time in KY's, it is a bit bizarre to blame people for not counting *hundreds* of kilometers. The distance is in fact ridiculous, if you look at a map; it's much smaller than that between your current home and the area where your ancestral dialect was spoken a few *hundreds* of years ago, or between that area and where PIE was spoken just *six* KY ago. To supply you with some examples which you may not wish to accept from somebody who is so shamelessly Russian (and angry at you).
TaylorS wrote:[...](though *septm has to come from PS *sabbatum, it just has to!).
You missed a *pharyngeal* in your Proto-Semitic, I think. Foolish sounds, they're always too many and in ridiculous inventories, and nobody cares about them anyway.
sangi39 wrote:I remember seeing one reconstruction online somewhere which included a bilabial or labiodental fricative and shifted the plain velars back to uvular and the palatalised velars back to velar leaving the labiovelars as they are with corresponding voiceless fricatives for each as well as a glottal plosive:

/p t k kʷ q/
/b d g gʷ ɢ/
/p' t' k' k'ʷ q' ʔ/
/ɸ s x xʷ χ h/
/m n/
/r l/
/w j/

/ɸ/ and /xʷ/ were proposed as two varients of *h3 with /χ/ and /x/ as two varients of *h2 and /ʔ/ and /h/ as two varients of *h1 although I can't remember whether they were suggesting they were allophones of a single phoneme which underwent different sound changes or six distinct phonemes which later merged into three.

IIRC the main point was that within the Anatolian branch, the six sounds "coloured" the vowels as indicated by their current labelling but one was deleted while the other was retained. I can't remember exactly what was said but it was something along the lines of, with h3 for example, that either /ɸ/ or /xʷ/ was deleted after vowel-colouring occurred while the non-deleted one was retained simply as ḫ. The same would then be true of /χ/ and /x/ where one is deleted and the other retained and so on for /h/ and /ʔ/. In other branches, however, the two sounds "merged" and were both subsequently deleted after vowel-colouring.
Really and absolutely fascinating stuff! Please, try to recall the souce!
sangi39 wrote:Personally I'm still a bit iffy on exactly what the laryngeals were and how many of them there might have been but I have a hard time imagining that if there were six of them that they'd all be between the velar and glottal POAs or that they'd all be fricatives.
...in a language where consonant labialization was in principle phonemic (on velars).
Etherman wrote:Nobody really knows how they were pronounced. I've seen them variously described as glottals, velars, uvulars, and pharyngeals, with or without voice and/or labialized distinctions.
It is interesting to observe how nobody has the guts to say "velars". Loans to/from Hittite don't seem to point to any PoA further back, AFAICT.

Also, do any reconstructions try to account for the fact that Hittite, actually, had two distinct "laryngeals"?

Posted: Mon Jul 05, 2010 12:33 am
by Morrígan
Basilius wrote:Also, will it make things too complex for some racists' brains learning that Illich-Svitych was an ethnic Ukrainian, and Dolgopolsky is a pureblooded Jew and Israeli?
While this is interesting, I really don't think this disqualifies them from being "Russian Nostraticists", in which "Russian" is to be taken as identifying a school of thought, rather than a nationality.

Nice touch on throwing the word "racist" around, btw.
Basilius wrote:Foolish sounds, they're always too many and in ridiculous inventories, and nobody cares about them anyway.
Are you always this dickish?
Basilius wrote:Also, do any reconstructions try to account for the fact that Hittite, actually, had two distinct "laryngeals"?
Unless am completely misunderstanding you, yes. They are called *h2 and *h3.

Posted: Mon Jul 05, 2010 1:31 am
by TaylorS
Geez, I think I struck a nerve. :roll:

A-A clearly originated in Africa, while my version of Nostratic originated in Central Asia around 12,000 to 14,000 years ago. There is no evidence, archeological or genetic, of a movement of people between Central Asia and NE Africa 14,000 years ago

Posted: Mon Jul 05, 2010 8:38 am
by Travis B.
Basilius wrote:Interesting. "Russian Nostraticists" who included Afro-Asiatic and Dravidian in Nostratic might refer to Illich-Svitych and Dolgopolsky. Whose reconstructed phoneme inventory is "ridiculously complex", I wonder? Or is this an example of how North American universities teach people to quote their sources?
Basilius wrote:Also, will it make things too complex for some racists' brains learning that Illich-Svitych was an ethnic Ukrainian, and Dolgopolsky is a pureblooded Jew and Israeli?
Basilius wrote:Before calling something foolish, you might consider the idea that when you measure time in KY's, it is a bit bizarre to blame people for not counting *hundreds* of kilometers. The distance is in fact ridiculous, if you look at a map; it's much smaller than that between your current home and the area where your ancestral dialect was spoken a few *hundreds* of years ago, or between that area and where PIE was spoken just *six* KY ago. To supply you with some examples which you may not wish to accept from somebody who is so shamelessly Russian (and angry at you).
Basilius wrote:You missed a *pharyngeal* in your Proto-Semitic, I think. Foolish sounds, they're always too many and in ridiculous inventories, and nobody cares about them anyway.
Basilius wrote:Really and absolutely fascinating stuff! Please, try to recall the souce!
Basilius wrote:...in a language where consonant labialization was in principle phonemic (on velars).
Basilius wrote:It is interesting to observe how nobody has the guts to say "velars". Loans to/from Hittite don't seem to point to any PoA further back, AFAICT.

Also, do any reconstructions try to account for the fact that Hittite, actually, had two distinct "laryngeals"?
:roll:

Posted: Mon Jul 05, 2010 12:16 pm
by Kai_DaiGoji
Basilius wrote:Also, will it make things too complex for some racists' brains learning that Illich-Svitych was an ethnic Ukrainian, and Dolgopolsky is a pureblooded Jew and Israeli?
Seriously, a discussion of PIE laryngeals is turning into a flamewar? Now I've seen everything.

Posted: Mon Jul 05, 2010 1:15 pm
by MrKrov
You have never seen everything on the internet.