The Great Proto-Indo-European Thread

Discussion of natural languages, or language in general.
Fixsme
Sanci
Sanci
Posts: 32
Joined: Fri Jan 02, 2015 1:45 pm
Location: Paris, France

Re: The Great Proto-Indo-European Thread

Post by Fixsme »

Hmmm. Statgasm!! :-D

I need to read this article in detail. But I would not be that sure about the conclusion. Most of their results say "it looks Kurgan but Anatolian could be true also" a bit like Atkinson and Gray one, "Anatolian is true but it could be also Kurgan." (p. 27) And the results are really unstable, change one parameter in the models and you switch from Kurgan to Anatolian.

Is there a third solution? Between 8000 BP and 6500BP?

By the way, is using "Ancient Greek" correct? Since it's probably the Attic dialect of the 5th century BC. Do you think other Greek dialects should be used?
Could Mycenian greek have been used? I feel so sorry for the Greek Branch...

User avatar
linguofreak
Lebom
Lebom
Posts: 123
Joined: Mon Jun 13, 2005 10:39 pm
Location: Somewhere
Contact:

Re: The Great Proto-Indo-European Thread

Post by linguofreak »

Valdeut wrote:
linguofreak wrote:Anybody want to poke holes in this?
Correct me if I'm misunderstanding what you're proposing. But the core idea seems to be that PIE had a nice symmetrical system of four series based on two contrasts: plosive–fricative and voiceless–voiced. In particular, the proposal seems to be that the voiced aspirates and ”laryngeals” were two symmetrical series distinguished only in that one of them was voiced and the other voiceless.
More or less. What led me to the idea was that the "voiced aspirates" as voiced aspirates seem highly improbable given the lack of voiceless aspirates, and that the similarities between Grimm's law and the second Germanic sound shift suggest that they might well have been fricatives. Meanwhile, the laryngeals could be syllabic, so they were likely either fricatives or approximants. The three laryngeals we know of colored vowels in a way that suggests their points of articulation matched those of the post-alveolar stops in PIE. We already know of a labiovelar approximant other than h3 (w), which suggests they were fricativesfricatives. To not collide with the "voiced aspirates", if those were fricatives, there would have to be a voicing distinction. A similar argument lead to grouping /s/ with the laryngeals.

If there is no other patterning between ”laryngeals” and voiced aspirates, why would you think that there should be any sort of symmetry just in terms of what segments exists? Why both PL and KL?
As mentioned above, the thing that put it into my head is that both look like they should be fricatives. The voicing distinction came from the fact that a distinction was needed to keep the series from colliding if both were fricatives. Once I had determined that s would have the same voicing as the laryngeals in this case, things looked so symmetric except for PL that I thought PL was unlikely to be empty if both the laryngeals and voiced aspirates were fricatives.
And traditional *s and *dʰ, did they really differ only in that one was [z] and one was [s]? So the word *misdʰós would have had either a cluster [sz] or [zs]? And what about the cluster *sd and *st? They have very different outcomes in Sanskrit compared to *dʰt.
The voicing assimilation issue, especially between s and z, is where I myself see the biggest problem with the idea.

I did just think of another idea for the identity of the laryngeals: could the laryngeals have been nasals? Out of the five places of articulation in PIE, the known nasals cover the front two, and the laryngeals cover the back three. So each serious fits into the gap in the other. Both are known to have been syllabic in PIE. On the other hand, the laryngeals became h in the languages that preserved them long enough for them to be historically attested, and that seems more likely for a fricative than a nasal.

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: The Great Proto-Indo-European Thread

Post by KathTheDragon »

I highly doubt that *dʱ could have been /z/, given its reflex of *d /ð/ in Proto-Germanic.

User avatar
Hallow XIII
Avisaru
Avisaru
Posts: 846
Joined: Sun Nov 04, 2012 3:40 pm
Location: Under Heaven

Re: The Great Proto-Indo-European Thread

Post by Hallow XIII »

bashkir
陳第 wrote:蓋時有古今,地有南北;字有更革,音有轉移,亦勢所必至。
R.Rusanov wrote:seks istiyorum
sex want-PRS-1sg
Read all about my excellent conlangs
Basic Conlanging Advice

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: The Great Proto-Indo-European Thread

Post by hwhatting »

@ linguofreak: can you give a full table with all the equivalents in tradtional notation? Perhaps it's just me, but I don't fully understand which constituents of your system give which results in the traditional reconstruction.

User avatar
Pogostick Man
Avisaru
Avisaru
Posts: 894
Joined: Sat Jul 25, 2009 8:21 pm
Location: Ohio

Re: The Great Proto-Indo-European Thread

Post by Pogostick Man »

linguofreak wrote:I did just think of another idea for the identity of the laryngeals: could the laryngeals have been nasals? Out of the five places of articulation in PIE, the known nasals cover the front two, and the laryngeals cover the back three. So each serious fits into the gap in the other. Both are known to have been syllabic in PIE. On the other hand, the laryngeals became h in the languages that preserved them long enough for them to be historically attested, and that seems more likely for a fricative than a nasal.
Rhinoglottophilia. *ŋ > x is attested somewhere in Mayan.
(Avatar via Happy Wheels Wiki)
Index Diachronica PDF v.10.2
Conworld megathread

AVDIO · VIDEO · DISCO

Valdeut
Sanci
Sanci
Posts: 56
Joined: Sun Mar 02, 2014 12:16 pm
Location: Sweden
Contact:

Re: The Great Proto-Indo-European Thread

Post by Valdeut »

linguofreak wrote:More or less. What led me to the idea was that the "voiced aspirates" as voiced aspirates seem highly improbable given the lack of voiceless aspirates, and that the similarities between Grimm's law and the second Germanic sound shift suggest that they might well have been fricatives. Meanwhile, the laryngeals could be syllabic, so they were likely either fricatives or approximants. The three laryngeals we know of colored vowels in a way that suggests their points of articulation matched those of the post-alveolar stops in PIE. We already know of a labiovelar approximant other than h3 (w), which suggests they were fricativesfricatives. To not collide with the "voiced aspirates", if those were fricatives, there would have to be a voicing distinction. A similar argument lead to grouping /s/ with the laryngeals.
Either the voiced aspirates or the laryngeals could have been fricatives, but not really both at the same place of articulation. Voicing alone is not enough of a distinction given that, from what I can tell, there are no alternations between voiced aspirates and laryngeals (but perhaps it's just that no one has been looking?). In contrast, you find a lot of voicing assimilation of stops in the history of the IE languages.

The reflexes of PIE voiced aspirates can very well have been voiced fricatives in later Pre-Proto-Germanic, meaning that they remained unchanged by Grimms Law before later hardening to stops to different extent in different daughter languages. Whatever the voiced aspirates really where in PIE, they don’t need to have remained unchanged until Grimm’s Law. But at that point, they wouldn't contrast with any voiceless fricatives except *s (and the reflex of *dʰ is then more likely to have been [ð] than [z]).

Almost all suggestions I've seen for pronunciation of the laryngeals are dorsal, radical or glottal fricatives or a glottal stop. It is commonly observed that laryngeals mostly pattern with *s in terms of places they can occupy in a root. So apart from the extra labial laryngeal, you're system is quite standard in that respect.
linguofreak wrote:I did just think of another idea for the identity of the laryngeals: could the laryngeals have been nasals? Out of the five places of articulation in PIE, the known nasals cover the front two, and the laryngeals cover the back three. So each serious fits into the gap in the other. Both are known to have been syllabic in PIE..
"Syllabic laryngeals” are quite different from syllabic resonants, though. See this for more information (you may have to download the PDF):
https://www.academia.edu/4165302/The_Ph ... inal_draft_

Also, if they were nasal consonants, you would probably expect there to be a tendency towards place assimilation before following stops. If not in PIE itself at least in the history of some daughter branches.

User avatar
linguofreak
Lebom
Lebom
Posts: 123
Joined: Mon Jun 13, 2005 10:39 pm
Location: Somewhere
Contact:

Re: The Great Proto-Indo-European Thread

Post by linguofreak »

hwhatting wrote:@ linguofreak: can you give a full table with all the equivalents in tradtional notation? Perhaps it's just me, but I don't fully understand which constituents of your system give which results in the traditional reconstruction.

Code: Select all

Phonetic
     P     T     K    Q    W
S [p]  [t]   [k]  [q]  [qʷ]
Z [b]  [d]   [g]  [ɢ]  [ɢʷ]
F [f]  [s]   [x]  [χ]  [χʷ]
V [v]  [z]   [γ]  [ʁ]  [ʁʷ]

Phonemic

     P     T     K     Q     W
S  [p]   [t]  [ḱ]   [k]   [kʷ]
Z  [b]   [d]  [ǵ]   [g]   [gʷ]
A  [bʰ]  [dʰ] [ǵʰ]  [gʰ]  [gʷʰ]
L  [h₁]  [s]  [h₁]  [h₂]  [h₃]

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: The Great Proto-Indo-European Thread

Post by KathTheDragon »

Note that I pair up the two fricative series the other way around.

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: The Great Proto-Indo-European Thread

Post by hwhatting »

KathAveara wrote:Note that I pair up the two fricative series the other way around.
Yes, that makes more sense (and I note linguofreak leaves that possibility open). But still, I have a few quibbles:
The distinction between "phonemic" and "phonetic is used in a strange way here. If it means, e.g., the Media Aspirata (MA) were underlying fricatives but realised as voiced stops with aspiration, what sense does the distinction even make here? Normally, the phonemic value ought to at least be one of the phonetic realisations, i.e., one should be able to indicate under which circumstances /v/ (or /f/) was realised as [v], as [ bh], as [f] etc. If it means something like "the MA were aspirated voiced stops alright, but 'underlyingly' they were fricatives, because that allows us to pair them symmetrically with the laryngeals", well, this is useless - no phonological system is obliged to be symmetrical. The same concerns linguofrek's two /h1/'s - yes, it creates a nice symmetry, but the goal should be to reconstruct a system for which there is evidence, not a nice symmetrical table. Let's keep in mind that the voiced stop series also has a glaring gap, with /b/ being at best marginal; most examples of [ b] that we have were actually allophones of /p/ until the loss of /h3/, and roots like *bel- may very well be loans. So, if there would be corroborating evidence for that nice symmetrical system (like "labial" /h1/ behaving differently from "velar" /h1/, or the laryngeals (+ /s/) behaving and developing in a similar way to the MA, that would mean something; but being symmetrical doesn't make a reconstruction better than being asymmetrical.
/h3/ is normally reconstructed as voiced, based on the voicing effect it has, while the other laryngeals are not. So putting it in the same series as the other laryngeals requires an explanation either way - if all laryngeals were voiced fricatives, why is there no evidence for voicedness of /h1/ and /h2/? If they were all unvoiced, why does /h3/ have the voicing effect?
In the languages that split off earliest, the MA behave like stops. They become fricatives only in two branches, Italic (where this happens only to the MAs, which would support linguofreak's system) and in Germanic, where they pair with the plain unvoiced stops. In most branches they merge with the voiced stops. For me, all this makes most sense if they were indeed stops; the fricativisation in Italic (and Germanic) can be easily explained from their aspiration.
As a meta-remark, it's of course always tempting to try making the excentric PIE stop system more "normal" and symmetrical, but exactly the point that it has been changed in all daughter languages is a good argument for it actially having been excentric and unstable.

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: The Great Proto-Indo-European Thread

Post by KathTheDragon »

Given that unusual systems like PIE's are so unstable, they must have arisen from much more normal systems (which is likely what LF is creating, and certainly what my system is). My rationale is that once weird things happened to the voiceless fricatives (in this case, a merger between **ɸ and **xʷ, followed by debuccalisation to *h (corresponding to *h1), you have an excentic and unstable system, which was fixed in most branches by hardening the voiced fricatives to stops (plus aspiration in Greek and Indo-Iranian), or creating new voiceless fricatives in Germanic and Italic. Note that Germanic and Italic are two of the west-most branches of IE, so a family-wide change of voiced fricative > voiced stop could easily have missed just those two branches (though perhaps not entirely - observe that in Germanic, the voiced fricatives do have allophonic stop realisations). Ultimately, though, debating the phonetics of this series is pointless, as neither stance can be proven.

Edit:
hwhatting wrote:Let's keep in mind that the voiced stop series also has a glaring gap, with /b/ being at best marginal; most examples of [ b] that we have were actually allophones of /p/ until the loss of /h3/, and roots like *bel- may very well be loans. So, if there would be corroborating evidence for that nice symmetrical system (like "labial" /h1/ behaving differently from "velar" /h1/, or the laryngeals (+ /s/) behaving and developing in a similar way to the MA, that would mean something; but being symmetrical doesn't make a reconstruction better than being asymmetrical.
/h3/ is normally reconstructed as voiced, based on the voicing effect it has, while the other laryngeals are not. So putting it in the same series as the other laryngeals requires an explanation either way - if all laryngeals were voiced fricatives, why is there no evidence for voicedness of /h1/ and /h2/? If they were all unvoiced, why does /h3/ have the voicing effect?
I don't agree that *h₃ was voiced - the only example I've seen is in the reduplicated present of *peh₃: *pibh₃e-. But note that this can also be explained by word-initial devoicing of *b, which also offers an explanation of why *b is virtually unknown in initial position, but is somewhat less uncommon in root-final position. I also searched through the LIV for roots where a voiceless stop and *h₃ can come into contact, and I found one: *h₂eḱh₃, attested in Vedic Sanskrit, Old Norse, and Greek. This suggests that this voicing rule is bogus, invented to explain one peculiar alternation involving *b.

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: The Great Proto-Indo-European Thread

Post by hwhatting »

KathAveara wrote:Given that unusual systems like PIE's are so unstable, they must have arisen from much more normal systems (which is likely what LF is creating, and certainly what my system is).

If that's what LF is doing, then it's fine; it's also something I don't want to comment on much - after all, who knows how many series pre-PIE had that collapsed into the PIE system (you have four, may be there were five, or seven?); if we do not find internal evidence or outside confirmation (from Indo-Ugric, Nostratic, or whatever else grouping PIE may be descended of), this is unprovable, and symmetricity in itself isn't a proof. After all, the pre-system may have been asymmetric and riddled with gaps that led to its being sorted into the PIE system.
KathAveara wrote:I don't agree that *h₃ was voiced - the only example I've seen is in the reduplicated present of *peh₃: *pibh₃e-. But note that this can also be explained by word-initial devoicing of *b, which also offers an explanation of why *b is virtually unknown in initial position, but is somewhat less uncommon in root-final position.
Hmmm... there indeed don't seem to be too many good examples of voicing by H3, although I think I've seen some more examples of voicing by the Hoffmann suffix in the literature. Still, the root in *pib(H3)eti is clearly *peH3. Of course, you seem to argue that /b/ was always devoiced word-initially, so that the root may actually have been **beH(3), but until you give me more examples of such a development, that looks very much like special pleading as well. (And it's also incompatible with roots like *bel-, which in case of such a devoicing rule must belong to a later diachronic stratum).
I also searched through the LIV for roots where a voiceless stop and *h₃ can come into contact, and I found one: *h₂eḱh₃, attested in Vedic Sanskrit, Old Norse, and Greek.
Well, the laryngeal is actually doubtful in this root and only established based on Greek ἄκολος, a word concerning which all of Frisk, Beekes, and Chantraine even doubt of whether it belongs to that root and isn't maybe a loan. (And even if it belonged to that root, there would be other possibilities to explain the /o/ than by a laryngeal.)

So I concede that the evidence for H3 being voiced is limited, but OTOH I don't see a reason to throw away a good explanation for *pibeti and *abon- only because one wants to put all three laryngeals into one series for the sake of a neat phonemic symmetricity.
Ultimately, though, debating the phonetics of this series is pointless, as neither stance can be proven.
I agree.

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: The Great Proto-Indo-European Thread

Post by KathTheDragon »

An additional issue is that if *h₃ voiced a previous voiceless stop, wouldn't you expect *h₂ at the very least to have devoiced a previous voiced or breathy-voiced stop? But roots like *(s)kedh₂ and *bʱedʱh₂- clearly show that this isn't the case (and note that roots of this shape a fairly common.

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: The Great Proto-Indo-European Thread

Post by hwhatting »

KathAveara wrote:An additional issue is that if *h₃ voiced a previous voiceless stop, wouldn't you expect *h₂ at the very least to have devoiced a previous voiced or breathy-voiced stop? But roots like *(s)kedh₂ and *bʱedʱh₂- clearly show that this isn't the case (and note that roots of this shape a fairly common.
I don't see how one follows from the other (phonemes from different series often behave differently in that respect). But your example shows that /h1/ and /h2/ actually don't line up with /s/ in all respects, as devoicing (plus de-aspiration) of stops before /s/ is well established for PIE.

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: The Great Proto-Indo-European Thread

Post by KathTheDragon »

I'll concede that point. But then there comes the issue of when in PIE these processes actually date to, which is likely unrecoverable. But note that *s is already special in PIE - only *s participates in s-mobile. Based on that, you might expect *s to behave slightly differently to other present voiceless fricatives.

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: The Great Proto-Indo-European Thread

Post by Salmoneus »

KathAveara wrote:An additional issue is that if *h₃ voiced a previous voiceless stop, wouldn't you expect *h₂ at the very least to have devoiced a previous voiced or breathy-voiced stop? But roots like *(s)kedh₂ and *bʱedʱh₂- clearly show that this isn't the case (and note that roots of this shape a fairly common.
If, however, we were to imagine that the 'voiced' stops and fricatives in PIE were actually glottalised in some way, then suddenly that would make perfect sense. A glottalised *h3 could easily glottalise adjacent voiceless stops into voiceless ejectives, while leaving the true voice (/voiced aspirate) series unaffected (or perhaps allophonically glottalising them, with this allophony failing to become phonemic due to the absence of an existing glottalised voiced series). Meanwhile the unglottalised, voicing-neutral *h1, *h3 have no effect on neighbouring stops.

This would still leave it odd why *s did have an effect, but that's a problem with any theory, I'd have thought. Perhaps it could be said that *s became fixed as voiceless by default (and hence started devoicing preceding stops) later on?

Now if only there were some independent other reason to imagine that the 'voiced' stops were really glottalised...

[Regarding s-mobile: I thought that was considered more likely a morphophonemic process rather than a strictly phonological one? That ambiguity arose over initial *s due to the large number of words ending in *s.
In any case, are we sure there was no 'h-mobile'? I seem to recall that initial laryngeals before consonants only show up in some daughters, and that their presence is sometimes unpredictable? Could be a similar effect with *h2 at least, due to the collective/feminine ending?]
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
WeepingElf
Smeric
Smeric
Posts: 1630
Joined: Wed Mar 08, 2006 5:00 pm
Location: Braunschweig, Germany
Contact:

Re: The Great Proto-Indo-European Thread

Post by WeepingElf »

Salmoneus wrote:If, however, we were to imagine that the 'voiced' stops and fricatives in PIE were actually glottalised in some way, then suddenly that would make perfect sense. A glottalised *h3 could easily glottalise adjacent voiceless stops into voiceless ejectives, while leaving the true voice (/voiced aspirate) series unaffected (or perhaps allophonically glottalising them, with this allophony failing to become phonemic due to the absence of an existing glottalised voiced series). Meanwhile the unglottalised, voicing-neutral *h1, *h3 have no effect on neighbouring stops.
Thanks! An interesting idea which for some reason hasn't been posted here yet!
Salmoneus wrote:This would still leave it odd why *s did have an effect, but that's a problem with any theory, I'd have thought. Perhaps it could be said that *s became fixed as voiceless by default (and hence started devoicing preceding stops) later on?
PIE wasn't static in time, and the stop-devoicing effect of *s may belong to a later period (one in which the traditional stop system was in place) than the stop-"voicing" effect of *h3.
Salmoneus wrote:Now if only there were some independent other reason to imagine that the 'voiced' stops were really glottalised...
Well put - all we can say about the glottalic model right now is that it could correctly describe some time stage of PIE, but we yet lack the forceful evidence that it did.
Salmoneus wrote:[Regarding s-mobile: I thought that was considered more likely a morphophonemic process rather than a strictly phonological one? That ambiguity arose over initial *s due to the large number of words ending in *s.
In any case, are we sure there was no 'h-mobile'? I seem to recall that initial laryngeals before consonants only show up in some daughters, and that their presence is sometimes unpredictable? Could be a similar effect with *h2 at least, due to the collective/feminine ending?]
S-mobile indeed probably was a morphophonemic process of some sort rather than a purely phonological one. That weird *s may have been some sort of prefix (a demonstrative?). If we had no direct attestation of Arabic but only the loanwords it left in various European languages, we would have an "al-mobile" phenomenon in those loanwords (e.g. alchemy vs. chemistry). Of course, we know Arabic, and can say what this "al-mobile" was ;) I have hit upon a similar phenomenon which I call "alpha mobile" in my studies of substratum loanwords in western IE languages. Example: German Amsel 'blackbird' < *amasula vs. Latin merula 'blackbird' < *masula. I suspect the alpha mobile to have been an article in the hypothetical donor language.

But s-mobile is not the only evidence against *s forming a class with the laryngeals. Another, for instance, is the non-existence of a "syllabic *s". Yet, *s is of course a sibilant, and sibilants often behave differently than "flat" (non-sibilant) fricatives. So this still doesn't surely show that the laryngeals weren't voicess fricatives.

Initial laryngeals before consonants regularly show up as vowels in Greek and Armenian, and are lost everywhere else. I am currently not aware of any irregularities here that could be compared to s-mobile.
...brought to you by the Weeping Elf
Tha cvastam émi cvastam santham amal phelsa. -- Friedrich Schiller
ESTAR-3SG:P human-OBJ only human-OBJ true-OBJ REL-LOC play-3SG:A

User avatar
WeepingElf
Smeric
Smeric
Posts: 1630
Joined: Wed Mar 08, 2006 5:00 pm
Location: Braunschweig, Germany
Contact:

Re: The Great Proto-Indo-European Thread

Post by WeepingElf »

Regarding *h3: What kind of phoneme could have effected both glottalization of voiceless stops and backing [æ] to something like [ɑ] or [ɒ]?
...brought to you by the Weeping Elf
Tha cvastam émi cvastam santham amal phelsa. -- Friedrich Schiller
ESTAR-3SG:P human-OBJ only human-OBJ true-OBJ REL-LOC play-3SG:A

sirdanilot
Avisaru
Avisaru
Posts: 734
Joined: Sat Aug 18, 2007 1:47 pm
Location: Leiden, the Netherlands

Re: The Great Proto-Indo-European Thread

Post by sirdanilot »

WeepingElf wrote:Regarding *h3: What kind of phoneme could have effected both glottalization of voiceless stops and backing [æ] to something like [ɑ] or [ɒ]?
Isn't it pretty well known that *h3 is commonly regarded as a voiced labialized velar/uvular/pharyngeal fricative? The exact point of articulation is debatable as is anything when you are reconstructing so far back of course.

User avatar
WeepingElf
Smeric
Smeric
Posts: 1630
Joined: Wed Mar 08, 2006 5:00 pm
Location: Braunschweig, Germany
Contact:

Re: The Great Proto-Indo-European Thread

Post by WeepingElf »

sirdanilot wrote:
WeepingElf wrote:Regarding *h3: What kind of phoneme could have effected both glottalization of voiceless stops and backing [æ] to something like [ɑ] or [ɒ]?
Isn't it pretty well known that *h3 is commonly regarded as a voiced labialized velar/uvular/pharyngeal fricative? The exact point of articulation is debatable as is anything when you are reconstructing so far back of course.
That's a non-answer. I don't ask what *h3 is commonly regarded to be, but which kind of sound is likely to show the observed effects.
...brought to you by the Weeping Elf
Tha cvastam émi cvastam santham amal phelsa. -- Friedrich Schiller
ESTAR-3SG:P human-OBJ only human-OBJ true-OBJ REL-LOC play-3SG:A

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: The Great Proto-Indo-European Thread

Post by Zaarin »

WeepingElf wrote:
sirdanilot wrote:
WeepingElf wrote:Regarding *h3: What kind of phoneme could have effected both glottalization of voiceless stops and backing [æ] to something like [ɑ] or [ɒ]?
Isn't it pretty well known that *h3 is commonly regarded as a voiced labialized velar/uvular/pharyngeal fricative? The exact point of articulation is debatable as is anything when you are reconstructing so far back of course.
That's a non-answer. I don't ask what *h3 is commonly regarded to be, but which kind of sound is likely to show the observed effects.
The combined pharyngealization of [æ] and glottalization of consonants could suggest [ʔˤ]; or perhaps [ʔʷ], which is marginally attested in Tlingit.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

sirdanilot
Avisaru
Avisaru
Posts: 734
Joined: Sat Aug 18, 2007 1:47 pm
Location: Leiden, the Netherlands

Re: The Great Proto-Indo-European Thread

Post by sirdanilot »

WeepingElf wrote:
sirdanilot wrote:
WeepingElf wrote:Regarding *h3: What kind of phoneme could have effected both glottalization of voiceless stops and backing [æ] to something like [ɑ] or [ɒ]?
Isn't it pretty well known that *h3 is commonly regarded as a voiced labialized velar/uvular/pharyngeal fricative? The exact point of articulation is debatable as is anything when you are reconstructing so far back of course.
That's a non-answer. I don't ask what *h3 is commonly regarded to be, but which kind of sound is likely to show the observed effects.
I see no reason to speculate on sounds other than those commonly speculated as to the value of *h3, e.g. a labialized voiced fricative articulated somewhere in the back of the throat, e.g. /xʷ/ /ɣʷ/ /ʕʷ/. What is so unlikely about these sounds that you feel the need to postulate a different sound for *h3? You would need a good reason to disprove that *h3, if something like that can be reconstructed, would be one of the sounds I mention here which are commonly regarded as good candidates for what *h3 could represent. This is like asking 'hey what is 2+2? oh and mind you I do not want to hear the conventional answer '4' but I want to speculate myself without showing any argument why the conventional answer '4' might not be satisfactory'.

@Zaarin: You need the labial component so the first one is kinda out how else are you going to account for rounding. The second does not account for pharyngeal/lowering effects, which is why /ʕʷ/ is more likely. Other possibilities that I would speculate on would be /qʷʼ/ /xʷ/ or things like that, but I am pretty sure there are reasons why this is not so because I couldn't fathom why I would be the first one to think up of them.


But you know what I honestly believe is the case? We cannot know, because proto-languages are constructs to help us prove relationship between languages rather than actual languages which were ever actually extant single monolithic spoken languages at any point in history. So *h3 has certain features (suc as guttural, rounded, voiced, fricative/approximant) which help us establish a relationship between related languages. But it might not ever have been a sound, however it would be pronounced, that was ever once uttered in a single monolithic 'proto-indo-european' language. It might be that in the original Urheimat of the then small IE language family, you had a group of related languages which had this one phoneme (which we conveniently call *h3 here) which was pronounced differently in the various languages but might have shared some features so it makes sense for us to call it a single phoneme. For example, /xʷ/, /ʕʷ/, /qʷ/, /ɣʷ/ and so on, perhaps some existing as allophones of each other.

User avatar
2+3 clusivity
Avisaru
Avisaru
Posts: 454
Joined: Fri Mar 16, 2012 5:34 pm

Re: The Great Proto-Indo-European Thread

Post by 2+3 clusivity »

http://www.bbc.com/news/science-environment-31695214

Some BBC back-page fluff on migrations into Europe c. 8,000 - 4,500 years ago.
linguoboy wrote:So that's what it looks like when the master satirist is moistened by his own moutarde.

vokzhen
Avisaru
Avisaru
Posts: 352
Joined: Sat Aug 09, 2014 3:43 pm
Location: Iowa

Re: The Great Proto-Indo-European Thread

Post by vokzhen »

sirdanilot wrote:This is like asking 'hey what is 2+2? oh and mind you I do not want to hear the conventional answer '4' but I want to speculate myself without showing any argument why the conventional answer '4' might not be satisfactory'.
It's a lot more like asking what addition gives you 4 as a result, and just that the most straightforward answer is 2+2 or 3+1 doesn't discount 4+0 or -1+5 as possible answers. Given further qualifications for h3 may be equivalent of giving qualifications that require one of the numbers to be ≥4.

There's a pretty clear glottal-pharyngeal connection. You've got the Semetic languages shifting ejectives to pharyngeals, Circassian ejectives are pharyngealized and can drop the glottalization entirely, and Semitic and Northeast Caucasian often have a "voiced pharyngeal" that is a pharyngealized glottal or glottalized pharyngeal. Ejective fricatives in Tlingit have pharyngeal constriction, though less than in normal pharyngealization. A couple Mayan languages have epiglottals as final allophones of glottal stops, iirc. And I believe +RTR correlates with creakiness just as +ATR does with breathiness. Though probably not relevant to PIE, you've also got uvulars that relatively frequently change into both glottals and pharyngeals.

User avatar
jal
Sumerul
Sumerul
Posts: 2633
Joined: Tue Feb 06, 2007 12:03 am
Location: Netherlands
Contact:

Re: The Great Proto-Indo-European Thread

Post by jal »

sirdanilot wrote:But you know what I honestly believe is the case?
Yes, because you mentioned it before. And it is irrelevant. No-one is claiming what you are claiming is claimed. Since "strawman" seems to be your middle name, I'm not surprised.
It might be that in the original Urheimat of the then small IE language family, you had a group of related languages which had this one phoneme (which we conveniently call *h3 here) which was pronounced differently in the various languages but might have shared some features so it makes sense for us to call it a single phoneme. For example, /xʷ/, /ʕʷ/, /qʷ/, /ɣʷ/ and so on, perhaps some existing as allophones of each other.
This is as meaningless as saying "I don't believe there's a common ancestor to organism X, just a group of related species that all contributed to it's genome". Like species, languages have the equivalent of horizontal gene transfer, but that doesn't mean that related languages by definition don't have a shared common ancestor. Yes, the picture will be complex, but without any counter evidence, throwing your hands up and say "we can't know" is just weaseling out.


JAL

Post Reply