The Great Proto-Indo-European Thread

[Salmoneus]

I have seen the laryngeals represented as h with subscript e, a, o instead of 1, 2, 3. Quite neat, if you ask me.

It's probably as transparent as you're going to get in regards to their function.

But it does make it harder to posit more than three - what do you call h4, in that system?

[KathTheDragon]

Well, it's not a problem for me, since I don't think there were more than three.

[WeepingElf]

I have seen the laryngeals represented as h with subscript e, a, o instead of 1, 2, 3. Quite neat, if you ask me.

It's probably as transparent as you're going to get in regards to their function.

But it does make it harder to posit more than three - what do you call h4, in that system?

If there is one Indeed, Mallory & Adams, who distinguish between *h₂ and *h₄, write *h_a where they don't know which of their two a-colouring laryngeals to reconstruct.

[Neek]

I have seen the laryngeals represented as h with subscript e, a, o instead of 1, 2, 3. Quite neat, if you ask me.

It's probably as transparent as you're going to get in regards to their function.

But it does make it harder to posit more than three - what do you call h4, in that system?

If there is one Indeed, Mallory & Adams, who distinguish between *h₂ and *h₄, write *h_a where they don't know which of their two a-colouring laryngeals to reconstruct.

I think h_e, h<sub>a[/b], ho[/b] is perfectly fine with me. Perhaps h4 should be ha for Mallory & Adams, just cuz. It looks so succinct!

[quote="Sleinad Flar"]The "troll-worthy" PIE version already exist. At dnghu.org http://dnghu.org/Indo-European-Language-Europe/.[/quote]

That website is so awful I can't actually look at the content. Wow.</sub>

[jal]

I think h_e, h_a, h_o is perfectly fine with me. Perhaps h₄ should be h_a for Mallory & Adams, just cuz. It looks so succinct!

Fixed.


JAL

[KathTheDragon]

The "troll-worthy" PIE version already exist. At dnghu.org http://dnghu.org/Indo-European-Language-Europe/.

That website is so awful I can't actually look at the content. Wow.

They do seem to have missed the whole point of reconstructing laryngeals. What's more, they still claim that what they reconstruct is ancestral to Greek, which is clearly wrong.

[jal]

What's more, they still claim that what they reconstruct is ancestral to Greek, which is clearly wrong.

What, are they Greek nationalists or something?


JAL

[KathTheDragon]

Not a clue.

[Salmoneus]

The "troll-worthy" PIE version already exist. At dnghu.org http://dnghu.org/Indo-European-Language-Europe/.

That website is so awful I can't actually look at the content. Wow.

They do seem to have missed the whole point of reconstructing laryngeals. What's more, they still claim that what they reconstruct is ancestral to Greek, which is clearly wrong.

I've not looked at their theories in a long time. Why do you single out the claim to be ancestral to Greek? They're trying to reconstruct PIE, which is indeed ancestral to Greek, so surely the problem would just be that you disagree with their reconstruction, rather than anything with the ancestral-to-greek bit itself? Or am I missing something?

[Neek]

Flipping through their grammar, I don't get the jist that they're assuming Greek is the "least unadultered PIE dialect," you know, that sort of bullshit linguistic purity what-if question that gets thrown around (What if Latin didn't break off into various dialects? What would it look like today?-sort of idiocy). What I do get out of it is that at least one of the compilers was very much in love with Ancient Greek to the point that he idealized Late PIE as being merely pre-Proto-Greek.

But that's the least of my concern for the text. Despite what it says on the tin, it's a revival grammar. The title itself, A Grammar of Modern Indo-European should be indicative enough. They're not saying it in the text, but it's a reference grammar for a dead-language revival akin to Hebrew, just without an ethic body that's interested or a movement worth the intellectual effort. It's also not a grammar of PIE, but in their eyes, LIE (Late Indo-European, post-Hittite) so they can avoid the laryngeal debate and the grammatical controversies introduced by Hittite while still pander to the Nostraticism.

Oh, and their orthographical decisions are terrible and they should feel bad.

[KathTheDragon]

The "troll-worthy" PIE version already exist. At dnghu.org http://dnghu.org/Indo-European-Language-Europe/.

That website is so awful I can't actually look at the content. Wow.

They do seem to have missed the whole point of reconstructing laryngeals. What's more, they still claim that what they reconstruct is ancestral to Greek, which is clearly wrong.

I've not looked at their theories in a long time. Why do you single out the claim to be ancestral to Greek? They're trying to reconstruct PIE, which is indeed ancestral to Greek, so surely the problem would just be that you disagree with their reconstruction, rather than anything with the ancestral-to-greek bit itself? Or am I missing something?

Ok, I clearly wasn't very clear with what I meant. What I meant to say is that they're not reconstructing laryngeals, and claiming that this post-laryngeal PIE is still directly ancestral to Greek, despite the triple-reflex of laryngeals.

[Neek]

They also forgot the centum-satem distinction. In fact, it looks pretty sloppy all around. Gold star, guys!

[KathTheDragon]

Actually, they deliberately ignored the front-back velar distinction since they believe that the distinction is secondary.

[Neek]

I was trying to find where they even acknowledge it in the text and I couldn't find it. Either way, I think you can't reconstruct anything satem without that distinction. I wonder what use their grammar is without it.

[KathTheDragon]

In this text on PIE phonology, they first discuss the status of the 'palatovelars', before moving onto various synchronic sound changes. It's worth at least a read.

[jal]

They also forgot the centum-satem distinction. In fact, it looks pretty sloppy all around. Gold star, guys!

I have no knowledge about the finer details of this distinction, but in the document I just checked, which is I think the one were talking about (i.e. a-grammar-of-modern-indo-european-third-edition.pdf), there's plenty of discussion about the centum/satem split; Adobe Reader reports 52 instances of "satem" in that document (and 33 of centum). So whatever you think of their analysis, I think it's too bold a claim to state they "forget the centum-satem distinction".


JAL

[Pabappa]

In this text on PIE phonology, they first discuss the status of the 'palatovelars', before moving onto various synchronic sound changes. It's worth at least a read.

Much of that, though, is just copy pasted from Wikipedia with little editing. They even deleted the laryngeals in the "Hirt's Law" section, and didn't bother to edit the text that says "Hirt’s law, named after Hermann Hirt who postulated it originally in 1895, is a Balto-Slavic sound law which states in its modern form that the inherited Proto-Indo-European stress would retract to non-ablauting pretonic vowel or a syllabic sonorant if it was followed by a consonantal (non-syllabic) laryngeal that closed the preceding syllable." I had to go back to the original Wikipedia text to make sense of what they were saying. Unless they say somewhere else that they're using long vowels as a shortcut for vowel plus laryngeal that just looks like lazy editing to me. Even if they do they should have at least edited the Wikipedia text to say "long vowel" instead of "vowel plus laryngeal". Anyone who didnt realize they were taking their writing mostly from Wikipedia would probably be lost.

[Chagen]

Oh, and their orthographical decisions are terrible and they should feel bad.

Eh, Dnghu's not actually too bad except for using <c> for the voiced labiovelar, which is just WHAT.

If I had to romanize PIE I'd probably do it like this:

/m n/
<m n>

/p b bʰ t d dʰ ḱ ǵ ǵʰ k g gʰ kʷ gʷ gʰʷ/
<p b bh t d dh c j jh k g gh q gẉ ghẉ>

/s h₁ h₂ h₃ hₓ/
<s ḥ x ħ ḥ>

/r l j w/
<r l y w>

/e o i u a ṛ l̩ m̩ n̩/
<e o i u a ṛ ḷ ṃ ṇ>

I didn't use <h> for any of the laryngeals to distinguish things like /gh₃/ <gḥ> from /gʰ/ <gh>. Also things kinda broke down near the labiovelars, because this language apparently though a fucking voiced aspirated labiovelar was a good idea. Damn <ghẉ> is ugly, but I can't think of anything else. All laryngeals with unknown values will be assumed to be h₁. Thorn clusters are as <CCh>, e.g <djh> (as <dhjh> and <djh> don't seem to contrast in PIE except maybe in a tiny amount of words).

Anyway, Schleicher's Fable, in both the "original" (using the 2013 Byrd version; here I'm writing the effects of laryngeals on vowels like he is, but I normally wouldn't do that; I don't know if a <xa> for instance came from <xe> or <xo> so I'm playing it safe) and romanized forms:

More: show

h₂áu̯ei̯ h₁i̯osméi̯ h₂u̯l̥h₁náh₂ né h₁ést, só h₁éḱu̯oms derḱt. só gʷr̥hₓúm u̯óǵʰom u̯eǵʰed; só méǵh₂m̥ bʰórom; só dʰǵʰémonm̥ h₂ṓḱu bʰered. h₂óu̯is h₁ékʷoi̯bʰi̯os u̯eu̯ked: “dʰǵʰémonm̥ spéḱi̯oh₂ h₁éḱu̯oms-kʷe h₂áǵeti, ḱḗr moi̯ agʰnutor”. h₁éḱu̯ōs tu u̯eu̯kond: “ḱludʰí, h₂ou̯ei̯! tód spéḱi̯omes, n̥sméi̯ agʰnutór ḱḗr: dʰǵʰémō, pótis, sē h₂áu̯i̯es h₂u̯l̥h₁náh₂ gʷʰérmom u̯éstrom u̯ept, h₂áu̯ibʰi̯os tu h₂u̯l̥h₁náh₂ né h₁esti. tód ḱeḱluu̯ṓs h₂óu̯is h₂aǵróm bʰuged.

xáwey ḥyosméy xwḷḥnáx né ḥést, só ḥécwoms derct. só gẉṛḥúm wójhom wejhed; só méjhṃ bhórom; só djhémonṃ xṓcu bhered. xówis ḥéqoybhyos wewked: "djhímonṃ spécyox ḥecwoms-qe xájeti, cḗr moy aghnutor". ḥécwōs tu wewkond: "cludhí, xowey! tód spécyomes ṇsméy aghnutor cḗr: djhémō, pótis, sē xawyes xwḷḥnáx ghẉérmom wéstrom wept, xáwibhyos tu xwḷḥnáx né ḥesti. tód cecluwṓs xówis xajróm bhuged.

Seeing the various "translations" of Schleicher's Fable (was that fable invented wholesale by Schleicher himself or did he base it on something?) over time, it's rather humorously adorable how much the first few scholars of PIE fetishized Sanskrit. The original made by Schleicher himself looks so much like Sanskrit it's ridiculous; imagine his reaction to what we NOW think PIE was like.

One thing I find odd about PIE is the odd inconsistency between the palatovelars and labiovelars next to their respective glides. Labiovelars became plain next to /w/, but palatovelars remain the same next to /j/, as the above fable shows with words like <spécyomes>. Sanskrit preserves this somewhat; there are worlds like yujyate "it is joined".

[KathTheDragon]

One thing I find odd about PIE is the odd inconsistency between the palatovelars and labiovelars next to their respective glides. Labiovelars became plain next to /w/, but palatovelars remain the same next to /j/, as the above fable shows with words like <spécyomes>. Sanskrit preserves this somewhat; there are worlds like yujyate "it is joined".

Quite simply because the 'palatoverlars' weren't palatalised, just that the 'plain velars' were more backed, ie. /k q/ for <ḱ k>.

[WeepingElf]

Quite simply because the 'palatoverlars' weren't palatalised, just that the 'plain velars' were more backed, ie. /k q/ for <ḱ k>.

We don't really know; but the 'palatovelars' are much more frequent than the 'plain velars', so they probably had a less marked place of articulation, and /k q/ is more likely than /c k/ here. I prefer the agnostic terms 'front velars' and 'back velars' here.

[Chagen]

One thing I find odd about PIE is the odd inconsistency between the palatovelars and labiovelars next to their respective glides. Labiovelars became plain next to /w/, but palatovelars remain the same next to /j/, as the above fable shows with words like <spécyomes>. Sanskrit preserves this somewhat; there are worlds like yujyate "it is joined".

Quite simply because the 'palatoverlars' weren't palatalised, just that the 'plain velars' were more backed, ie. /k q/ for <ḱ k>.

http://rationalwiki.org/wiki/Extraordin ... y_evidence

It seems like a good idea, but I can't trust it; for instance, why would Sanskrit unfailingly change normal velars to palatal fricatives (one later becoming a palatal affricate/stop)? Velars (and voiceless stops in general) are pretty damn stable throughout the world, but palatalized velars becoming palatal fricatives/affricates we have quite a bit of attestation for around the world.

If the palatovelars were actually normal plain velars, I would expect to find them reflexed as plain velars in nearly all the branches; half the family keeping them plain but the half randomly turning them into palatalized fricatives is just bizarre. One could say "they palatalized before front vowels and analogy fixed any damage/irregular paradigms", but in Sanskrit, palatovelars do not become <c j> before front vowels like plain velars did; it has *ḱe > śa but *ke > ca; *ḱeyey > Skrt. śaye "he/she/it lies" but *kʷe > PII. ke > Skrt. -ca "and".

None of the other plain voiceless stops underwent anything similar*, as well (though you could explain that away with "the velars were in a much weirder situation", I guess).

/////////

WeepingElf: Then we run into the problem that having velars that are front enough to have some kind of palatal affect as the least marked velars is incredibly rare in languages.

Because of this, I feel that the palatovelars arrived through some sound change that occured soon before the dialects split. Perhaps originally there were only plain velars and labiovelars (a well-attested state of affairs cross-lingusitically), and then the plain velars palatalized before front vowels, with analogy fixing any problems (similar to what happened in Sanskrit), then for whatever reason this became phonemic, right before the dialects split, with the dialects changing this rather unstable state of affairs.

You would even take this further and say that there were only plain velars, which labialized before back vowels and palatalized before front vowels with analogy fixing things; but given the extensive ablaut seen in the language, this would result in some problems; if *ken- and *kon- become *ḱen- and *kʷon, which variant should win out for all forms of the root (since we see no evidence for consonant mutation in PIE)?

[WeepingElf]

Fair. If the three velar series preserve, as I fancy, early pre-PIE (perhaps Proto-Indo-Uralic?) vowel features (before what I call the "Great Vowel Collapse", the merger of all non-high vowels into */a/, which would then split again in the course of the development of ablaut), one would indeed expect the front velars having been palatalized. Apparently, both front vowels and rounded vowels were more frequent than back unrounded vowels (and front rounded vowels did not occur, otherwise we'd see a fourrth - labio-palatovelar - series in PIE). That no attested IE language kept the system intact probably shows that it was typologically unstable.

[Herr Dunkel]

Satemisation of /*q *k *ɕ/ or something isn't really that hard to believe. It's not an extraordinary claim; changes ten times over weirder have happend many times over.

[KathTheDragon]

My assumed course of events is the following: we start with /k q/, and /q/ fronts to /k/. In some dialects (the satem daughters), the original /k/ fronts in response to /c/, and further dialect-specific changes ensue. In the other dialects (centum), the two dorsals merge.

[CatDoom]

Satemisation of /*q *k *ɕ/ or something isn't really that hard to believe. It's not an extraordinary claim; changes ten times over weirder have happend many times over.

Further, it's entirely reasonable to believe that a change like this happened once, and propagated throughout an areal group of PIE dialects. The alternative is that /c/ or /kʲ/ backed to /k/, an (almost?) unattested sound change that would have to have occurred independently in multiple dialect clusters, including some that were likely quite distant, geographically. Positing a phoneme as bizarre as /ɢʱ/ is enough to give anyone pause, but it still reqires fewer miracles than the alternative. It may provide some indirect evidence for the glottalic theory, though obviously nothing sufficient to really justify it; a "glottalic" reconstruction of pre-PIE is an attractive idea, however.

In any event, it seems to be pretty uncontroversial in the current literature that there was nothing palatal about the frontmost velar series.