The Great Proto-Indo-European Thread

Discussion of natural languages, or language in general.
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 take things Kloekhorst says with a pinch of salt and a very critical eye, since he's said several things which I think are utter bullshit.

User avatar
Tropylium
Avisaru
Avisaru
Posts: 512
Joined: Sun Aug 07, 2005 1:13 pm
Location: Halfway to Hyperborea

Re: The Great Proto-Indo-European Thread

Post by Tropylium »

WeepingElf wrote:Alwin Kloekhorst seems to think that *h2 and *h3 were uvular stops in PIE; his web site teases us of a manuscript in which he says that there is evidence of this in Anatolian;
If I had to guess, probably the Lycian reflexes as k q g.
[ˌʔaɪsəˈpʰɻ̊ʷoʊpɪɫ ˈʔæɫkəɦɔɫ]

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 »

KathTheDragon wrote:I take things Kloekhorst says with a pinch of salt and a very critical eye, since he's said several things which I think are utter bullshit.
Yes, many things he has written are ones I do not agree with at all, such as his reconstruction of the Proto-Anatolian (and Early PIE) stop system, and his recent defence of the perfect theory of the hi-conjugation. Some of his ideas are interesting nevertheless. I don't know what I should think of this one; the laryngeals may have been uvular stops once, but probably in PIE1 at its latest; I am of the opinion that in PIE2, they already were fricatives. The Lycian stop reflexes appear to be a late innovation, as the cuneiform spellings in Luwian suggest fricatives as in Hittite.
...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
Soap
Smeric
Smeric
Posts: 1228
Joined: Sun Feb 16, 2003 2:57 pm
Location: Scattered disc
Contact:

Re: The Great Proto-Indo-European Thread

Post by Soap »

Im not aware of ther being reflexes of h2 & h3 in Hittite but Id like to see how the author reconstructs words with laryngeals in between consonants. e.g. if the word for father is /pqter/ or something like /pəqter/ instead. Im not sure how we know what the values of the letters were in lycian anyway ... couldnt they have been like the Romans etc who used letters for more than one sound?
Sunàqʷa the Sea Lamprey says:
Image

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 »

Soap wrote:Im not aware of ther being reflexes of h2 & h3 in Hittite
You must've been living under a rock, then, because Hittite is quite famous for being the first discovered language to have consonantal reflexes of laryngeals in . Examples are ḫanza, ḫant- "forehead" < *h₂ent-, ḫāraš, ḫaran- "eagle" < *h₃eron-.
Im not sure how we know what the values of the letters were in lycian anyway ... couldnt they have been like the Romans etc who used letters for more than one sound?
Kloekhorst has actually done a study on this.

User avatar
Howl
Sanci
Sanci
Posts: 69
Joined: Sat Oct 07, 2017 7:43 pm
Location: Campinia

Re: The Great Proto-Indo-European Thread

Post by Howl »

WeepingElf wrote:
KathTheDragon wrote:I take things Kloekhorst says with a pinch of salt and a very critical eye, since he's said several things which I think are utter bullshit.
Yes, many things he has written are ones I do not agree with at all, such as his reconstruction of the Proto-Anatolian (and Early PIE) stop system, and his recent defence of the perfect theory of the hi-conjugation. Some of his ideas are interesting nevertheless.
For me, it's his attack on Narten and E/O acrostatic ablaut patterns that makes me cringe. I think the Leiden model is weak. Somehow, with the magic wand of accent shifts, everything is derived from some clean paradigm. And anything that points in a different direction is reasoned away in a lawyer like way.

But I am curious how these laryngeals could have been uvular stops when (1) they behave like fricatives in the rules that determine the structure of a PIE syllable and (2) they reduce to zero or *h, but never *k, in all PIE languages.

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 »

Well, some of the pre-stages the Leiden model assumes are plausible, such as an original R(Ø)-S(é)- in the amphikinetic accusative, and it's even possible that the nominative there had S(Ø) as well, but that's strictly speaking beyond the comparative method.

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 »

I shall address three matters in this post.

1. Alwin Kloekhorst has an idea about how the Lydian dative singular ending emerged. As you may know, Lydian, one of the younger Anatolian languages, has a strange-looking dative singular ending, (this is the conventional transcription of a Lydian letter which probably represented a palatal lateral). Various explanations have been suggested, such as the Proto-Anatolian adverbial suffix *-li (as in Hittite language designations), or a connection to the Hittite pronominal genitive -êl. Now Kloekhorst suggests that PIE *y has simply become λ in word-final position in Lydian, so the Lydian ending is just the regular development of the PIE dative singular *-ey. Sounds nice, but I don't know how good this really is.

2. I have recently looked around at this site. Looks interesting, but doubtful. This is Carlos Quines, the Spanish guy who brought us "Modern Indo-European" (regularized PIE proposed as a European auxlang), who examines the origin of PIE in context of genetics and archaeology. Some of his ideas are definitely non-canonical. For instance, he doubts that the Corded Ware culture, conventionally associated with Northwest IE, was IE at all; rather, he assumes a Uralic language here! This is mainly based on genetics: he claims that R1b was the dominant Y-DNA haplogroup of the PIE speakers, while Corded Ware was mostly R1a. R1b would have been brought to western Europe by the Bell Beaker people, whom he has as descendants of Yamnaya bypassing Corded Ware. All this is hard to stomach, and I am very doubtful. I have grown weary of language-genetics connections of this sort - language shifts (i.e., populations switching to another, more prestigious language) happen all too often (I have first-hand experience of this, being grown up in a Northern German village that was shifting from Low to High German). Also, AFAIK, Bell Beaker originated in the Iberian Peninsula, and probably has nothing to do with IE (other than seeding NW IE languages with loanwords from Vasconic or whatever).

3. I am trying to figure out who spoke PIE0, PIE1, PIE2 and PIE3, respectively, and when. I think that PIE0 (the "pre-Caucasianization" - see for instance here for the concept of a Caucasian substratum in PIE) stage close to Proto-Indo-Uralic - may have been the language of the Khvalynsk culture north of the Caspian Sea ca. 5000 BC. These were either pastoralists or still hunter-gatherers. The Khvalynsk culture later merged with - or rather took over - their western neighbours, the Dniepr-Donets culture north of the Black Sea, who were argriculturalists, resulting in what one could call "Pre-Yamnaya". (Is this Sredny Stog, or is Sredny Stog an offshoot of Dniepr-Donets that had not taken over by Khvalynsk? Help would be appreciated!) The language of Dniepr-Donets is unknown, but it may have been related to the Abkhaz-Adyghean languages of the Northwestern Caucasus, and brought such things as ejectives, the three-way split of the velars, and a reduced vowel system (the Great Vowel Collapse) into PIE0 which thus changed into PIE1, spoken north of the Black and Caspian Seas around 4500 BC. This would change into PIE2, the common ancestor of PIE3 and Anatolian, spoken by the Early Yamnaya around 3500 BC, with the rise of ablaut and a shift from agglutinating to fusional morphology (and the ejectives shifting to implosives). An offshoot of PIE1 moving west could have been Aquan, the language of the Old European Hydronymy (which I explore in my Hesperic conlangs). PIE3 would be later Yamnaya around 3000 BC; this is the ancestor of the non-Anatolian IE languages and the PIE found in the mainstream handbooks.
...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
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 »

WeepingElf wrote:
2. I have recently looked around at this site. Looks interesting, but doubtful. This is Carlos Quines, the Spanish guy who brought us "Modern Indo-European" (regularized PIE proposed as a European auxlang), who examines the origin of PIE in context of genetics and archaeology. Some of his ideas are definitely non-canonical. For instance, he doubts that the Corded Ware culture, conventionally associated with Northwest IE, was IE at all; rather, he assumes a Uralic language here!
That's nonsense.
This is mainly based on genetics: he claims that R1b was the dominant Y-DNA haplogroup of the PIE speakers, while Corded Ware was mostly R1a.
Great. You know who else was mostly R1a? The Indo-Iranians. And the Tocharians. Also the Slavs, Balts, Germans, Greeks, Armenians, and iirc Albanians.
You know who WASN'T R1a? Uralic speakers, except where, in Europe, they've intermarried with local Slavic and Baltic populations.
R1b would have been brought to western Europe by the Bell Beaker people, whom he has as descendants of Yamnaya bypassing Corded Ware. All this is hard to stomach, and I am very doubtful.
Well, that bit's accurate. R1b was introduced to western Europe by the Bell Beaker people, who were descendants of Yamnaya, and who probably did bypass Corded Ware somehow. [my theory is that they represent a more southerly, Danubian route, while CWC represents the northern plains route. But it's possible that the initial migration was more mixed and the separation happened later]. But the CWC=non-PIE stuff is indeed nonsense.
I have grown weary of language-genetics connections of this sort - language shifts (i.e., populations switching to another, more prestigious language) happen all too often (I have first-hand experience of this, being grown up in a Northern German village that was shifting from Low to High German).
I wouldn't go that far. Genuine language shifts from one language family to another (rather than one variety to a related dialect) occur very rarely without at least some migration, and the more we look the more we find that language (and other culture) shifts are more associated with migration than we thought. Language shifts undoubtedly do happen, but it's parsimonious not to use them as an excuse at every opportunity.

Also, AFAIK, Bell Beaker originated in the Iberian Peninsula, and probably has nothing to do with IE (other than seeding NW IE languages with loanwords from Vasconic or whatever).
There's an ambiguity here in the name 'Bell Beaker'. Bell beakers as a pottery type did originate in Iberia, along with apparently some other cultural traits. From there they spread across western Europe. The bell beaker PEOPLE, however, as found across western and central europe and the british isles, NOT including Iberia (at least at first), were a steppe tribe extremely closely related to the CWC people, but tending to have R1b instead of R1a. How exactly this tribe from the east picked up a cultural trait from the west is not clear.
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
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 »

WeepingElf wrote:
2. I have recently looked around at this site. Looks interesting, but doubtful. This is Carlos Quines, the Spanish guy who brought us "Modern Indo-European" (regularized PIE proposed as a European auxlang), who examines the origin of PIE in context of genetics and archaeology. Some of his ideas are definitely non-canonical. For instance, he doubts that the Corded Ware culture, conventionally associated with Northwest IE, was IE at all; rather, he assumes a Uralic language here!
That's nonsense.
This is mainly based on genetics: he claims that R1b was the dominant Y-DNA haplogroup of the PIE speakers, while Corded Ware was mostly R1a.
Great. You know who else was mostly R1a? The Indo-Iranians. And the Tocharians. Also the Slavs, Balts, Germans, Greeks, Armenians, and iirc Albanians.
You know who WASN'T R1a? Uralic speakers, except where, in Europe, they've intermarried with local Slavic and Baltic populations.
R1b would have been brought to western Europe by the Bell Beaker people, whom he has as descendants of Yamnaya bypassing Corded Ware. All this is hard to stomach, and I am very doubtful.
Well, that bit's accurate. R1b was introduced to western Europe by the Bell Beaker people, who were descendants of Yamnaya, and who probably did bypass Corded Ware somehow. [my theory is that they represent a more southerly, Danubian route, while CWC represents the northern plains route. But it's possible that the initial migration was more mixed and the separation happened later]. But the CWC=non-PIE stuff is indeed nonsense.
I have grown weary of language-genetics connections of this sort - language shifts (i.e., populations switching to another, more prestigious language) happen all too often (I have first-hand experience of this, being grown up in a Northern German village that was shifting from Low to High German).
I wouldn't go that far. Genuine language shifts from one language family to another (rather than one variety to a related dialect) occur very rarely without at least some migration, and the more we look the more we find that language (and other culture) shifts are more associated with migration than we thought. Language shifts undoubtedly do happen, but it's parsimonious not to use them as an excuse at every opportunity.

Also, AFAIK, Bell Beaker originated in the Iberian Peninsula, and probably has nothing to do with IE (other than seeding NW IE languages with loanwords from Vasconic or whatever).
There's an ambiguity here in the name 'Bell Beaker'. Bell beakers as a pottery type did originate in Iberia, along with apparently some other cultural traits. From there they spread across western Europe. The bell beaker PEOPLE, however, as found across western and central europe and the british isles, NOT including Iberia (at least at first), were a steppe tribe extremely closely related to the CWC people, but tending to have R1b instead of R1a. How exactly this tribe from the east picked up a cultural trait from the west is not clear.
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:
WeepingElf wrote:
2. I have recently looked around at this site. Looks interesting, but doubtful. This is Carlos Quines, the Spanish guy who brought us "Modern Indo-European" (regularized PIE proposed as a European auxlang), who examines the origin of PIE in context of genetics and archaeology. Some of his ideas are definitely non-canonical. For instance, he doubts that the Corded Ware culture, conventionally associated with Northwest IE, was IE at all; rather, he assumes a Uralic language here!
That's nonsense.
I concur with you, even if I wouldn't put it as bluntly. We don't know for sure which language the Corded Ware people spoke, but Northwest IE is a more plausible idea than Uralic. Sure, I have seen much more idiotic claims, but I think this one is already beyond the pale.

It also tells a lot that this is from a guy who seriously peddles a regularized PIE as a pan-European auxlang, which is kinky enough; and while there are people who have kinky ideas in one field and good ones in another, people who are kinky everywhere are much more common.
This is mainly based on genetics: he claims that R1b was the dominant Y-DNA haplogroup of the PIE speakers, while Corded Ware was mostly R1a.
Great. You know who else was mostly R1a? The Indo-Iranians. And the Tocharians. Also the Slavs, Balts, Germans, Greeks, Armenians, and iirc Albanians.
You know who WASN'T R1a? Uralic speakers, except where, in Europe, they've intermarried with local Slavic and Baltic populations.
I know; and this means that Quines writes is, as you put it, nonsense - or at least wild speculation not concerned with facts.
R1b would have been brought to western Europe by the Bell Beaker people, whom he has as descendants of Yamnaya bypassing Corded Ware. All this is hard to stomach, and I am very doubtful.
Well, that bit's accurate. R1b was introduced to western Europe by the Bell Beaker people, who were descendants of Yamnaya, and who probably did bypass Corded Ware somehow. [my theory is that they represent a more southerly, Danubian route, while CWC represents the northern plains route. But it's possible that the initial migration was more mixed and the separation happened later]. But the CWC=non-PIE stuff is indeed nonsense.
It is. All sensible interpretation of the evidence suggest that the most likely language of the CWC is the northwestern dialect of Late PIE, i.e. the common ancestor of Italic, Celtic, Germanic and Balto-Slavic.

But I am not sure about R1b being introduced by the Bell Beaker people, I have read completely different stories. Alas, all this genetics stuff is both hard to make sense of if you are not a geneticist, and IMHO only loosely connected to language families as language shifts aren't uncommon and mismatches are evident (how about Basques being overwhelingly R1b, for instance?).
I have grown weary of language-genetics connections of this sort - language shifts (i.e., populations switching to another, more prestigious language) happen all too often (I have first-hand experience of this, being grown up in a Northern German village that was shifting from Low to High German).
I wouldn't go that far. Genuine language shifts from one language family to another (rather than one variety to a related dialect) occur very rarely without at least some migration, and the more we look the more we find that language (and other culture) shifts are more associated with migration than we thought. Language shifts undoubtedly do happen, but it's parsimonious not to use them as an excuse at every opportunity.
Fair. Most children pick up the language of their germane parents, and language shifts usually occur
where a language is more prestigious than another, which probably was hardly ever the case in the Paleolithic.

Also, AFAIK, Bell Beaker originated in the Iberian Peninsula, and probably has nothing to do with IE (other than seeding NW IE languages with loanwords from Vasconic or whatever).
There's an ambiguity here in the name 'Bell Beaker'. Bell beakers as a pottery type did originate in Iberia, along with apparently some other cultural traits. From there they spread across western Europe. The bell beaker PEOPLE, however, as found across western and central europe and the british isles, NOT including Iberia (at least at first), were a steppe tribe extremely closely related to the CWC people, but tending to have R1b instead of R1a. How exactly this tribe from the east picked up a cultural trait from the west is not clear.
This is also pretty much my opinion on the matter. The Bell Beaker culture has a similar (not identical) distribution as the Old European Hydronymy, which I suspect being a residue of a PIE1-descended language family I call "Aquan". IMHO, not all people in that area were "Bell Beaker people", though; I suspect the Bell Beaker culture to be the remains of a diaspora of some sort, perhaps a merchant caste (strontium isotope analyses on some Bell Beaker skeletons, such as the Amesbury Archer, revealed that many of them had travelled far); these people would not have ventured far beyond where the language and customs of the local people were familiar, hence the close correlation between BB and OEH.

The mixture of eastern genes and language (not sure about the genes, though!) and western cultural traits does seem somewhat odd, but why shouldn't an outlier develop a cultural influence that radiates back into older turf? Consider Western, and more specifically, North American, culture. The people spread across the continent from east to west, but yet the most culturally influential region is now - California (Hollywood cinema and all that), which was a late addition. Perhaps the Iberian Peninsula was the California of the Aquan world, so to speak.
...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
Howl
Sanci
Sanci
Posts: 69
Joined: Sat Oct 07, 2017 7:43 pm
Location: Campinia

Re: The Great Proto-Indo-European Thread

Post by Howl »

I am very skeptical about theories in which the Caucasian languages influenced PIE. These theories always serve to give PIE more of a profile like the Caucasian languages of today. But we don't have a clue what those languages were like more than 5000 years ago. And I don't believe in linguistic refrigerators that keep the phonetic structure of a language the same over millennia.

And who were those mythical Caucasian females that the people on the Pontic-Caspian steppe supposedly mated with? Everyone in the Caucasus mountains had a genetic component, called ANF (Anatolian Neolithic Farmer) that was not present in the people on the steppe.

Also, if IE came from the steppe and language change must always be accompanied by genes, then how did Hittite get to Anatolia? There are no steppe genes in the Hittite samples.

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 »

nota bene: there are no Hittite samples. There are a small number of samples from the Hittite polity (some of which iirc have been argued to have steppe influence), but none of them are from clearly ethnically Hittite, high-caste burials; they may therefore be Hattians, who formed the majority of the ancestry of most people in the "Hittite" empire.
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
Howl
Sanci
Sanci
Posts: 69
Joined: Sat Oct 07, 2017 7:43 pm
Location: Campinia

Re: The Great Proto-Indo-European Thread

Post by Howl »

Salmoneus wrote:nota bene: there are no Hittite samples. There are a small number of samples from the Hittite polity (some of which iirc have been argued to have steppe influence), but none of them are from clearly ethnically Hittite, high-caste burials; they may therefore be Hattians, who formed the majority of the ancestry of most people in the "Hittite" empire.
That is one possible explanation. But it does have its weaknesses. How much evidence is there for a caste system within the Hittite empire? And why didn't those aristocratic genes find their way into the lower classes? Also, if the gene samples do not confirm to your theory, can you just say that they did not sample the right population? With this line of reasoning, even an India PIE homeland becomes impossible to falsify, since "they haven't sampled the true Aryans, yet'".

User avatar
Frislander
Avisaru
Avisaru
Posts: 836
Joined: Mon Feb 29, 2016 6:34 am
Location: The North

Re: The Great Proto-Indo-European Thread

Post by Frislander »

Howl wrote:I am very skeptical about theories in which the Caucasian languages influenced PIE. These theories always serve to give PIE more of a profile like the Caucasian languages of today. But we don't have a clue what those languages were like more than 5000 years ago. And I don't believe in linguistic refrigerators that keep the phonetic structure of a language the same over millennia.
Well we can make a pretty good guess as to how those languages were a few thousand years ago given that Nakh-Daghestanian must clearly be of a quite large time depth and the languages have all stayed fairly similar phonology-wise, so their proto-language must have been fairly similar too, and I see no reason for that not to be the case for the other to families as well (similarly the fact that all the families of the Pacific-Northwest must bee reconstructed with quite similar inventories attests to the long time depth of the contact situation and the relative stability of their phonologies over time). And while a language family can differentiate into distinct sub-branches with radically different profiles we also know of many cases where the daughter languages have maintained fairly similar typological profiles over thousands of years; just look at haw similar the Na-Dené or Mayan languages are in their phonology and grammar despite clearly being several thousands of years old. I therefore don't find it at all unreasonable that the Caucasus region should have had a similar typological profile thousands of years ago to what it has today.
https://frislander.tumblr.com/

First known on here as Karero

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 »

While I agree with the principle that there's no reason to assume sudden change in the Caucasus without evidence, I think your argument there's a little weak. After all, Mayan and Na-Dene aren't the only families to show considerable typological uniformity after thousands of years of divergence. Another classic example is Romance, where the languages have maintained a great deal of -in some cases quite detailed - similarity despite 1500 years of divergence.

The problem is, Romance DIDN'T maintain that typological similarity. It innovated it, across the whole of the continent at once. No sane reconstruction of Proto-Romance would look anything like attested (even vernacular) Latin!


But of course, there's no specific reason to think that that happened in the Caucasus. The typological divergence between Caucasian families discourages the idea of a recent sprachbund, while Caucasoid features in PIE encourage the idea of at least some areal similarities, if not genetic connection, in the distant past.



Howl: in the specific case of the Hittites, yes, we know that there were ethnic distinctions in the polity. Most of the core area ruled by the Hittites had, until a short time before, been ruled by the Hattians - hence the name of the Hittites, which comes from foreign names for their country as "land of the Hattians" - who were a non-IE group. The Hattians were conquered by the Hittites. We know there was preservation of the Hattian religion and (at least in liturgical contexts) language well into the Hittite era. So it's not at all surprising that if you randomly pick five (iirc) people from the Hittite era who weren't from the ruling class, you wouldn't get any obvious Hittite genes.

If the Hittite ruling class don't look IE, that's more surprising (as long as they aren't very late, since the Hittite ruling class was iirc in turn Hurrianised later on). What would be more surprising still would be early Hittite samples from Nesh, or samples from other Anatolian groups where there isn't such a clear multiethnicity.


The potentially disruptive news, btw, is that some people believe that there are seemingly Anatolian names found in Eblaite records of the inhabitants of "Armi", an unknown, presumably highland Anatolian, polity. This would be really important, because those names would be found in Anatolia at a time contemporary to Yamnaya culture north of the steppe, and apparently not looking like a distinct recent migrant class. Given the lack of any obvious migration from the steppe before then, and the lack of anything obviously migrationy about the alleged Anatolians in Armi, that would not only prove a much earlier separation between PIE and PA (effectively we'd be firmly in Indo-Hittite territory, conceptually), but would also make the prehistory of the language really puzzling. Given the very low levels of steppe genes in anatolia, and the high level of non-steppe genes on the steppe, it would suggest that Indo-Hittite developed south of the caucasus and then migrated north, either via the caucasian piedmont*, or more radically through europe (radical because this could even put LBK back on the table as Indo-Hittite**).

*culturally this has always been suspected - there seems, for instance, to be a development from mesopotamian ziggurats to armenian mound tombs to steppe kurgans, as well as some transmission of metallurgical culture. But there's no obvious genetic pathway for this (it turns out Maykop, the obvious link, wasn't that genetically related).

**the radical theory there would be that mesolithic europe was indeed Indo-Hittite, that Anatolian is a back-migration from Europe, and that an influx of farmer women brough the language to the steppe. However, since the cultural and genetic ties to europe are even weaker than those to the caucasus, this seems less intuitively likely.
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
Howl
Sanci
Sanci
Posts: 69
Joined: Sat Oct 07, 2017 7:43 pm
Location: Campinia

Re: The Great Proto-Indo-European Thread

Post by Howl »

I don't believe that language always has to follow the genes. For example, there are numerous cases in history where an invading people eventually ended up speaking the language of the native people. And mechanisms like elite dominance and the need for a lingua franca make it possible that a population takes up a new language without a massive transfer of genes. Of course the speakers of the new language still have to be present in the population, and they will leave a genetic signal. But that signal can be more subtle. Now, this gives more leeway for competing models for the spread of languages. But that is just the consequence of acknowledging that we can't know more about the past than what we really do know.

And if I ignore genetics and potential links to other language families for a moment, I would say that a Balkan origin of PIE is the best model. The biggest diversity of PIE dialects (Celtic, Italic, Illyrian, Dacian, Thracian, Phrygian, Greek, Balto-Slavic) can be found in and around the Balkan, while the steppe just has Indo-Iranian. Also, the Balkan has the neolithic technologies and population density to support the major linguistic expansion that PIE went through.

But the farmers on the Balkan came from Anatolia. And a southern origin of PIE would make it very hard to explain the obvious commonalities with Uralic and the Altaic language families. So for me, PIE must have been an EHG language. And then the best candidate for a PIE origin would be the EHG communities just east of the Balkan. We know that these communities entered Balkan from the late 4th millennium BCE onwards. And this ties in neatly with the 'through the Balkan' models of Anatolian.

Vijay
Smeric
Smeric
Posts: 2244
Joined: Sat Feb 06, 2016 3:25 pm
Location: Austin, TX, USA

Re: The Great Proto-Indo-European Thread

Post by Vijay »

Howl wrote:I don't believe that language always has to follow the genes.
If it did always have to follow the genes, my native language certainly wouldn't be English of all things.

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 »

Howl wrote:I don't believe that language always has to follow the genes. For example, there are numerous cases in history where an invading people eventually ended up speaking the language of the native people. And mechanisms like elite dominance and the need for a lingua franca make it possible that a population takes up a new language without a massive transfer of genes. Of course the speakers of the new language still have to be present in the population, and they will leave a genetic signal. But that signal can be more subtle. Now, this gives more leeway for competing models for the spread of languages. But that is just the consequence of acknowledging that we can't know more about the past than what we really do know.
Fine. Genes and languages do not always go together - they often do, but they don't always do. The same is true with language and material culture (consider the many nations speaking whatever languages but all using coke bottles, to take a modern example), so such notions as "Corded Ware spoke NW IE" must be taken with a grain of salt and can be questioned. Languages spread faster than genes; material culture spreads faster than either genes or languages. Yet, when genes and material culture go together, chance is that language also goes with them - simply because it is people moving around, carrying with them their whole cultural heritage (e.g. Anglo-Americans are both genetically and culturally very close to the English - and speak the same language). But even then, caution is in order.
And if I ignore genetics and potential links to other language families for a moment, I would say that a Balkan origin of PIE is the best model. The biggest diversity of PIE dialects (Celtic, Italic, Illyrian, Dacian, Thracian, Phrygian, Greek, Balto-Slavic) can be found in and around the Balkan, while the steppe just has Indo-Iranian. Also, the Balkan has the neolithic technologies and population density to support the major linguistic expansion that PIE went through.
The Balkan homeland is IMHO the least ridiculous among the alternative PIE homeland proposals I have heard of. But the Balkan Peninsula is a region which has received waves upon waves of immigration, and its linguistic diversity is often overrated. The ancient Balkan looks like a mess of independent IE branches, sure, but this is because we don't know those fragmentarily attested languages well enough to classify them. It may well turn out that Illyrian, Thracian, Dacian, Macedonian and what else those are named are no farther removed from each other than, say, the Italic languages, and that Greek may just be a member of this branch, too (and Albanian simply the last modern residue of this besides Greek). That would mean that the ancient Balkan Peninsula was not more diverse than ancient Italy.

Also, not all the languages you mentioned went through the Balkan Peninsula at all. Celtic, Italic and Balto-Slavic, according to mainstream theory, did not - they are, together with Germanic, considered members of the Northwestern dialect group which is held to have evolved from the PIE dialect spoken by the Corded Ware people. South Slavic and Romanian are of course late arrivals on the Balkan Peninsula, the other NW IE languages have nothing to do with it at all.

What regards the steppe, it is a typical "spread zone"; for as long as we can tell, language after language spread out there. In historical times, we have the Huns, the Avars, the Magyars, the Mongols and finally the Turks, and the Iranian peoples who dominated it in Hellenistic and Roman times were like that, too. In the Middle Ages, the "Great Tartary" as it was known then was often called the vagina gentium for that reason. This is easy to explain: on the steppe, the grass is always greener in the west. The farther west you go, the more temperate the climate becomes - and the richer the pastures. Hence, people after people moved west on the steppe, and often beyond it.

Most mainstream IEists conjecture that PIE was spoken by the Yamnaya culture in what is now Ukraine and southern Russia, and the reasons for that are IMHO good. The culture reconstructed on the ground of what can be reconstructed for the PIE language matches what we know of Yamnaya quite well, and there is archaeological and genetic evidence of westward movements from there.
But the farmers on the Balkan came from Anatolia. And a southern origin of PIE would make it very hard to explain the obvious commonalities with Uralic and the Altaic language families. So for me, PIE must have been an EHG language. And then the best candidate for a PIE origin would be the EHG communities just east of the Balkan. We know that these communities entered Balkan from the late 4th millennium BCE onwards. And this ties in neatly with the 'through the Balkan' models of Anatolian.
The early Neolithic farmers can't have brought PIE with them because they did not know wheeled vehicles and some other things which PIE has well-reconstructed words for (and hunter-gatherers are a fortiori out for the same reasons). The Yamnaya did know these things, and could have brought PIE to the Balkan Peninsula and the rest of Europe, and probably did. (The infamous Mongol-like "Kurgan nomads" are a strawman, though. The Yamnaya were not much like Mongols; in the western part of their homeland at least, they were sedentary farmers, even if herding held more prestige than growing crops.)
...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
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 »

A correction: the only people we can be pretty sure derive from the Corded-Ware Culture are the Indo-Iranians. Indo-Iranians seem culturally and genetically derived from the Sintashta of the Caspian steppe - the people who developed fire cults and chariots. But Sintashta weren't in-situ descendents of Yamna. Instead, the Sintashta seem culturally AND genetically (and so probably linguistically) derived from CWC! Specifically, Sintashta (like modern indo-Iranians) show greatly elevated levels of western european farmer and hunter-gatherer genes; they clearly replaced the older, 'pure' steppe culture in the area. (they also show cultural traits from the west, like a sudden interest in metalwork).


So, what happened seems to be:
- very early (early Yamnaya or pre-Yamnaya) steppe people invaded northern europe. They absorbed some local genes and technologies, creating CWC. Other steppe people remained on the steppe.
- centuries later, CWC people migrated back onto the steppe, overrunning the 'native' steppe folk. The most famous of these were the Sintashta, but other steppe groups (like the Srubna to their west) seem genetically very similar, so Sintashta seems just to be a particularly succesful part of a general migration.
- the Sintashta - now armed with chariots and metal weapons, and seemingly more mobile than their predecessors - invaded India and Iran

But also worth mentioning: the Sintashta, or another late (i.e. CWC-derived) steppe culture very genetically close to Sintashta, invaded Armenia, replacing the Hurrians. And likewise, the Sintashta or a very similar steppe culture invaded Greece - the Mycenaeans are basically what you get if you mix minoans with sintashta (or, say, Srubna).


So what we're left with thinking, I think is that CWC was the OPPOSITE of Northwestern. Instead, it looks more directly associated with Graeco-Aryan!

[A superficial demonstration: the graeco-aryan branches are all associated with R1a, as is CWC. Whereas italo-celtic, like pre-CWC steppe cultures, were R1b]


But of course, it's worth remembering how few IE languages have survived. Other than Balto-Slavic (whose location until the last thousand years or so is totally unknown), no IE language family "originally" spoken north of the caucasus anywhere between Greece/Bohemia and Kazakhstan (i.e. Indo-Iranian Sintashta) (aside from a few scant attestations of Thracian) That's a vast area! And likewise, we don't have any IE language originally spoken in Western Europe - that entire area has been overrun by speakers of Italic, Celtic and Germanic (which itself is attested relatively late) - the earlier languages of Iberia, France, Britain and possibly even Germany are unknown.
[And of course we have no representatives of the IE migrations into Siberia (except possibly Tocharian?), and only a few dead relics of the IE migrations into anatolia.]
So any ideas of early IE dialectology are lacking most of the evidence they'd need...
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 »

Hmm, interesting. But there is another major branch of IE dominated by R1a: Balto-Slavic. Of course, Balto-Slavic and Indo-Iranian share a number of isoglosses, most prominently the satem shift and the ruki rule. Perhaps this "NW IE vs. Greco-Aryan" thing is vastly overrated as the "satem vs. centum" thing used to be. Indeed, the genetics seem to be more in tune with the "satem vs. centum" theory.

According to these maps, "South Yamna" (in the open steppe) was mostly R1b while "North Yamna" (in the forest steppe) was mostly R1a. R1b seems to have been brought into the Pontic steppe by the Khvalynsk culture, while Dniepr-Donets apparently was R1a.

Now, if Indo-Iranian is R1a/Corded Ware, and Greek perhaps also, this does not leave many languages for R1b: basically just Anatolian and Italo-Celtic. But Italo-Celtic IMHO originated in the Pannonian basin, even if those people were mostly R1a. Languages do not always go together with Y-DNA; there are two possibilities how populations with different Y-DNA haplogroup profiles may end up speaking the same language: language shift and language transmission through the maternal line (i.e., Mom and Dad speaking different native languages, and the children picking up Mom's language).

All this brought me to a new idea, of which I don't know yet what it is worth: Late PIE (PIE3 in my Adradosian numbered stage scheme) was spoken not by the Yamnaya but by the Corded Ware culture! Early Yamnaya/Pre-Yamnaya (earlier than CW) would have been Early PIE (PIE2), and Late Yamnaya (contemporary to CW) a different branch ancestral to Anatolian. Aquan, the hypothetical language of the Bell Beaker people, may belong there, too. I used to place Aquan as a branch that branched off even earlier than Anatolian, but the main characteristic feature of Aquan - the dominance of the vowel */a/ and the absence of qualitative (not so sure of quantitative) ablaut - can be explained by secondary loss of the distinction between *e, *a and *o - three vowels which would have huddled closer together in Early than in Late PIE, anyway. (Of course, we know so little about Aquan that we cannot really classify it - it is not even certain that it has anything to do with IE!)
...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
Howl
Sanci
Sanci
Posts: 69
Joined: Sat Oct 07, 2017 7:43 pm
Location: Campinia

Re: The Great Proto-Indo-European Thread

Post by Howl »

WeepingElf wrote: The early Neolithic farmers can't have brought PIE with them because they did not know wheeled vehicles and some other things which PIE has well-reconstructed words for (and hunter-gatherers are a fortiori out for the same reasons). The Yamnaya did know these things, and could have brought PIE to the Balkan Peninsula and the rest of Europe, and probably did.
You might want to check Google before making such a proposition. The first evidence of wheeled vehicles appears around 3500 BCE in three different cultures: Sumer (Mesopotamia), Maykop (North-Caucasus) and Cucuteni–Trypillia (Central Europe). So the Balkan was actually one of the first places to have wheeled vehicles.

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 »

Howl wrote:
WeepingElf wrote: The early Neolithic farmers can't have brought PIE with them because they did not know wheeled vehicles and some other things which PIE has well-reconstructed words for (and hunter-gatherers are a fortiori out for the same reasons). The Yamnaya did know these things, and could have brought PIE to the Balkan Peninsula and the rest of Europe, and probably did.
You might want to check Google before making such a proposition. The first evidence of wheeled vehicles appears around 3500 BCE in three different cultures: Sumer (Mesopotamia), Maykop (North-Caucasus) and Cucuteni–Trypillia (Central Europe). So the Balkan was actually one of the first places to have wheeled vehicles.
Sure, these all had wheeled vehicles, and are therefore not out of the question as PIE speakers on that ground. I did not question the Balkan hypothesis on the ground that they "did not have wheeled vehicles", what I did was to reject the Anatolian hypothesis. I am aware of the model according to which PIE came from Neolithic Anatolia, but the living IE language all descend from an intermediate node in the Balkans at a time when they already had wheeled vehicles. Fair.

Yet, we have good evidence of contact between PIE and Proto-Uralic, as the latter is full of loanwords from PIE. This makes a PIE homeland north of the Black Sea more likely. Also, as I said yesterday, the ancient Balkan Peninsula probably wasn't really much more linguistically diverse than ancient Italy. All those Paleo-Balkan languages are provisionally classified as separate branches of IE because they are too poorly known to make a final judgment on their classification. This is like those splinter genera (Orrorin, Sahelanthropus, Ardipithecus etc.) at the base of the hominin family tree: paleoanthropologists set up separate genera for all these species because they yet don't know how they are related and thus don't know which "real" genus which species belongs to. And lastly, a hotbed of diversity in a language family is likely to be near the original homeland, but it can emerge secondarily off the centre, e.g. if branch after branch moves into it.

In my opinion, the Maikop culture probably spoke Abkhaz-Adygean (NW Caucasian), and Cucuteni-Trypillian an extinct language related to that of the LBK culture. But that's little else than mere speculation.

As for the Corded Ware idea I posted earlier today, no, that is not my new opinion. It's just a "raw thought" that emerged in my mind when I considered what Salmoneus had posted. Further consideration led me to consider this idea possible, but not very likely. The conventional Yamnaya theory explains things better, even if the Y-DNA haplogroups seem to mismatch. (And the similarity of this idea with notions held about 100 years ago which since then have been abandoned for good reasons is disconcerting.)

All this tells probably is that correlations between language families, archaeological cultures and genetic relationships are far from perfect. Pots don't speak languages, they say, but genes don't speak languages either. No matter how you have it, you can't get around languages spreading to populations with different Y-DNA haplogroup profiles, either by language shift (people shifting to more prestigious languages) or maternal transmission (children acquiring their mother's rather than their father's language); both are known to happen often enough to explain such discrepancies.
...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
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 »

WeepingElf wrote:Hmm, interesting. But there is another major branch of IE dominated by R1a: Balto-Slavic. Of course, Balto-Slavic and Indo-Iranian share a number of isoglosses, most prominently the satem shift and the ruki rule. Perhaps this "NW IE vs. Greco-Aryan" thing is vastly overrated as the "satem vs. centum" thing used to be. Indeed, the genetics seem to be more in tune with the "satem vs. centum" theory.
I think this is quibbling over minutiae.
Obviously, there are complications in dialectology: Greek lacks satem, Balto-Slavic lacks augment. Which of these occured first - and hence which is the 'genuine' 'genetic' commonality and which is the later sprachbund effect doesn't seem a really important question to me.
According to these maps, "South Yamna" (in the open steppe) was mostly R1b while "North Yamna" (in the forest steppe) was mostly R1a. R1b seems to have been brought into the Pontic steppe by the Khvalynsk culture, while Dniepr-Donets apparently was R1a.
I don't think any of that can be said with confidence - I think that site is just guessing.
More generally: I don't think it makes sense to talk of R1b being introduced to the steppe by Kvalynsk (which was not ancestral to Yamnaya anyway). What makes you think it wasn't already there? R1b is known from Italy from 14,000 BP, and from the Baltic by 7,000 BP. It was all over the place.

[really important detail: when people talk about R1a and R1b in the PIE sense, they're specifically refering to specific male lineages that would have diverged only a century or two before the PIE expansions - so they can't, for instance, represent a European substrate (R1b in general has been in Europe forever, but the specific PIE clades that now dominate western europe weren't).]
Now, if Indo-Iranian is R1a/Corded Ware, and Greek perhaps also, this does not leave many languages for R1b: basically just Anatolian and Italo-Celtic.
Fairness is not the motivating factor here, so this doesn't really matter. [although you can also include Germanic in the R1b category].

This seems like an accident of history. The "R1b" languages of western europe have been wiped out by Romance and Celtic. [assuming Basque wasn't an original fellow-traveller]. The "R1b" languages of eastern europe have have been wiped out by Slavic. The "R1b" languages of Siberia have been wiped out by... well, everyone.
But Italo-Celtic IMHO originated in the Pannonian basin, even if those people were mostly R1a.
Were they?[/quote]
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
Circeus
Sanci
Sanci
Posts: 53
Joined: Thu Mar 11, 2004 2:15 pm
Location: Montreal, Canada

Re: The Great Proto-Indo-European Thread

Post by Circeus »

WeepingElf wrote:And lastly, a hotbed of diversity in a language family is likely to be near the original homeland, but it can emerge secondarily off the centre, e.g. if branch after branch moves into it.
Clearly you have no familiarity with the concept of evolutionary radiation. While radiation is better seen in linguistics within the field of semantics (indeed zompist himself basically brings it up in the Lexicopedia), it has also occasionally been invoked within the field of historical liguistics.
Golden age set the moral standard, the Silver Age revised it, the Bronze Age broke free of it and the Rust Age ran wild with it. -- A. David Lewis

We're all under strict orders not to bite the newbies. -- Amaya

Post Reply