European languages before Indo-European

Discussion of natural languages, or language in general.
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: European languages before Indo-European

Post by Salmoneus »

Yes, Neolithic Britain and Ireland were populated by migrants from the Near East. I can't immediately find anything saying whether this was primarily Cardial or LBK, but both groups were fairly closely related anyway.

EDIT: looking back at my previous update on this thread: apparently at least some neolithic British were of Iberian origin - i.e. the route was Cardial, not LBK. This could then be connected culturally to the 'Atlantic Megalithic' cultures. However, a) there may have been some migration from LBK as well, who knows? and b) Cardial and LBK were presumably different culturally (the former being very seagoing and megalith-building), but seem almost indistinguishable genetically. So their language families may well have been closely related (though, of course, they needn't necessarily have been).
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: European languages before Indo-European

Post by WeepingElf »

It seems as if LBK originated in Anatolia and Cardial in the Levant. The languages may have been unrelated; there seem to have been lots of small units in the Near East before Semitic and Indo-European took over (we have Hattic, Hurrian-Urartian, Sumerian, Elamite, all unrelated to each other or to anything else, and perhaps a few others that disappeared without leaving written records; and of course the fascinating mess we have in the Caucasus). Semitic probably entered the Levant too late to have been the language of the Cardial founders; indeed, the arrival of (Pre-Proto-)Semitic from Egypt may have been the event that set the Cardial founders on the move. Basque may be a Cardial language (then, Iberian would probably be so, too, and Basque and Iberian indeed related to each other), so the language of Neolithic Britain may have been related to Basque.
...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: European languages before Indo-European

Post by Salmoneus »

WeepingElf wrote:It seems as if LBK originated in Anatolia and Cardial in the Levant.

Citation? Because what I've gathered - as mentioned above - is that both came from Anatolia; we can deduce this because a) they are very similar, and b) they both have WHG genes right from the beginning, which were present in Anatolia but probably not any further south, so far as we know. This would also conflict with the third neolithic migration: the WHG-free migration into southern Greece, Cyprus, Crete, etc, which is hypothesised to come from the Levant.

In fact, Mathieson et al's "The Genomic History of Southeastern Europe" says the Cardial and LBK populations cluster so tightly with the Balkan Neolithic population that it seems as though there was only a single migration from Anatolia to the Balkans, which then later on (hundreds, perhaps a thousand years?) split into two (or three, if you count the ones who remained in the Balkans). This seems pretty plausible - if you're the north Balkans you follow the Danube, and if you're on the Balkan coast, then later exemplars show that a maritime lifestyle is easy to adopt.
The languages may have been unrelated; there seem to have been lots of small units in the Near East before Semitic and Indo-European took over (we have Hattic, Hurrian-Urartian, Sumerian, Elamite, all unrelated to each other or to anything else, and perhaps a few others that disappeared without leaving written records; and of course the fascinating mess we have in the Caucasus).
Anything's possible - just because two populations are almost identical doesn't mean they might not have different languages. Parsimony, however...

I think we also need to bear some other facts in mind...
- the first migration into Europe would have taken place something like three thousand years, or at least two thousand years, before Sumerian is attested. So the situation at the time of Hattic, Hurrian etc is not really relevant.
- Sumeria is a long way from western Anatolia; Elam is a bloody world away
- the Hurrians at least are documented as much later arrivals

- there were at least two migrations from (seemingly) the Caucasus and/or northwest Iran. The Pelopennese neolithic migration that lacked WHG possessed some CHG, unlike Anatolian neolithic samples. In other words, descendents of ancient caucasian hunter gatherers (whose genes are also found in early Iranian farmers and in modern Caucasians) migrated, probably via the Levant, into southern Greece. Later, however, Minoans and Myceneans (who were probably PIE people who had invaded and mingled with a local Minoan-adjacent native Greek population) have much more CHG, and so do Copper and Bronze Age anatolians. This could mean there was a second wave of migration that reached all the way to Crete (we might expect Minoan and Hattic to be related, then), or of course there could have been multiple waves. [Hurrian has been suggested as a Caucasian migration; likewise, one theory is that Anatolian IE was a migration from the Caucasus].

[unfortunately I don't think we have enough sample of the various bronze age civilisations to work out which group is which?]

It seems unlikely that Cardial languages, which departed before any genetic influence from the Caucasus is encountered, would be related to Kartvelian, but of course not impossible (Kartvelian may have been imposed on the local population more recently).
Semitic probably entered the Levant too late to have been the language of the Cardial founders; indeed, the arrival of (Pre-Proto-)Semitic from Egypt may have been the event that set the Cardial founders on the move.
I think Semitic enters too late? But I don't think we know for sure?
Basque may be a Cardial language (then, Iberian would probably be so, too, and Basque and Iberian indeed related to each other), so the language of Neolithic Britain may have been related to Basque.
That's certainly a possibility. Or, Basque may have been a steppe language (the only one of its branch to survive being drowned out by its IE neighbours), or Basque might be a survival from the Mesolithic. Or something else.
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: European languages before Indo-European

Post by WeepingElf »

Salmoneus wrote:
WeepingElf wrote:It seems as if LBK originated in Anatolia and Cardial in the Levant.

Citation? Because what I've gathered - as mentioned above - is that both came from Anatolia; we can deduce this because a) they are very similar, and b) they both have WHG genes right from the beginning, which were present in Anatolia but probably not any further south, so far as we know. This would also conflict with the third neolithic migration: the WHG-free migration into southern Greece, Cyprus, Crete, etc, which is hypothesised to come from the Levant.
What I posted was an attempt at a synthesis of various secondary sources I found on the Web, and apparently, I got some things wrong. The primary sources (geneticits's research papers) are, as I said earlier, too difficult for me, and the secondary sources are often imprecise, obsolete and sometimes just plain wrong.
In fact, Mathieson et al's "The Genomic History of Southeastern Europe" says the Cardial and LBK populations cluster so tightly with the Balkan Neolithic population that it seems as though there was only a single migration from Anatolia to the Balkans, which then later on (hundreds, perhaps a thousand years?) split into two (or three, if you count the ones who remained in the Balkans). This seems pretty plausible - if you're the north Balkans you follow the Danube, and if you're on the Balkan coast, then later exemplars show that a maritime lifestyle is easy to adopt.
Fine. So Cardial and LBK parted ways only on the Balkan Peninsula, which would make related languages spoken by them seem likely. What these were related to, is of course anyone's guess.
The languages may have been unrelated; there seem to have been lots of small units in the Near East before Semitic and Indo-European took over (we have Hattic, Hurrian-Urartian, Sumerian, Elamite, all unrelated to each other or to anything else, and perhaps a few others that disappeared without leaving written records; and of course the fascinating mess we have in the Caucasus).
Anything's possible - just because two populations are almost identical doesn't mean they might not have different languages. Parsimony, however...
You seem to prefer models in which large areas have been linguistically uniform even in prehistoric times and no linguistic lineage ever existed that isn't either spoken today or preserved in written form. Parsimonious, indeed, but correct? It seems to me that in prehistoric times, diversity was more like the norm than uniformity, before there were fewer forces at work that could enforce linguistic uniformity.

Sure, the linguistic situation of the Ancient Near East in the Bronze Age was probably not the same as in the Neolithic. But I think that those factors that lead to more language uniformity, such as trade networks and states projecting power over large areas, were stronger in the Bronze Age than in the Neolithic, so the Neolithic Near East may have been even more diverse linguistically. Alas, we don't know.
I think we also need to bear some other facts in mind...
- the first migration into Europe would have taken place something like three thousand years, or at least two thousand years, before Sumerian is attested. So the situation at the time of Hattic, Hurrian etc is not really relevant.
- Sumeria is a long way from western Anatolia; Elam is a bloody world away
- the Hurrians at least are documented as much later arrivals
Yes, the distances are considerable. There are also some scholars who attempt to link Hurrian-Urartian to Nakh-Daghestanian; I have no opinion on that because I don't know these two families well enough.
- there were at least two migrations from (seemingly) the Caucasus and/or northwest Iran. The Pelopennese neolithic migration that lacked WHG possessed some CHG, unlike Anatolian neolithic samples. In other words, descendents of ancient caucasian hunter gatherers (whose genes are also found in early Iranian farmers and in modern Caucasians) migrated, probably via the Levant, into southern Greece.
Yes, I had confused them with the Cardials.
Later, however, Minoans and Myceneans (who were probably PIE people who had invaded and mingled with a local Minoan-adjacent native Greek population)
The Myceneans spoke an early form of Greek. What the Minoans spoke is anyone's guess; their scripts have not been sufficiently deciphered yet. What the people in mainland Greece spoke before the Myceneans came (usually called "Pelasgian") is even more anyone's guess; it seems plausible to conjecture a relationship to Minoan, but this is uncertain. Well, we don't even know whether Eteocretan (an unintelligible language of Crete, written in Greek letters around 500 BC) was a descendant of Minoan or a later arrival from wherever.
have much more CHG, and so do Copper and Bronze Age anatolians. This could mean there was a second wave of migration that reached all the way to Crete (we might expect Minoan and Hattic to be related, then), or of course there could have been multiple waves. [Hurrian has been suggested as a Caucasian migration; likewise, one theory is that Anatolian IE was a migration from the Caucasus].
This makes sense. Minoan, "Pelasgian" and Hattic may have formed a single language family, probably unrelated to that of the Cardial and LBK people. What regards Anatolian IE, I consider a migration via the Bosporus more likely, but that is (1) uncertain and (2) beside the point here.
[unfortunately I don't think we have enough sample of the various bronze age civilisations to work out which group is which?]

It seems unlikely that Cardial languages, which departed before any genetic influence from the Caucasus is encountered, would be related to Kartvelian, but of course not impossible (Kartvelian may have been imposed on the local population more recently).
I seem to remember reading somewhere (but I don't remember where) that there is some evidence that Kartvelian originated in NE Anatolia, and was pushed into Georgia by the incoming Indo-Europeans, first Anatolian and later Armenian. But I don't know what's the evidence for that.

I would also never claim that the Cardial and LBK languages were related to Kartvelian. That's just an idea I spin in my conlang projects - possible, but highly uncertain.
Semitic probably entered the Levant too late to have been the language of the Cardial founders; indeed, the arrival of (Pre-Proto-)Semitic from Egypt may have been the event that set the Cardial founders on the move.
I think Semitic enters too late? But I don't think we know for sure?
Nobody knows for sure when Semitic entered the Levant. But this is pretty much beside the point as the misunderstanding that Cardial originated in the Levant has been cleared up now. Whenever Semitic entered the Levant, it has nothing to do with Cardial and LBK!
Basque may be a Cardial language (then, Iberian would probably be so, too, and Basque and Iberian indeed related to each other), so the language of Neolithic Britain may have been related to Basque.
That's certainly a possibility. Or, Basque may have been a steppe language (the only one of its branch to survive being drowned out by its IE neighbours), or Basque might be a survival from the Mesolithic. Or something else.
All this is possible. So we get these scenarios:

1. Basque could be a surviving Cardial language. Then the Cardial languages have nothing to do with Kartvelian (as Basque and Kartvelian are quite certainly unrelated), nor probably the LBK languages; and the language of Neolithic Britain may also be related to Basque.

2. Basque could be the language of some sort of steppe allies of the Indo-Europeans (perhaps from North Caucasia or southwestern Siberia), travelling on the latter family's coat-tails, so to speak. Such confederations of linguistically unaligned tribes have been in existence on the steppe for as long as we know; consider the Turkic tribes that rode alongside the Mongols, or the mottled retinue of Attila the Hun. (There even was a splinter of Ossetian in Medieval Hungary.) Then Basque would have nothing to do with the Cardial or whatever languages that were spoken in Neolithic Europe, and the Neolithic European languages are anyone's guess.

3. Basque is a surviving Mesolithic language, though the Basques seem to be descendants of Neolithic farmers. This is not impossible; as I observed here, surviving Mesolithic hunter-gatherers could have become warriors (it is not a long way from a hunter to a warrior: a hunter at least knows how to handle weapons and kill living things; he is better at such skills than a farmer) and established themselves as an élite over some farmers and imposed their language on them. Such things may have happened several times in different parts of Europe; you once wrote here that pre-Neolithic genes had a sort of "comeback" later. In this scenario, Basque has nothing to do with either the Cardial or the LBK languages.
...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: European languages before Indo-European

Post by Salmoneus »

WeepingElf wrote:
Salmoneus wrote:
WeepingElf wrote:It seems as if LBK originated in Anatolia and Cardial in the Levant.

Citation? Because what I've gathered - as mentioned above - is that both came from Anatolia; we can deduce this because a) they are very similar, and b) they both have WHG genes right from the beginning, which were present in Anatolia but probably not any further south, so far as we know. This would also conflict with the third neolithic migration: the WHG-free migration into southern Greece, Cyprus, Crete, etc, which is hypothesised to come from the Levant.
What I posted was an attempt at a synthesis of various secondary sources I found on the Web, and apparently, I got some things wrong. The primary sources (geneticits's research papers) are, as I said earlier, too difficult for me, and the secondary sources are often imprecise, obsolete and sometimes just plain wrong.
I agree, it's hard to get a handle on these things, particularly due to the lack of comprehensive overviews. I'd recommend that Mathieson paper, though - even if you don't ge through the fine print and the tables (and personally I can only grasp a little of those details myself), he lays out a pretty clear plain-English summary.
You seem to prefer models in which large areas have been linguistically uniform even in prehistoric times and no linguistic lineage ever existed that isn't either spoken today or preserved in written form. Parsimonious, indeed, but correct?
I think my default position is that in times of big cultural and genetic expansions, there are likely to be large language families; but that as time goes on, things may become more complicated. in the case of Europe, since the Glacial Maximum we have indeed seen a series of vast demic replacements that covered at the very least the majority of the continent, and I do think this is likely to suggest large, expansive language families - though some survivals in the mountains or peninsulas are not impossible.
Now, if we were talking about some parts of pre- or non-Bantu Africa, my default position might be quite different...

I also think it's worth pointing out that massive language families are the norm almost everywhere in the world. Take out Indo-European, Sino-Tibetan, Austro-Tai*, Pama-Nyungan, Turkic, Uralic, Afroasiatic (or just Semitic, really), Niger-Congo, Dene-Yeneseian and Amerind**, and you're not left with much of the map open. These expansive families show that even thousands of years after the initial expansion, familial relationships are still recognisable. So I think again that the onus is on those who want to hypothesise a very different situation in prehistoric Europe.

*I'm assuming Austronesian and Tai-Kadai (or whatever we call it) are related - this seems linguistically increasingly likely, and entire plausible from culture and genetics AIUI.

**Amerind is obviously a special case here (with the possible exception of Afroasiatic), because, as the family with the deepest language depth, with most of its expansion a very long time ago, and with in many cases limited attestations, and almost no historical records, it's not been possible yet (and perhaps will never be possible) to prove the relationship linguistically. However, genetically and archaeologically, it seems almost certain that Amerind is a single family with a time depth of 15-18k years. Even so, if you discount Amerind, if you just take families like Ge, Cariban and Tupi (which are probably branches of a single family), Arawakan, Uto-Aztecan, Algic, Siouxan and Iroquoian, you've already covered most of the Americas.


That all said, I do in general, all else being equal, think it's better to assume that where there's no evidence, there's nothing interesting, even if that leads to overlooking some lost languages, rather than speculating about hypothetical entities for which we have no evidence.

Sure, the linguistic situation of the Ancient Near East in the Bronze Age was probably not the same as in the Neolithic. But I think that those factors that lead to more language uniformity, such as trade networks and states projecting power over large areas, were stronger in the Bronze Age than in the Neolithic, so the Neolithic Near East may have been even more diverse linguistically. Alas, we don't know.
I think demic mass expansions paired with the expansion of cultural packages, particularly something as overwhelming as agriculture, are likely tied to linguistic expansion more often than not. Of course, I'm not saying that one village discovered agriculture and then conquered the world. But generally with these game-changing innovations, only a couple of groups in an area get in on the game early enough to benefit, in terms of cultural power.

What I suspect happened was that the early neolithic families expanded across a large area, but that after a certain length of time, neighbouring groups (perhaps mesolithic 'barbarians', or just newly neolithised peoples seeing the more advanced tribes across the river) back-migrated (invaded, either politically or just through economic migration) into that core area, creating a more confused picture.

So we'd see an expansion of people across anatolia, mesopotamia and the levant as agriculture spread, probably carrying one, or a small number at least, of language families. But later, newly neolithised tribes to the northeast (the 'Caucasian' element) and the southwest (Semitic) underwent their own expansions, both into the core neolithic area (proven linguistically for semitic, seen genetically for caucasian) and in other directions (caucasians expanded into eastern europe and central asia, semites back into africa). This I think is the situation we probably see in the historic bronze age. Since for thousands of years the core areas (mesopotamia, anatolia and the levant) would have been more developed than neighbouring areas, there was probably a longstanding dynamic of periodic barbarian invasions (Hurrians, Gutians, Mitanni, etc).

From the genetic point of view it's worth pointing out a concept called 'Basal Eurasian'. Basal Eurasians were the first non-Africans to split off from the non-African population - in other words, east asian and west european hunter gatherers were more related to each other than to basal eurasians. This is significant because according to some studies the neolithic populations of anatolia and europe were about 50% comprised of a 'basal eurasian' population that had remained isolated from all other populations for thousands and thousands and thousands of years - possibly a population hidden away in Arabia? We don't know. No actual 'basal eurasian' samples have been found, and it's possible that it's just an artifact of the analysis. But if there were such a population somewhere in the middle east, and they did suddenly expand to interbreed with their neighbours around the time of the neolithic revolution, it would favour the model of one group originally innovating agriculture and demically expanding with its export, at least at first. But this is all speculative.
This makes sense. Minoan, "Pelasgian" and Hattic may have formed a single language family, probably unrelated to that of the Cardial and LBK people.
I think this would be my default assumption at this point, yes, although I admit that it's pure assumption.
I seem to remember reading somewhere (but I don't remember where) that there is some evidence that Kartvelian originated in NE Anatolia, and was pushed into Georgia by the incoming Indo-Europeans, first Anatolian and later Armenian. But I don't know what's the evidence for that.
I don't know about that. But it certainly doesn't sound implausible.
Nobody knows for sure when Semitic entered the Levant. But this is pretty much beside the point as the misunderstanding that Cardial originated in the Levant has been cleared up now. Whenever Semitic entered the Levant, it has nothing to do with Cardial and LBK!
Well, it might. Semitic seems to have spread very quickly. It might have temporarily spread further (i.e. into Anatolia before being replaced by later migrants from the east), or it might have pushed the earlier inhabitants of the levant into anatolia, triggering a chain migration.
But I suspect they were too late to do that.
Basque may be a Cardial language (then, Iberian would probably be so, too, and Basque and Iberian indeed related to each other), so the language of Neolithic Britain may have been related to Basque.
That's certainly a possibility. Or, Basque may have been a steppe language (the only one of its branch to survive being drowned out by its IE neighbours), or Basque might be a survival from the Mesolithic. Or something else.
All this is possible. So we get these scenarios:

1. Basque could be a surviving Cardial language. Then the Cardial languages have nothing to do with Kartvelian (as Basque and Kartvelian are quite certainly unrelated), nor probably the LBK languages; and the language of Neolithic Britain may also be related to Basque.

2. Basque could be the language of some sort of steppe allies of the Indo-Europeans (perhaps from North Caucasia or southwestern Siberia), travelling on the latter family's coat-tails, so to speak. Such confederations of linguistically unaligned tribes have been in existence on the steppe for as long as we know; consider the Turkic tribes that rode alongside the Mongols, or the mottled retinue of Attila the Hun. (There even was a splinter of Ossetian in Medieval Hungary.) Then Basque would have nothing to do with the Cardial or whatever languages that were spoken in Neolithic Europe, and the Neolithic European languages are anyone's guess.

3. Basque is a surviving Mesolithic language, though the Basques seem to be descendants of Neolithic farmers. This is not impossible; as I observed here, surviving Mesolithic hunter-gatherers could have become warriors (it is not a long way from a hunter to a warrior: a hunter at least knows how to handle weapons and kill living things; he is better at such skills than a farmer) and established themselves as an élite over some farmers and imposed their language on them. Such things may have happened several times in different parts of Europe; you once wrote here that pre-Neolithic genes had a sort of "comeback" later. In this scenario, Basque has nothing to do with either the Cardial or the LBK languages.
Yes, I think those are the three most likely options. Regarding the third option:
- yes, mesolithic genes rebounded independently across the continent; we don't know if that was local interbreeding (the farmers finally starting to date the guys in the marshes) or some cultural change (farmer polities finally break down, inviting takeovers by big strong hunters). The interesting thing is that the mesolithic genes are much more often passed through the male line than the female - farmer girls had children with hunter boys. To modern eyes, this suggests the hunters had taken over in many placed - but we don't know enough about the cultures to really know. It could be that farmer matriarchs interbred with hunter slaveboys; or it could just be that gradually, over the centuries, the bigger, stronger hunters surreptitiously sexually outcompeted the smaller, paler farmer men despite the latter remaining dominant culturally.
- in any case, modern Basque genes are mostly from the Steppe - they're more mesolithic and more neolithic than most of europe, but they're not outside the general pattern. [the real holdovers are the sardinians, who are almost pure neolithic]. What this means is that genetically the Basque have been mostly replaced twice. If we assume that a neolithic language survived one turnover, we could also assume that a mesolithic language survived two.

Regarding the second option: I agree in theory, and I don't think this option should be overlooked, as it often is. However, one demerit is that we should remember that the steppe pastoralists of the neolithic were not the same steppe pastoralists of the iron age - they were less mobile and somewhat less warlike. We can't necessarily extrapolate the huge mounted multiethnic confederations of Attila and Temujin back onto the PIE. However, we can't necessarily rule out that option either.

We should also for completeness mention Option 4: Basque is a small-scale migration from north africa. i don't think we would necessarily be able to spot that genetically, particularly since the parent north african population is probably extinct now anyway, and given the thousands of years of african geneflow into iberia that followed. Basque is of course on 'the wrong side' of Iberia; but Iberian, presumably a macro-Basque language, is not.



Personally, I'd probably rank these options 1>2>4>3, although the exact ordering is just a guess really. I do think the most parsimonious option is that Basque is a survivor of the Cardial expansions and that yes, probably the pre-Bell-Beaker British would have spoken a language very distantly related to Basque.
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: European languages before Indo-European

Post by WeepingElf »

Salmoneus wrote:
WeepingElf wrote:What I posted was an attempt at a synthesis of various secondary sources I found on the Web, and apparently, I got some things wrong. The primary sources (geneticits's research papers) are, as I said earlier, too difficult for me, and the secondary sources are often imprecise, obsolete and sometimes just plain wrong.
I agree, it's hard to get a handle on these things, particularly due to the lack of comprehensive overviews. I'd recommend that Mathieson paper, though - even if you don't ge through the fine print and the tables (and personally I can only grasp a little of those details myself), he lays out a pretty clear plain-English summary.
I have taken a look at the paper. As you say, it is hard to grasp, but at least, the summary is quite clear and intelligible.
You seem to prefer models in which large areas have been linguistically uniform even in prehistoric times and no linguistic lineage ever existed that isn't either spoken today or preserved in written form. Parsimonious, indeed, but correct?
I think my default position is that in times of big cultural and genetic expansions, there are likely to be large language families; but that as time goes on, things may become more complicated. in the case of Europe, since the Glacial Maximum we have indeed seen a series of vast demic replacements that covered at the very least the majority of the continent, and I do think this is likely to suggest large, expansive language families - though some survivals in the mountains or peninsulas are not impossible.
I concur with you largely, at least what regards Europe. Do you remember the overview I posted in September in this thread (now also found here for easier reference)? Nice big families, essentially just half a dozen of them, including IE and Uralic. The conservative ("splitter") classification of North American languages, which I earlier used as a model for pre-IE Europe, is certainly not the final word. Most of those small units are rather shallow, with time depths in the range of 1000-2000 years, and some of the units proposed by the "lumpers" may be hardly deeper than IE. Also, there is little reason to assume that the diversity patterns were similar in those two regions.
Now, if we were talking about some parts of pre- or non-Bantu Africa, my default position might be quite different...
Yes, pre-Bantu Africa must have been a crazy quilt of small families - though most of them may have had clicks as an areal trait - almost everything non-Bantu south of the Equator, and some Bantu languages too, has clicks.
I also think it's worth pointing out that massive language families are the norm almost everywhere in the world. Take out Indo-European, Sino-Tibetan, Austro-Tai*, Pama-Nyungan, Turkic, Uralic, Afroasiatic (or just Semitic, really), Niger-Congo, Dene-Yeneseian and Amerind**, and you're not left with much of the map open. These expansive families show that even thousands of years after the initial expansion, familial relationships are still recognisable. So I think again that the onus is on those who want to hypothesise a very different situation in prehistoric Europe.

[footnotes snipped]

That all said, I do in general, all else being equal, think it's better to assume that where there's no evidence, there's nothing interesting, even if that leads to overlooking some lost languages, rather than speculating about hypothetical entities for which we have no evidence.
Fair - large families are the norm, hotbeds of diversity such as the Caucasus or New Guinea are rather the exception.
Sure, the linguistic situation of the Ancient Near East in the Bronze Age was probably not the same as in the Neolithic. But I think that those factors that lead to more language uniformity, such as trade networks and states projecting power over large areas, were stronger in the Bronze Age than in the Neolithic, so the Neolithic Near East may have been even more diverse linguistically. Alas, we don't know.
I think demic mass expansions paired with the expansion of cultural packages, particularly something as overwhelming as agriculture, are likely tied to linguistic expansion more often than not. Of course, I'm not saying that one village discovered agriculture and then conquered the world. But generally with these game-changing innovations, only a couple of groups in an area get in on the game early enough to benefit, in terms of cultural power.
I think we agree on this one, I think. Genes and languages do not always travel together, but they often do. Most people speak the native language of their parents. And in a demic expansion events, those migrants bring in both their genes and their languages.
What I suspect happened was that the early neolithic families expanded across a large area, but that after a certain length of time, neighbouring groups (perhaps mesolithic 'barbarians', or just newly neolithised peoples seeing the more advanced tribes across the river) back-migrated (invaded, either politically or just through economic migration) into that core area, creating a more confused picture.
Quite likely, the Mesolithic languages were reduced to small pockets, especially in mountain ranges, by the Neolithic demic expansion, but not entirely wiped out. The back-migration for which we have genetic evidence (if I understand this correctly) may have led to secondary expansions of such pockets.
So we'd see an expansion of people across anatolia, mesopotamia and the levant as agriculture spread, probably carrying one, or a small number at least, of language families. But later, newly neolithised tribes to the northeast (the 'Caucasian' element) and the southwest (Semitic) underwent their own expansions, both into the core neolithic area (proven linguistically for semitic, seen genetically for caucasian) and in other directions (caucasians expanded into eastern europe and central asia, semites back into africa). This I think is the situation we probably see in the historic bronze age. Since for thousands of years the core areas (mesopotamia, anatolia and the levant) would have been more developed than neighbouring areas, there was probably a longstanding dynamic of periodic barbarian invasions (Hurrians, Gutians, Mitanni, etc).
There is a number of what appear to have been Neolithic Wanderwörter shared by IE and Semitic, and partly also by Caucasian languages. These lexical resemblances are often cited as evidence of Nostratic or an IE-Semitic connection, but it is more likely that these words were adopted by the languages where we find them now. We have no idea what this "original Fertile Crescent Neolithic language" was, and whether there are attested or even living residues of it. (For a living residue, Kartvelian would IMHO the best candidate, but this is of course sheer speculation. OK, I admit it right-out: I am a Kartvelophile. That language family is just so rocking cool that ... but that is of course no valid reason to posit lost branches of Kartvelian all around!)
This makes sense. Minoan, "Pelasgian" and Hattic may have formed a single language family, probably unrelated to that of the Cardial and LBK people.
I think this would be my default assumption at this point, yes, although I admit that it's pure assumption.
I seem to remember reading somewhere (but I don't remember where) that there is some evidence that Kartvelian originated in NE Anatolia, and was pushed into Georgia by the incoming Indo-Europeans, first Anatolian and later Armenian. But I don't know what's the evidence for that.
I don't know about that. But it certainly doesn't sound implausible.
Nobody knows for sure when Semitic entered the Levant. But this is pretty much beside the point as the misunderstanding that Cardial originated in the Levant has been cleared up now. Whenever Semitic entered the Levant, it has nothing to do with Cardial and LBK!
Well, it might. Semitic seems to have spread very quickly. It might have temporarily spread further (i.e. into Anatolia before being replaced by later migrants from the east), or it might have pushed the earlier inhabitants of the levant into anatolia, triggering a chain migration.
But I suspect they were too late to do that.
Semitic indeed seems to have spread like bushfire through the Near East, and there may have been branches that were later lost again.
So we get these scenarios:

1. Basque could be a surviving Cardial language. Then the Cardial languages have nothing to do with Kartvelian (as Basque and Kartvelian are quite certainly unrelated), nor probably the LBK languages; and the language of Neolithic Britain may also be related to Basque.

2. Basque could be the language of some sort of steppe allies of the Indo-Europeans (perhaps from North Caucasia or southwestern Siberia), travelling on the latter family's coat-tails, so to speak. Such confederations of linguistically unaligned tribes have been in existence on the steppe for as long as we know; consider the Turkic tribes that rode alongside the Mongols, or the mottled retinue of Attila the Hun. (There even was a splinter of Ossetian in Medieval Hungary.) Then Basque would have nothing to do with the Cardial or whatever languages that were spoken in Neolithic Europe, and the Neolithic European languages are anyone's guess.

3. Basque is a surviving Mesolithic language, though the Basques seem to be descendants of Neolithic farmers. This is not impossible; as I observed here, surviving Mesolithic hunter-gatherers could have become warriors (it is not a long way from a hunter to a warrior: a hunter at least knows how to handle weapons and kill living things; he is better at such skills than a farmer) and established themselves as an élite over some farmers and imposed their language on them. Such things may have happened several times in different parts of Europe; you once wrote here that pre-Neolithic genes had a sort of "comeback" later. In this scenario, Basque has nothing to do with either the Cardial or the LBK languages.
Yes, I think those are the three most likely options. Regarding the third option:
- yes, mesolithic genes rebounded independently across the continent; we don't know if that was local interbreeding (the farmers finally starting to date the guys in the marshes) or some cultural change (farmer polities finally break down, inviting takeovers by big strong hunters). The interesting thing is that the mesolithic genes are much more often passed through the male line than the female - farmer girls had children with hunter boys. To modern eyes, this suggests the hunters had taken over in many placed - but we don't know enough about the cultures to really know. It could be that farmer matriarchs interbred with hunter slaveboys; or it could just be that gradually, over the centuries, the bigger, stronger hunters surreptitiously sexually outcompeted the smaller, paler farmer men despite the latter remaining dominant culturally.
Fair. The hunter takeover scenario would suggest that the community may have adopted the language of the Mesolithic new élite, which is less likely in the slaveboy scenario. The sexual outcompetition scenario is somewhere between. If you ask me, the likelihood decreases from hunter takeover via sexual outcompetition to slaveboys.
- in any case, modern Basque genes are mostly from the Steppe - they're more mesolithic and more neolithic than most of europe, but they're not outside the general pattern. [the real holdovers are the sardinians, who are almost pure neolithic]. What this means is that genetically the Basque have been mostly replaced twice. If we assume that a neolithic language survived one turnover, we could also assume that a mesolithic language survived two.

Regarding the second option: I agree in theory, and I don't think this option should be overlooked, as it often is. However, one demerit is that we should remember that the steppe pastoralists of the neolithic were not the same steppe pastoralists of the iron age - they were less mobile and somewhat less warlike. We can't necessarily extrapolate the huge mounted multiethnic confederations of Attila and Temujin back onto the PIE. However, we can't necessarily rule out that option either.
Yes. The Proto-Indo-Europeans were nothing like Huns or Mongols! They were sedentary agro-pastoralists. They had horses; but they also had houses and arable fields. The existence of agricultural and architectural terminology, the latter probably pertaining to more permanent structures than tents or yurts, in PIE proves that. Also, they apparently kept pigs, which are unsuitable to pastoral nomadism. And of course, their homeland north of the Black Sea is famously suited to agriculture. Attila's and Temujin's mottled retinues cannot be used as a model here!
We should also for completeness mention Option 4: Basque is a small-scale migration from north africa. i don't think we would necessarily be able to spot that genetically, particularly since the parent north african population is probably extinct now anyway, and given the thousands of years of african geneflow into iberia that followed. Basque is of course on 'the wrong side' of Iberia; but Iberian, presumably a macro-Basque language, is not.
This is an idea I overlooked! It is not certain whether Basque and Iberian are related, but they look very similar for "unrelated" languages. Some (though not all) Iberian names can be interpreted by means of Basque (e.g., the recurrent city name Iliberi, which looks a lot like Basque hiri berri 'new town' - Basque hiri 'town' could be a loanword from Iberian, but basic adjectives like 'new' are not as easily borrowed); while Basque comparison has not yielded a breakthrough in comprehending Iberian, there seems to be something to that. So the idea that Basque and Iberian originated in North Africa, where their relatives were later clobbered by Berber, makes sense.
Personally, I'd probably rank these options 1>2>4>3, although the exact ordering is just a guess really. I do think the most parsimonious option is that Basque is a survivor of the Cardial expansions and that yes, probably the pre-Bell-Beaker British would have spoken a language very distantly related to Basque.
Yes, the Cardial hypothesis is the most parsimonious; I would put the steppe confederation hypothesis lowest. This leaves the hunter backflow hypothesis (in its hunter takeover variant) and the North African hypothesis somewhere in the middle.
...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
mèþru
Smeric
Smeric
Posts: 1984
Joined: Thu Oct 29, 2015 6:44 am
Location: suburbs of Mrin
Contact:

Re: European languages before Indo-European

Post by mèþru »

I don't know about genetic origins, but the earliest LBK sites were in the area of Czechia and Hungary.
ìtsanso, God In The Mountain, may our names inspire the deepest feelings of fear in urkos and all his ilk, for we have saved another man from his lies! I welcome back to the feast hall kal, who will never gamble again! May the eleven gods bless him!
kårroť

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

Re: European languages before Indo-European

Post by WeepingElf »

What speaks for the Cardial hypothesis is further that the ancient toponymy of Sardinia seems to show affinity to Iberian and Basque. And isn't Sardinia considered almost purely Cardial? Yet, the very name of the island is reminiscent of Šrdš, the name of one of the "Sea Peoples" in Egyptian chronicles, who probably originated in the Aegean, and one may suspect a Paleo-Sardinian - Etruscan link on those grounds.

But as is well known, old place names can be deceptive, as the case of the Old European Hydronymy shows which were attributed to IE by Krahe, to Vasconic by Vennemann, both with arguments that are not easily dismissed, and considered to be the linguistic equivalent to ley lines by various other scholars. So the place name argument is not all that strong.

On the other hand, it may be that the Šrdš fared like the Franks in Gaul, or the Turkic Bulgars on the Balkan - they conquered the land, gave it their name, but lost their language.
...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

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

Re: European languages before Indo-European

Post by hwhatting »

WeepingElf wrote:On the other hand, it may be that the Šrdš fared like the Franks in Gaul, or the Turkic Bulgars on the Balkan - they conquered the land, gave it their name, but lost their language.
Well, as a minimum at least once - presumably, the didn't speak a Romance language before the Romans conquered them (if indeed the Šrdš have anything to do with the Sards) . ;-)

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

Re: European languages before Indo-European

Post by WeepingElf »

hwhatting wrote:
WeepingElf wrote:On the other hand, it may be that the Šrdš fared like the Franks in Gaul, or the Turkic Bulgars on the Balkan - they conquered the land, gave it their name, but lost their language.
Well, as a minimum at least once - presumably, the didn't speak a Romance language before the Romans conquered them (if indeed the Šrdš have anything to do with the Sards) . ;-)
That's not really what I mean. I mean that the Šrdš, whatever their language, adopted the language spoken on Sardinia before their conquest, whatever that may have been. Of course neither language was Romance! But as you say, the identification of the Šrdš with Sardinia is highly uncertain.
...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: European languages before Indo-European

Post by WeepingElf »

Something more on the "Basque problem". It seems as if I had misunderstood something in the news about Neolithic origins of the Basques. This Wikipedia page says that the Atapuerca skeletons whose DNA was compared to modern Basques are from the Late Neolithic to Early Bronze Age ("These individuals lived between 3,500 and 5,500 years ago"), from a time long after the Neolithicization of the Iberian Peninsula. So the "Basques descending from Neolithic farmers" is merely a truism, and there is no reason to connect them to a hypothetical "Cardial family" on genetic grounds. In fact, the Basques are so overwhelmingly R1b that a genetic connection to G2a-heavy people from Neolithic Anatolia is unlikely. So all those speculations about Basque being a Cardial language go out of the window! Also, there is no reason to assume that it is related to the language of the LBK farmers. If Basque is related to Iberian, it could be a Paleo-Mediterranean language. Or it could be Paleo-Atlantic.

And according to another Wikipedia page the British Isles were Neolithicized from Central Europe, so one would expect a Neolithic language there that is descended from the LBK language, unless a Paleo-Atlantic language survives there which may or may not be related to Basque.
...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: European languages before Indo-European

Post by WeepingElf »

Another take at the Bell Beaker people

I am currently involved in an interesting discussion with Salmoneus about the genetics and origins of the Indo-Europeans, and in the flow of that debate and associated research, I got a new idea about the Bell Beaker culture.

First, it appears as even if the Bell Beaker cultural package originated in the Iberian Peninsula, the Bell Beaker people outside the Iberian Peninsula descend from people from the Pontic Steppe, though probably from an early offshoot, earlier than PIE proper, while in the Iberian Peninsula, the genetic steppe admixture is slight to non-existent; only in the northwest of the peninsula, steppe admixture plays a significant role. However, the Bell Beaker people do not seem to descend from the Corded Ware people who probably were the ancestors of the IE branches of northern Europe.

So my new idea about the Bell Beaker people is this. A group of steppe people from the south of the Pontic steppe (which was dominated by the Y-DNA haplogroup R1b, while the north was predominantly R1a) spread out across central and western Europe in the Copper Age; they parted ways from those who would become the speakers of PIE proper around 4500 BC, which was the time of the first steppe incursion into the Balkan Peninsula, and fanned out across western Europe around 3000 BC. These are the speakers of the hypothetical Aquan language family, related to IE, and leaving traces in the Old European Hydronymy. Some of them ended up in the northwest of the Iberian Peninsula (where the OEH is present). Then, Bell Beakers arose in the peninsula as a cross-linguistic cultural phenomenon, which caught on in all the peoples of the peninsula, no matter whether Basques, Iberians, Tartessians - or Aquans. What then happened was that the Bell Beaker fashion spread out of the Iberian Peninsula among the Aquan-speaking communities of France, Italy, Britain etc., until it reached the eastern boundaries of the Aquan family.

However, the Bell Beaker people seem to have been a sort of elite or diaspora in Aquan-speaking lands. Apparently, the Bell Beaker people moved around a lot; for instance, the Amesbury Archer has been revealed of having grown up in Switzerland by strontium isotope analysis of his teeth. They thus probably were travelling merchants, trading in metal, salt and other commodities that weren't available everywhere. North of the Pyrenees, they would only venture where the locals would speak an intelligible language, i.e. an Aquan one, explaining why BB and OEH are largely coterminous outside the Iberian Peninsula.
...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
mèþru
Smeric
Smeric
Posts: 1984
Joined: Thu Oct 29, 2015 6:44 am
Location: suburbs of Mrin
Contact:

Re: European languages before Indo-European

Post by mèþru »

Sounds like an interesting story prompt
ìtsanso, God In The Mountain, may our names inspire the deepest feelings of fear in urkos and all his ilk, for we have saved another man from his lies! I welcome back to the feast hall kal, who will never gamble again! May the eleven gods bless him!
kårroť

Post Reply