February 21, 2013

Indo-Europeans galore

A whole bunch of papers in a journal I hadn't heard of before, but written by some of the best known people in this field. I see lots of "Buy pdf" links to accompany them, so if you want to spare me the expense, feel free to e-mail me a copy.

Anyway, here's a list of titles for reference:


Anthony, David Two IE phylogenies, three PIE migrations, and four kinds of steppe - p. 1-21
Balanovsky, Oleg; Utevska, Olga; Balanovska, Elena Genetics of Indo-European populations: the past, the future - p. 23-35
Blazek, Vaclav Indo-European zoonyms in Afroasiatic perspective - p. 37-54
Burlak, Svetlana Languages, DNA, relationship and contacts - p. 55-67
Dybo, Anna Language and archeology: some methodological problems. 1. Indo-European and Altaic landscapes - p. 69-92
Dybo, Vladimir Dialectal variation of Proto-Indo-European in the light of accentological research - p. 93-108
Gamkrelidze, Tamaz; Ivanov, Vyacheslav Indo-European homeland and migrations: half a century of studies and discussions - p. 109-136
Kullanda, Sergey Early Indo-European social organization and the Indo-European homeland - p. 137-144
Mallory, J.P. Twenty-first century clouds over Indo-European homelands - p. 145-154
Kornienko, Tatiana 'Archaeological Research in Northern Mesopotamia and North Caucasus' [Chronicle] - p. 155-162
Korovina, Eugenia 'Problems of the Indo-European Homeland' [Chronicle] - p. 163-166

UPDATE: Someone sent me the Anthony paper. Here is the main figure which suggests a correspondence between the linguistic phylogeny of Ringe et al. and the archaeological reconstruction of out-of-steppe movements of its author:


One of the weakest points of the steppe model is its treatment of Anatolians. Movement marked #1 perhaps takes some populations in an indistinct way into the east Balkans, but does not take them all the way to Anatolia.

The steppe model must propose some kind of early round-the-Black Sea mechanism to bring the Anatolians to their historical seats. This is not trivial; it might seem like a small jump from Thrace (where the endpoint of the first migration, marked with 1 is placed) to Anatolia, but it is in fact in the southern parts of the peninsula that we first find the Anatolian speakers. And, Anatolian place names are also recorded in Greece and its immediate environs, and not at all in the proposed migration route. Thus, the arrow must take the Anatolians further west (to account for the evidence from the Aegean, and then bring them south and east). I don't find this very plausible.


51 comments:

MGN said...

We really need open access in the sciences :(

Jim said...

"I don't find this very plausible."

I don't think anybody is very satisfied with it. But remember, all the same difficulties work in the opposite direction - if it's hard to get the Anatolians into position from Thrace, it's justa as hard to get the Indo-Iranians into position from Anatolia.

Dienekes said...

it's justa as hard to get the Indo-Iranians into position from Anatolia.

Why is it hard? It's a short way from eastern Anatolia to the Indo-Iranians.

The steppe model fails here too, because it posits two expansions along pretty much the same route for the Tocharians and the Indo-Iranians, and must bring the Indo-Iranians all the way from wherever the (dialect continuum) existed. It is arguably a much more difficult proposition.

Jim said...

"Why is it hard? It's a short way from eastern Anatolia to the Indo-Iranians."

Well yes. They can just as well have spread eastward as southward. The problem may be the deep diffenrences between Hittite and II, one would expect such proximity to make for continued ocntact, but who knows what actually happened. Geographic distance isn't the only kind of lingusitic distance. Choctaw and Chickasaw started diverging 500 years ago all while both langauges tayed tright next door. Right next door, but without any friendly contact. Something happened.

Come to think of it something like what I am suggesting did happen to split Iranian frorm Indic. There was some kind of religious split that happened, with the two groups swapping their terms for gods and demons, exactly mirror-imaged.

"The steppe model fails here too, because it posits two expansions along pretty much the same route for the Tocharians and the Indo-Iranians,"

Two migrations along the same route at different times will keep the speech communities separate enough to yield the separation we see.

"... and must bring the Indo-Iranians all the way from wherever the (dialect continuum) existed."

And if II, Germanic and Balto-Slavic were in fact in some dialect continuum, this would have provided the delay between the Tocharians and the II speakers. Speculation.

"It is arguably a much more difficult proposition. "

Neither of these models works particualrly well. That's why the question has stayed unsettled for so long. That's why you can still get all kinds of strange proposals -people arguing for an Out-of-Indai model.

Michael Boblett said...

Maybe I'm just misreading the chart, but it seems to place Italic and Celtic closer to PIE than Greek and Armenian. Did I get this wrong??? Anyhow, Dienekes, I can't find your post referring the relatively basal nature of Greek, but I recall it as pretty convincing and in line with what I recall of early IE dispersals. So mark me down as confused - and hoping I just got the whole thing wrong!

terryt said...

"One of the weakest points of the steppe model is its treatment of Anatolians".

I agree. Before I'd even read your statement I questioned the route offered for number 1.

"The steppe model must propose some kind of early round-the-Black Sea mechanism to bring the Anatolians to their historical seats. This is not trivial"

Again I agree. I would have thought the Anatolian group of languages must have entered through, or around, the Caucasus mountains in order to fit a steppe model. Perhaps at the Black Sea end of the mountains but not necessarily so. The remainder of the diagram largely makes sense to me.

terryt said...

Sorry. I didn't consider this statement:

"The steppe model fails here too, because it posits two expansions along pretty much the same route for the Tocharians and the Indo-Iranians"

I don't really see a problem with two migrations. Perhaps they were separated in time with some development of aridity at some time.

"and must bring the Indo-Iranians all the way from wherever the (dialect continuum) existed. It is arguably a much more difficult proposition".

Not really. Indo-Iranian and Balto-Slavic may have been at opposite ends of that dialect continuum with a third point in the continuum being the Celtic/Latin group in the west. The continuum had developed after Tocharian had developed independently somewhere else.

andrew said...

The case that Anatolian languages are much younger than indicated on the chart (perhaps 2500 BCE) but are divergent due to strong substrate influence seems more plausible, in light of the admittedly thin pre-Hittite historical record. Language evolution is far more punctuated than most linguistic models used to data linguistic time depth assume.

In terms of a direction of migration, a route via the Caucasus cultures, which were global leaders in metallurgy - something that the speakers of the Anatolian languages were renowned for their relative superiority in, seems more plausible than one via mainland Greece and the immediate vicinity (that extended into the Western Coast of Anatolia) - one would expect more linguistic similarity between Mycenian Greek and the Anatolian languages if they had arrived via this route.

Jim said...

"The case that Anatolian languages are much younger than indicated on the chart (perhaps 2500 BCE) but are divergent due to strong substrate influence seems more plausible, "

This the first time I have ever heard about a substratum in Aantolian and I wonder why, when it is so obviouly possible, to the poiont of the burden of proof being anyone trying to discount it. But even with a substratum effect, Anatolian has primitive features you can't explain with a substratum effect. It wasn't some substratum effect that dropped pharyngeals back into Anatolian anfter they had fallen out in every other group.

"In terms of a direction of migration, a route via the Caucasus cultures,..."

The standard objection to this is that the IE speakers left no trace in that region. Well, neither did the Wintuan speakers leave any trace in Achomawi or the Yanan languages when they pased through the Pit River country on their way from eastern Oregon into the Central Valley.

Slumbery said...

Dienekes

"Why is it hard? It's a short way from eastern Anatolia to the Indo-Iranians."

You try to find the autosomal trace of the IE expansion in Europe and earlier stated that you suspect it is West Asian.

Now, what could be the autosomal trace of IE expansion at its South-Eastern range? West Asian is not good, because its strong presence most likely pre-dates IE here. Our expectations depend on the assumed source of the migration.

The point is: a straight out of Anatolia migration would bring a significant Mediterranean signal into Pakistan/India. Not necessarily a big one, but at least significant. I looked up your Globe13 spreadsheet. There is no such a signal, the Mediterranean component is at noise level or barely above (and still could come from the Steppe, even the Uygurs are more Mediterranean that the SE IE). At the other hand the North European admixture is clear.

Note: Anatolia had to more Mediterranean before the Turks.

Based on the autosomal make-up of the South-Eastern IE,it is almost certain that it came from the Steppe, while an eastward migration from Anatolia seems to be unlikely. Anatolia as an ultimate source of the group is still in the play however, because a longer way makes more likely the loosing of the original genetic elements.

Va_Highlander said...

@Dienekes: "The steppe model fails here too, because it posits two expansions along pretty much the same route for the Tocharians and the Indo-Iranians, and must bring the Indo-Iranians all the way from wherever the (dialect continuum) existed. It is arguably a much more difficult proposition."

There could have been a millennium or more between migrations, so not sure why that is so difficult.

A bigger problem, to my mind, is the date Anthony is claiming for a Tocharian--Early-PIE split. 3300 BCE is centuries before significant nomadic pastoralism appeared on the Eurasian steppe. If they weren't nomadic, why did the forbears of the Afanasevo culture migrate, on foot no less, some thousands of kilometers to the east? Without direct evidence of migration, or motive for migrating, it's hard to see this as more than just a plausible story, at best.

Alternatively, we know that the peoples of southern Turkmenistan expanded eastward across southern Central Asia as far as Sarazm, in north-western Tajikistan, in the late fourth millennium BCE. Anthony has suggested that arid conditions west of the Tien Shan range would have presented a barrier to movement from Tajikistan to Siberia, but in reality that is far from clear. The general climatic trend in the region, over the past few millennia, seems to have been toward greater aridity. It may well be that the area was considerably more humid 5,000 years ago than it is today.

Anonymous said...

"The point is: a straight out of Anatolia migration would bring a significant Mediterranean signal into Pakistan/India. Not necessarily a big one, but at least significant. I looked up your Globe13 spreadsheet. There is no such a signal, the Mediterranean component is at noise level or barely above (and still could come from the Steppe, even the Uygurs are more Mediterranean that the SE IE). At the other hand the North European admixture is clear."

Wrong wrong wrong. The Dodecad results you are seeing are not any kind of formal admixture test. All you are seeing is that there is a West Asian cluster that is picking up most of the "western" aspect of Pakistan/India structure.

Formal admixture is a different animal entirely from what you are seeing from Dodecad.

Get AdmixTools up and running and give us some results, and we can talk.

pconroy said...

I'd love to see the whole paper, but I think the map gets some things right and some wrong.

1. Anatolian was probably an early break away by an elite, ruling over a large substrate of different origin.

2. Celtic and Tocharian (Centum) probably developed North of the Caucasus and spread West and East.

3. I agree with 3a - that Celto-Itali-Thracian spread from modern Romania through the North Balkans, across the Po valley into Liguria, and from there split up. P-Celtic spread North up the Rhone, then into Central Europe and NW Europe, Italic spread South into Italy, and Q-Celtic spread South West into Iberia and maybe Western France. Of course it could have departed by boat also and skirted around the Med to Western Europe too.

4. Later a great sound shift occurred in the core area, and we get Satem. They spread as Balto-Slavic into the North European plain and North of the Alps into Europe. So label 3b IMO is wrong, and should be Balto-Slavic

5. Indo-Iranian spread South of the Caucasus as Indic and East then South as Iranian, contemporaneously with Baltio-Slavic

6. As I've emnetioned many times over the last 10 years on this blog and Razib's, Germanic is best seen as a Baltic substrate with a Celtic overlayer. As even the word Deutsche is derived from a Celtic word for People (Teuta). So in my model it's only when Celtic journeys up the Rhone and East into Central Europe, does German emerge in the admixture zone.

MOCKBA said...

Could the Anatolia-Steppes connection be provided by the (now mostly submerged) shoreline corridor of Caspian West shore? It is mostly grassland & gentle terrain, and it's been used for attempted mass migrations as recently as in VIII c. by the Arabs in their (ultimately unsuccessful) Khazar campaign.

Estimated sea levels of the Caspian may have been as much as 10 m lower than in precent (Fig. 8.3 here)

Grey said...

"it posits two expansions along pretty much the same route for the Tocharians and the Indo-Iranians"

It may not apply to this case but generally i think the IE story may include an early foot-based expansion and a later horse-based one.

.
"If they weren't nomadic, why did the forbears of the Afanasevo culture migrate, on foot no less, some thousands of kilometers to the east? Without direct evidence of migration, or motive for migrating, it's hard to see this as more than just a plausible story, at best."

If hunter-gatherers in a region have a pop. density of n and adjacent sheep herders have a potential pop. density of 5n on the same land then i think the sheep herders would naturally expand into the H-G's terriotory.

If we imagine a time when the global population are all H-Gs and then farming starts in one spot involving both crops and domesticated animals then what would happen?

There will be terrain which is suitable for both parts of the package and some which isn't suitable (at the time) for the crop part of the package but is suitable for a herding based alternative - with a lower max. pop. density than the full package maybe but maybe still higher than the H-Gs.

In a situation like that the farmer segment will expand in the directions where both parts of the package are viable while the herder segment expands in the directions where the full package isn't (yet) viable but herding is.

(Obviously in reality the farmer/herder split would have been a spectrum but the basic idea is the same.)

http://en.wikipedia.org/wiki/Cattle_drive

"Patsy Durack, for instance, left Queensland for the Kimberleys in Western Australia in 1885 with 8,000 cattle, arriving with only half that number some two years and two months later, completing a drive of some 3,000 miles"

Now that's cattle but only two years ish to go 3,000 miles across an empty landscape.

How many years would it have taken sheep-herding Tocharians on foot (maybe carrying seeds with them) if the land they crossed at the time was effectively empty i.e. only had scattered foragers in a flat, open ambush-challenged landscape?

Ten years? Four? Three?

And if so then when would they stop - maybe when they hit an insurmountable obstacle - maybe like a region where the H-Gs were in unusually large numbers - like a large flood-plain with abundant foraged resources - like maybe the Wei valley.

###

I think the most plausible route - just from the geography - would be an initial (foot-based) startpoint somewhere in the mountains to the east of the first farming spot followed by going *around* the Caucasus mntns along the coast of the Black Sea.

The Captain said...

Why Albanian language is not included there?

eurologist said...

And if II, Germanic and Balto-Slavic were in fact in some dialect continuum, this would have provided the delay between the Tocharians and the II speakers. Speculation.

Thing is - they aren't, by any detailed mainstream research. In the same vein, the IE home region proposed here is huge and overly expansive in space and time - just to justify extremely unlikely migrations and language domination.

More importantly, there are so many contradictions with known archaeological and linguistic results that the model displayed in this figure really earns no credibility.

Proto-IE timings are pretty well known in Europe and don't adhere to this scheme - but more importantly, one of the riddles is that proto-Iranian is so close to PIE perhaps a thousand or more years, later. This certainly does not call for a northern Black Sea home range - just one migrating clockwise from the actual source region.

Anonymous said...

Another thing Anthony's map omits. Nomads have summer and winter camps.

Where did these people go in the winter?

Va_Highlander said...

@pconroy: "...Tocharian (Centum) probably developed North of the Caucasus and spread West and East."

Based on what evidence? What is it specifically about Tocharian itself that forces us to locate its origin north of the Caucasus and nowhere else?

@Grey: "...If hunter-gatherers in a region have a pop. density of n and adjacent sheep herders have a potential pop. density of 5n on the same land then i think the sheep herders would naturally expand into the H-G's terriotory."

Over thousands of kilometers? Are you serious?

Also, could you please point to this western sheep-herding culture that expanded east across the Eurasian steppe in the fourth millennium BCE?

"There will be terrain which is suitable for both parts of the package and some which isn't suitable (at the time) for the crop part of the package but is suitable for a herding based alternative - with a lower max. pop. density than the full package maybe but maybe still higher than the H-Gs.

I agree and how kind of you to strengthen my argument for me. It is, for this very reason, more likely that the Afanasevo were a pastoral extension of the settled agriculture of the Zarafshan Valley of north-western Tajikistan. Obviously, there was no such settled agriculture on the Eurasian steppe in the fourth millennium BCE.

"Now that's cattle but only two years ish to go 3,000 miles across an empty landscape."

While you're looking up the word, anachronism, I can only repeat myself: "3300 BCE is centuries before significant nomadic pastoralism appeared on the Eurasian steppe."

Slumbery said...

CJ

I know what are the those results and know there are many things invisible for that type of general analysis. However this does not mean I am wrong, just that the result I brought up are not enough to decide the question. Still they are a valid sign against an Anatolia -> India IE migration.

Shifting the burden of proving to me is an old lawyer trick, but this is a symmetrical situation. Why not take the Admixtools, do the analysis and prove your theory yourself? I do not have the time anytime soon, but I am looking forward to your decisive results and will happily learn about the truth.

You say the North European in India/Pakistan is a "noise" picked up from West Asian. I think no, it is too high for it and its distribution in the wider region is also againts this. This maybe rather intuitive without deeper analysis, but at this point you showed even less.

Valikhan said...

Va_Highlander, "significant nomadic pastoralism appeared on the Eurasian steppe. If they weren't nomadic, why did the forbears of the Afanasevo culture migrate, on foot no less, some thousands of kilometers to the east?" "3300 BCE is centuries before significant nomadic pastoralism appeared on the Eurasian steppe."

Even more. Because the first ever NOMADIC pastoralists where Scythians/Sakas.
While all other pastoralists before them were just pastoralists as were ancient Greeks, Anatolians etc. They built permanent houses and had no such thing as seasonal migration.

eurologist said...

pconroy,

Pretty much all modern European linguists agree that proto-Germanic condensed in north-central Germany (modern central/SE Lower Saxony, Thuringia, Saxony-Anhalt; IMO: Globular Amphora --> Únětice --> Northern Urnfield --> Jastorf cultures). This region, from the earliest through late Bronze and Iron Ages, has been archeologically demonstrated to trade with and receive cultural inputs from mostly their southern neighbors (the culturally, but not linguistically later Celts north of the Alps, probably speaking a pre-proto Celtic/Italic at the time). It is only at the inception that there were cultural and likely linguistic influences from the western (not N nor NE) Black Sea area (Globular Amphora Culture). So, there is a connection with the spread of IE into the Balkans and probably also with that of pre-proto Celtic/Italic westward at around the same time. However, there is no linguistic contact to Baltic nor Slavic during this time (and earlier contact to Uralic than to those two, at any rate).

Of course, Ringe's tree makes no sense, what-so-ever: there was no late II/Balto-Slavic/Germanic continuum (except see * below), and Armenian/Greek split off before Celtic/Italic.

I am a bit agnostic about Anatolian IE, but IMO there is much evidence that European IE, and likely also II, originated on the western Black Sea shores during their cultural peak with the first occurrence of a densely populated area with a preponderance of large, rich villages/ small cities as its core element. There was trade and cultural exchange across the Black Sea with Anatolia, but I don't know if that was uni- or bi-directional. Finally, I still entertain the idea that non-Cardium early European agriculturalist spoke some pre-proto IE from NW Anatolia/ SE Balkans. That would help explain (i) the old ages of Anatolian and Greek/Armenian IE, and (ii) the easy later acceptance of a more modern PIE in northern/central Europe, and (iii) the 20% to 30% of non-IE substrate in Germanic (it would date back to the "revenge of the mesolithic" - i.e., the crumbling of LBK and re-introduction of non-LBK haplogroups).

The early spit-off times of Greek/Armenian and Tocharian can be simply and reasonably explained by assuming that the core pre-PIE area was early-on split into southern (SW Black Sea), central (W Black Sea) and eastern (NW Black Sea) branches. The central branch can be identified with (*) above (and including parts of Celtic/Italic) - but at a much earlier time than usually proposed.

Andrew Lancaster said...

I must be missing something about the logic of this closing remark, even though everyone seems to agree:

"The steppe model must propose some kind of early round-the-Black Sea mechanism to bring the Anatolians to their historical seats. This is not trivial; it might seem like a small jump from Thrace (where the endpoint of the first migration, marked with 1 is placed) to Anatolia, but it is in fact in the southern parts of the peninsula that we first find the Anatolian speakers. And, Anatolian place names are also recorded in Greece and its immediate environs, and not at all in the proposed migration route. Thus, the arrow must take the Anatolians further west (to account for the evidence from the Aegean, and then bring them south and east). I don't find this very plausible."

Actually any model of IE's early dispersal has to show complications in the SE European area, because there is a diversity there. Is there truly any proposal which can get around that?

Grey said...

"Over thousands of kilometers? Are you serious?"

Yes. There is no difference between that and Polynesians getting on a boat and picking a random direction to sail in.

There would have been a moment in time when the first farmers - or some offshoot - reached the edge of the steppe *before* it was occupied by anything but small scattered bands of HGs when the steppe was effectively an open sea - closing up again afterwards as it became more densely occupied. During that window a tribe could have travelled very great distances in a few years.

.
"Also, could you please point to this western sheep-herding culture that expanded east across the Eurasian steppe in the fourth millennium BCE?"

I'm not talking about a specific time - except maybe through working backwards. I'm making a general point that there would have been a moment in time when the steppe would have briefly been an open road.

.
"While you're looking up the word, anachronism, I can only repeat myself: "3300 BCE is centuries before significant nomadic pastoralism appeared on the Eurasian steppe."

I'm not talking about nomadic pastoralism on the steppe. I'm talking about a group of farmers and sheep herders who set out onto the steppe looking for a good spot to settle but don't find one until they hit the other end.

Davidski said...

Neither ADMIXTURE nor formal mixture tests like ADMIXTOOLS will help much, because no one knows how to interpret the results.

Dienekes thought he discovered signals of a Bronze Age migration from West Asia to Poland when he used ADMIXTOOLS. In fact, he was picking up the belated spread of Neolithic influence from Western Europe to Eastern Europe.

Such west to east movements have now been confirmed by ancient DNA. But sadly most people taking part in these debates don't even know that Northeastern Europe received agriculture last, and population movements didn't go in straight lines, but were dictated by topography and climate.

I think only a massive study on the SNP structure within R1a will finally show us what we all want to see.

Things are looking great at the moment, with all the subclades fitting a post-Neolithic expansion of R1a1a from Europe to the east and southeast. But of course we need a formal paper to show this.

Nirjhar007 said...

For people whom are interested in Indo-Iranians and and surely Indo-European studies for a different attitude please check this article and website-
http://new-indology.blogspot.in/2013/02/indo-iranians-new-perspectives.html

Nirjhar007 said...

Davidski said-
''I think only a massive study on the SNP structure within R1a will finally show us what we all want to see.

Things are looking great at the moment, with all the subclades fitting a post-Neolithic expansion of R1a1a from Europe to the east and southeast. But of course we need a formal paper to show this.''
We just need the aDNA of you know which Harappan site to clear everything.

Va_Highlander said...

@Valikhan: "Even more. Because the first ever NOMADIC pastoralists where Scythians/Sakas."

I suspect you'll be hard pressed to find Scythians/Sakas as such 5,000 years ago. The Scythians were an Iron-Age phenomenon, not Chalcolithic. Nomadic pastoralism developed gradually during the Bronze Age between the Don and Volga and the Irtysh and Yenisei rivers. It seems unlikely that it was just one, specific group of people.

@Grey: "Yes. There is no difference between that and Polynesians getting on a boat and picking a random direction to sail in."

There are very significant and obvious differences. The Polynesians actually had a means of transport, were already a seafaring people, and, if their population was growing on limited land, had a plausible motive for migration. Since these are precisely the points at issue, it is extremely difficult to see your analogy as even remotely appropriate.

"I'm talking about a group of farmers and sheep herders who set out onto the steppe looking for a good spot to settle but don't find one until they hit the other end."

Actually, no, with all due respect you're talking about fantasies in your head. You have already admitted that you cannot point to a sheep-herding culture expanding eastward in the fourth millennium BCE. You are still perversely imagining farmers and nomadic pastoralists on the Eurasian steppe 5,300 years ago, despite having no empirical evidence supporting such a claim. And you have yet to acknowledge that there were actual farmers and sheep herders already living in western Central Asia at the time.

Jim said...

3300 BCE is centuries before significant nomadic pastoralism appeared on the Eurasian steppe. If they weren't nomadic, why did the forbears of the Afanasevo culture migrate, on foot no less, some thousands of kilometers to the east? "

Va_Highlander, it worked for the Apacheans - the Navajo and Apeaches. They got all the way form Alberta to New Mexico and Arizona wihtout horses. Nomadism doesn't have to be pastoral.

"Over thousands of kilometers? Are you serious?"

Well, yeah. "Thousands of kilometers" is not some impossible distance.

Grey said...

Va
"You are still perversely imagining farmers and nomadic pastoralists on the Eurasian steppe 5,300 years ago, despite having no empirical evidence supporting such a claim."

Nope.

When the first farmers expanded out in various directions from the original start point sooner or later a group of them would have hit the steppe at a point in time before the steppe was heavily populated.

1) At that moment in time the steppe would have been an open road.

2) They wouldn't have known how far away a good spot to settle was.

I'm talking about a Volkwanderung - nothing to do with nomadic pastoralism and how long would a migration to the Tocharian region have taken in those circumstances, 3-5 years?

It's just an idea but i think it's a plausible way of getting non-nomadic Tocharians across the steppe.

Grey said...

"3300 BCE is centuries before significant nomadic pastoralism appeared on the Eurasian steppe"

That's actually the point. There was no-one to stop them.

Valikhan said...

Va_Highlander said...
"I suspect you'll be hard pressed to find Scythians/Sakas as such 5,000 years ago. The Scythians were an Iron-Age phenomenon, not Chalcolithic."

That's my point, actually. I re-phrase it: "There was no such thing as nomadic pastoralism until Iron Age and Scythians".

I can't imagine Don-Dnieper pig-breeders being Nomads.

andrew said...

"A bigger problem, to my mind, is the date Anthony is claiming for a Tocharian--Early-PIE split. 3300 BCE."

The case for this kind of date is laid out by J.P. Mallory in his book on the Tarim mummies. In a nutshell, while the Tocharians appears in the Tarim basin ca. 2000 BCE (and are genetically and in terms of physical anthropology similar to peoples of the European Steppe rather than East Eurasian), their material culture suggests that this last leg of their migration was from an archaelogical culture to the north of the Tarim basin which they show strong continuity with, and that culture had a continuous presence there until ca. 3000 BCE to 3300 BCE. Hence, the long migration from Europe would have had to have taken place around 3300 BCE to 3500 BCE. This is one of the most solid constraints on the minimum age of PIE. But, since Tocharian had no substrate to assimilate because it was moving into basically virgin territory and was a fairly modest sized population isolated from other linguistic communities for a long time, we would expect it to be the most conservative of the Indo-European languages relative to PIE much like Icelandic, in similar circumstances, is the most conservative of the Germanic language relative to Old Norse.

Computational linguistic methods make conservative languages with low rates of linguistic change like Tocharian look relatively young compared to their real age and make languages that would have been under great pressure due to substrate influence, areal linguistic pressures and rapid societal change like the Anatolian populations look older than they really are by these methods.

Jim said...

I re-phrase it: "There was no such thing as nomadic pastoralism until Iron Age and Scythians".

I wonder about this. It's probably true. Certinaly horse and charot warfare is older than the Iron Age, and the IE speakers pretty much invented that. But everything from points to cattle-based pastoralism and maybe horse pastoralism with them, and I can't think of one other instance where cattle-based pastoralism is nomadic. I don't know what that is, or why it would be.

Dienekes said...

while the Tocharians appears in the Tarim basin ca. 2000 BCE

The identification of the people who lived in the Tarim basin 2000 BCE with the historical Tocharians of 2+ ky later is pure conjecture.

It is similar to speaking of Slavs or Vlachs in the Balkans during antiquity just because there are Slavs and Vlachs who live there at present.

Onur Dincer said...

It is similar to speaking of Slavs or Vlachs in the Balkans during antiquity just because there are Slavs and Vlachs who live there at present.

Vlach is originally an exonym used to designate the Romance-speaking populations of the Balkans and the adjacent regions to the north. Those who are called Vlachs by foreigners have used words derived from the word Romanus to designate themselves ever since the Antiquity. They are ethno-linguistically remnants of the Ancient Roman Empire and are in no way post-Antiquity peoples in the Balkans and the adjacent regions to the north.

eurologist said...

It is similar to speaking of Slavs or Vlachs in the Balkans during antiquity just because there are Slavs and Vlachs who live there at present.

As Onur alluded to, the name "Vlach" is of Germanic origin and originally designated peoples who had "sold out" to the Romans, but soon (in that form, rather than the related "Welsh") came to mean specifically the inhabitants of southern Romania and nearby Romanian-speaking populations - i.e., since antiquity.

Given linguistic (and cultural) continuation, there surely is also a lot of genetic continuation.

eurologist said...

I can't think of one other instance where cattle-based pastoralism is nomadic.

Jim,

Africa, perhaps? Or early Texans? At least, they drove their cattle to far-away train stations at (hopefully) minimal loss.

Half a century ago, when I was younger, older people would still try to bring kids to order by exclaiming: "Ich schick' Dich in die Wallachei! - " roughly translated as "I'll send you to Valachia."

The connotation being, a region that sold out to Rome but did not benefit even in a couple of millennia, or a troubled region not too far from our door-steps, but one I could actually manage to send you to.

Rob said...

all current models are incorrect. Maybe something is in the pipeline which will solve the outstanding issues . . .

andrew said...

"The identification of the people who lived in the Tarim basin 2000 BCE with the historical Tocharians of 2+ ky later is pure conjecture."

The identification of the people who lived in the Tarim basin of 2000 BCE with those ca. 600 CE when the Uygurs overwhelmed them in a manner resulting in ethnocide (i.e. demise of their culture without necessarily involving genocidal killing) is supported by multiple factors:

1. Continuity in ancient DNA from varied time periods in Tarim mummies.
2. Continuity in physical anthropology in preserved remains.
3. Discontinuity in physical anthropology between the late Tarim basin peoples and the neighboring Indo-Iranians.
4. Continuity in the Tarim basin material culture.
5. The lack of evidence of more than one round of admixture of the current population genetics of the admixed East Eurasian-West Eurasian population of the region today.
6. The lack of linguistic affinity between the Tocharian languages at the time that they were first committed to writing and the Indo-Iranian languages of neighboring peoples like the Sythians.

The continuity hypothesis is supported by considerably more than mere conjecture, even if it isn't established beyond any reasonable doubt.

Jim said...

Eurologist,

There may be some transhumance in the Sudan, don't know, but I have never heard of any in Kenya and Tanzania. And that could still be wrong too.

Texas - you are probably referring to the cattle drives. For one thing, those drives were not about raising the acttle, not really a lifestyle, but simply taking them to market. Typically the ranchers who raised the cattle were not the people who drove them. So it was pretty much divorced from the pastoralism of the ranch.

And the other thing is that cattle pastoralism, both in Europe and Africa, revolves around dairying. Texas cattle were only for beef, and in the early years, pretty stringy, low-quality beef from half-wild Texas longhorn cattle. Considering that these cattle ranged free foorm any human supervision or care for periods of months, the term pastroalism itself almost doesn't really apply.

Slumbery said...

Jim

IX-X. century Hungarians were cattle herders and they were pretty much considered to be something like semi-nomads.

On the other topic I completely agree with you. People do not need to be nomads or to have horses to migrate long distances.

Jim said...

Right, Slumberry, transhumance is a form of nomadism. But it has nothing to do with long range migration.

Dienekes said...

transhumance is a form of _pastoralism_ and nomadism is also a form of pastoralism.

Rob said...

"transhumance is a form of _pastoralism_ and nomadism is also a form of pastoralism"


According to Kazhanov, transhumance is not true nomadism (? so maybe not a form of it either).

Jim said...

"transhumance is a form of _pastoralism_ "

It may formally be called transhumance, but there is a non-pastoral form of something that looks a lot like transhumance. In Califonia proper and the Mojave Desert foragers used to go back and forth over their territoy seasonally as different plant foods came inot season. In th Mojave for instnace people used to go between the desert floor in winter to the mountain meadows in summer.

"According to Kazhanov, transhumance is not true nomadism (? so maybe not a form of it either). "

Makes sense to me and he damned sure knows more about than I do, whoever he is; you have to draw the definitional lines somewhere.

Slumbery said...

Jim
If you define nomadic so strictly, then you will find them really rare. It is a continuum.
Simple transhumance was what some of our people did just a few hundred years ago, even my Grandmother still had stories about the Summer camp of our village's cattle herders (though probably she just heard them too). IX. century Hungarians were more nomadic, though not fully nomadic if we use a very strict definition.
(The point is, I did not mean simple transhumance, when I said they were semi-nomadic.)

Jaakko Häkkinen said...

Dienekes, remember that Aryan DID NOT SPREAD TO THE EAST from Anatolia! Proto-Aryan developed in the Caspian steppes. Anatolian hypothesis is very weak compared to the steppe hypothesis:
http://www.elisanet.fi/alkupera/Problems_of_phylogenetics.pdf

eurologist said...

Aryan DID NOT SPREAD TO THE EAST from Anatolia! Proto-Aryan developed in the Caspian steppes.

I don't see a conflict with that. There is a huge gap in time between pre-PIE or PIE and proto-Iranian (PI), so it is actually easier to account for PI as entering from outside of the steppes. Tocharian made it far East of the Caspian - likewise pre-PI (while diverging from PII) could have moved counter-clockwise around the Caspian with plenty of time left to account for the Uralic loanword timings. Then, only later it moved back south into Iran and Afghanistan etc.

Unknown said...

Based of R1b1 maps, which concentrate in e. turkey , w. iran, and the lower caucauses, it is likely that this region is where IE's originated. The argument that it came from the steepes or europe, has long been been diffused.

Pneumatikon said...

Fascinating. My own Y-Chromosome came from Anatolia by way of Crete, and a consensus is emerging that the Minoans spoke an Indo-Europan language. Which means we must have spoken one when we arrived.

I don't think PIE came from Central Asia at all. I think it's Anatolian.