November 13, 2010

Linguistic musings

West European linguistic stuff that seems so obviously Basque that it hurts:

Italy-centered:
  • bi-, "Latin" preffix meaning "two" or "twice", very much extended to other European languages. < Basque bi (two), maybe via Ligurian or some other pre-IE Italian language. 
  • sardine and related Sardinia. < sardin(-a/-e) (sardine) < sarda (fish school) + -in ('doer', 'maker', probably from verb egin: to do, to make)
General West IE:
  • -er/-ero, suffix indicating profession (i.e. carpenter) < proto-Germanic *-arjoz OR Latin -arium < Basque -ari (professional suffix) < ari (auxiliar verb of action)  <> arin (quick, fast) <> aritu (to hurry)
English:
  • ill and kill - oddly similar to Basque hil (pronounced like ill, to die or to kill) < proto-Vasco-Aquitanian *kil (h, silent in the South but aspired in the North, seems to derivate from lost k - or g). Mediated by old Norse illr apparently in the case of ill (but not kill, whose etymology is uncertain).
  • to as particle for infinitive verbs (i.e. to do). While it might have a precedent in old Anglosaxon, it does not seem to have any other Indoeuropean cognate with this meaning of verbal particle. But in Basque it is one (and the most common) of three infinitive suffixes: -tu (pronounced exactly like to), -i and -n
I also considered ash (Basque auts) but this one seems to have legitimate Indoeuropean origins (Sanskrit asa and others) and could be a borrowing by Basque or whatever else (I'm not discarding remote shared origin of Basque and Indoeuropean, which would be less surprising than the unlikely Dene-Caucasian hypothesis, honestly).  

Another word I have wondered about in English is black, which looks similar to Basque beltz, same meaning (compound variant bel-/-bel as in beldur (fear) or urbel (blackwater, a river near Burgos), vide also bele: raven), which could be from a conjectural Vascoid root *belaK. However there's an alternative etymology related to blank and, via that word, to Romance blanc, blanco (white), via highly conjectural proto-Germanic *blakaz < PIE *blegh (no other known derivatives, it seems). So uncertain.

Iberian languages:
  • Castilian izquierda, Catalan esquerra, Galician/Portuguese esquerda < Basque ezker (left) or Iberian iskeŕ (argued to mean left or hand), and maybe related to Basque esker(-tu), meaning to thank (also eskerrak, eskerrik mean thanks as noun, pl. - never found in singular form, same in this aspect to what happens in other languages of West Europe: gracias, thanks...). Oddly enough some linguists reject this etymology biased by the negative perception of the left hand in Indoeuropean and particularly Latin (sinestra), which does not seem to exist in Basque (see the discussion at Vasco-Caucasian blog by promising Catalan linguist Octavià Alexandre)
If you have any ideas or criticisms, please let me know. I'm mostly actively exploring so I welcome them (Indoeuropean Paleolithic Continuity freaks please abstain: that hypothesis is plain nonsense). 

    68 comments:

    1. Couple of comments:

      -er suffix:

      also extremely common in German - pretty much any noun or verb can be transformed thus, with the meaning "a person doing that." Examples: Fussball --> Fussballer, fallen -> faellen --> Holzfaeller (fall -> to make fall/ to fell --> lumberjack), glas --> Glaeser, etc.

      May also be related to the meaning of a property derived from a noun:
      Holz -> hoelzern (wooden)
      Glas -> glaesern (made of glass)
      Eisen -> eisern (iron-like)

      So, not a person doing it, but the material itself, in a way ("its spirit"). I have no idea if there are related suffixes in Slavic or Greek.

      ---

      English black / bleak / bleach (German bleich/bleichen)

      IMO from IE:
      Latin flagrare, greek phlegein, German flackern

      so, something that is burnt turns black. Interesting that it also almost turns into white (blanco) in the bleak/bleach version, but I think that is secondary (that is, from the "bleak" meaning of black - not from the fact that something burnt can also turn white).

      I agree "ill" doesn't look much IE.

      - "to"

      there is of course the German "zu" (pronounced tsoo) that in some situations still is used the same, even though in General it is not required/used to indicate the infinitive: "Ich bin gewohnt zu gehen," "Ich mache dies um dir zu gefallen" (I am used to walk(ing); I am doing this to please you/so you like me). They also both share the meaning "towards."

      ReplyDelete
    2. So, to clarify, "bleak, bleach" stems from: when life turns black, it looses its color. When someone's face is "bleich" (German for pale), it has lost its color, and "to bleach" means to make the color go away - all not originally to turn white or make something white; that is just a side-effect.

      ReplyDelete
    3. Fair enough. The to/-tu thing may be a coincidence (or even a Frankish loan into Basque) but, as we can see via 'ill', that something has a Germanic origin does not make impossible to have an even deeper Vascoid root, maybe from Megalithic times (or whatever). However if a pan-IE root is found (and legitimate), as happens with 'ash', then the Vascoid substrate is most unlikely.

      Not being a professional linguist (and in fact not too inclined towards linguistics) I can't explore these threads but limitedly. However I find other stuff that correlates Basque and IE (or some IEs, specially Western IE) in the weirdest places. For instance pronouns have a vowel sympathy that is at least surprising:

      I - Basque 'ni', Romance 'io' (It.), yo (Sp.), je (Fr.) (< Lat. 'ego' but distinct from Portuguese 'eu', which is more conservative), also German 'Ich', etc.

      You - Basque 'zu' (sing), zuek (pl.), Sp. tú, Fr. tu - however this one is tricky because Basque zu originally meant you (pl.) and migrated to singular somehow. The original thou is 'hi'.

      Also some verbs:

      To be - Basque 'izan', compare with PIE *ist (from memory) and modern 3rd person sing. 'is' (Eng.), 'es' (Sp.), etc. While there are hypothesis that 'izan' is an IE import into Basque, its constant use in sentences with auxiliar verb like 'izan da' (it has been) or 'izango naiz' (I will be) makes this kind of unlikely.

      Also compare Basque 'gara' with 'we are', 'zara' with 'you are', 'dira' with 'they are'... Also Sp. 'eres' ('you are' sing.) The sound correspondences (and exactly same meaning) are quite striking.

      So either they represent a substrate in Western IE, a most remote connection Basque-IE or a most unlikely coincidence. Or a bit of all, which can only be discerned, if at all, with great care in analysis - something beyond my abilities.

      ReplyDelete
    4. And even weirder stuff I never bothered really researching:

      to say - Basque 'esan' (to say)

      to carry - Basque ekarri (to bring/carry to - contrasts with eraman: to bring/carry from). However notice Spanish "acarrear", which seems to stem from 'carro' (cart, chariot) - yet is used only for carrying on oneself (either by pulling or more commonly on the shoulders) or on an animal.

      It would require wide and deep research, IMO. Verbs specially are most intriguing.

      ReplyDelete
    5. It's a lot of fun, though - so please don't get discouraged, and I really did not want to sound negative at all with my (similarly lay) contributions.

      - blank in Germanic also means "devoid of color/structure; too bright to have recognizable color/structure" - similar to the "devoid of color" in bleak/bleich, rather than the later "white" meaning.

      - I don't think the "to" thing can be coincidence. In Germanic, the "towards" (in the direction of reaching a goal/ outcome) meaning is quite prevalent - as in constructs like "to show, to proove <-- um zu zeigen/beweisen (German)" --> towards a proof. So "to cut" is also "towards making/achieving a cut," etc. Is there a similar relation in Basque?

      Some of your examples of similarities make me think that Basque deviated from some proto-IE early on (long before the spread of agriculture?).

      ReplyDelete
    6. Honestly, I would become interested if some linguist launched a hypothesis gathering IE, NE Caucasian/Hurro-Urartean (and probably Sumerian in this group) and Basque, which would indicate a Gravettian era connection I presume. However so far IE and the rest have not been linked, possibly because of the (extremely conjectural but popular) hypothesis of Dene-Caucasian and Nostratic blurs the matter.

      It is I think reasonable to imagine a time when West Eurasian languages or at least some of them were one. But it's so remote in time that it's most difficult to dig so deep.

      ReplyDelete
    7. I don't know much of the subject, but if IIRC, Gaulish atta (father) is similar to the Basque atta (father) [I wonder why the Basques would have taken this particular word to the Celts].
      Maybe it's a sign that when IE arrived (either in central Europe or "France" (as I don't know if "atta" was only Gaulish or also the same in other ancient Celtic regions)) the local population had a name for father that was so similar to "pater" that it remained alive in the population.

      Or... maybe it's just a coincidence that 2 neighboring populations had ended up with the same word for the same thing.
      After all the Hittite word for "father" is "attas" while it's "tati" in Luwian and "papa" in Palaic.

      ReplyDelete
    8. Basque aita, maybe archaic ata. But the problem with such words is that they are nearly universal, as they are reinvented once and again by babies when they "learn" to speak (ama, apa, ata...) so it's not any valid connection.

      ReplyDelete
    9. I would suggest that aurignacians (Y-I, mt-U4/U5) spoke a vascoid language. On the other hand I would suggest that gravettians (Y-R1b/R1a, mt-H/V) spoke a proto-IE language.

      ReplyDelete
    10. At the moment it's not even 100% sure that Aurignacian-tech-using people were not Neanderthals...

      Also I do not think there was room for more than a single language family. After language is a tool: you use it to speak with your neighbors, so you need to speak the same your neighbors do. With all the people concentrated in SW Europe since the LGM, there was nearly no room for two language families. But who knows? The trend would have been anyhow convergence towards whatever was spoken in the most populated regions: Franco-Cantabrian area and specially Dordogne.

      There's always the possibility that all modern European languages spread in Neolithic and afterwards. This would make sense even without demographic replacement of any sort, because farmers' languages would become central for trade and such.

      ReplyDelete
    11. "At the moment it's not even 100% sure that Aurignacian-tech-using people were not Neanderthals..."

      Didn't know about that. I suppose you mean it is generally assumed but nobody is 100% sure. Is somebody actually challenging that view? 

      "Also I do not think there was room for more than a single language family. After language is a tool: you use it to speak with your neighbors, so you need to speak the same your neighbors do. With all the people concentrated in SW Europe since the LGM, there was nearly no room for two language families. But who knows? The trend would have been anyhow convergence towards whatever was spoken in the most populated regions: Franco-Cantabrian area and specially Dordogne."

      That's my point. The convergence would be what we call IE. Vasconic substrAte would differ in importance among the various branches of IE. With germanic being a creole, mostly vasconic speakers adopting a very different language. Language Continuity blog posted recently a 1998 paper by Marcel Otte that hipothesizes that basque missed the post-glacial expansion and was relegated to a secondary place.
         

      "There's always the possibility that all modern European languages spread in Neolithic and afterwards. This would make sense even without demographic replacement of any sort, because farmers' languages would become central for trade and such."

      I don't find it convincing. For me the strenght of Renfrew's theory was the correlation between population replacement and language replacement. Now that the former seems limited to very few spots the theory of a general language replacement seems untenable.

      ReplyDelete
    12. One of the linguistic points commonly attributed to a Vasconic substrate is the residual traces of a base twenty number system in English and some other IE languages. Hence, "eleven, twelve, thirteen, fourteen . . . nineteen," as opposed to ten-one, ten-two, ten-three as in Chinese. I don't know if there is any merit to this suggestion and do note that linguistic base twenty arose independently in Mesoamerica.

      ReplyDelete
    13. My working assumption is that cremation is a pretty good marker of the arrival of IE in most of Europe (including Greece) and India, with a few instances of cremation being adopted from nearby IE cultures (e.g. Etruscan). Thus, I find the evidence strongly suggestive of Urnfield cultures being the first IE language societies in most of Europe, and Cemetery H culture ca. 1900 BCE being a marker of IE culture's arrival in India. This would place the earliest IE cultures in Europe at ca. 2000 BCE in the Pannonian Plain and along the middle Danube which seems about right for the major expansion phase of IE societies as they transition from minor nomadic culture to future dominant European and South Asian culture. The severe drought in that general area around that time (also coinciding with the Akkadian Empire collapse) is a plausible trigger for that transition. I am inclined to favor the view that puts the Hittites in Anatolia not much more than a few hundred years before they are first attested in historical sources.

      Given the relative cultural continuity from the earliest European Neolithic until IE languages arrive, there is good reason to think that there was linguistic continuity as well and that the pre-Indo-European languages spoken immediately prior to Indo-European replacement largely descend directly back to at least the Neolithic, although it isn't inconceivable that there could have been perhaps one language shift in the early Bronze Age, between the two. In the Southern-Atlantic area there could have been a yet earlier language shift at the Neolithic, or earlier (perhaps during or post-LGM) in the Upper Paleolithic.

      The cultural evidence also seem to argue for three main language groups in early Neolithic Europe (without opining one way or the other about Upper Paleolithic roots): One Northern hunter-gatherer (maybe Uralic; maybe a lost language family or a language related to some relict Paleo-Siberian language like Yenessian), one for Southern Mediterrean-Atlantic areas now dominated by Y-DNA R1b (whether or not R1b was actually part of the founder group), and one for LBK areas now dominated by Y-DNA R1a (again without regard to whether R1a was part of the founder group).

      I think it is highly plausible that the LBK branch was related to Hurrian and other pre-IE Anatolian languages, with Northeast Caucasian plausibly being the closest related living member of that family.

      This would be suggestive of Basque, the Vasconic/Villanovan substrate of Western Europe, Etruscan and the other Tyrsenian languages (Rhaetic and Lemnian) sharing a linguistic family, quite possibly shared by Minoan. This family of languages may have had more time to develop regional distinctions if it is the original non-IE substrate at least back to the Neolithic if not older, since the LBK was relatively later. Thus, Basque and Etruscan might have been more different than Hurrian and "North LBK."

      This would also mean that Northeast Caucusian, while it might be the closest living language relative of Basque, might be only related at the distant macro-language family level that divided LBK Europe from Western and Southern Europe.

      Sumerian and Elamite would probably be related to one of these two language families. My intuition would be that both would belong in the LBK/North Caucasian side because of the overlap in related domains like the use of seals in commerce by Vinca, Sumerian and Harappan populations indicating a common starting point. I am also inclined to think that Harappan (although not Dravidian) were part of a Hurrian-LBK-Sumerian-Elamite-Hattic-Northeast Caucusian language family, given the IVC's likely origins and continued cultural ties (including the common use of seals).

      ReplyDelete
    14. [I broke my commments into three due to comment size limitations.]

      Your Western European-Basque correspondences, of course, fit the the idea of a widespread vasconic substrate in Western and Southern Europe.

      ReplyDelete
    15. Hi, Alberto.

      "Is somebody actually challenging that view?"

      Well it is not any strong opposition but the factual reality, which "Neanderfans" don't miss occasion to remind you, is that we lack of a clear fossil record between Mousterian (Neanderthals) and Gravettian (Cro-Magnôn subtype of modern human, and others like the more "Australid" Combe-Capelle skull). We have a couple of Neanderthal skulls in contested Chatelperronian contexts and we have an isolated contemporary AMH in Pestera cu Oase (without industrial context, though generally assumed to be proto-Aurignacian of some sort). There are also some teeth here and there.

      However, as Hoffecker points out, the Palestinian UP culture known as Ahmarian is clearly associated to modern Homo sapiens, what in turn brings us to other Aurignacoid industries. Another apparently solid evidence of H. sapiens in relation to Aurignacoid industries is found in Altai (while the previous Mousterian is associated to Neanderthals and the mysterious Denisova finger, IMO an H. erectus, at least by mtDNA lineage).

      So I do think that Aurignacian and in general Aurignacoid industries most probably attest the colonization of West Eurasia by H. sapiens but certainty is lacking.

      "That's my point. The convergence would be what we call IE".

      No. IE cannot be that old. IE is generally estimated to be some 6-10,000 years old. A lot more recent. The very pope of Linguistics (and Indoeruopean linguistics), Sergei Starostin himself claimed some time ago that linguistic research could not reach much deeper than some 10 Ka ago (though he gave no reason for that claim other than "because I think so").

      So whatever would be as old as the LGM or post-LGM expansion we should not be able to find any clear similitude and would remain obscure... unless linguistics advances a lot or, what is the same, someone proves Starostin wrong.

      Vasconic might be that old, but mostly because we know almost nothing about other Vasconics than Basque itself. It could also be more recent, Neolithic for instance.

      "Language Continuity blog posted recently a 1998 paper by Marcel Otte that hipothesizes that basque missed the post-glacial expansion and was relegated to a secondary place".

      Nonsense. Basque would have been absorbed, annihilated. There was no room for two language families. There is no room even today most probably.

      Ayhow, I already posted a warning against IEPC freaks. Ozpa! Which in colloquial Basque means: "get out of here".

      I am simply not going to bother with such fringe pseudo-science, seriously.

      "Now that the former seems limited to very few spots the theory of a general language replacement seems untenable".

      I knew Renfrew was wrong all along but he is still much more correct than you will ever be it seems to me. His hypothesis at least made some sense, even if it is not comparable at all with that of Gimbutas, which is the most solid model ever for IE spread.

      Gimbutas has a shrine in this blog.

      ReplyDelete
    16. "One of the linguistic points commonly attributed to a Vasconic substrate is the residual traces of a base twenty number system in English and some other IE languages. Hence, "eleven, twelve, thirteen, fourteen . . . nineteen," as opposed to ten-one, ten-two, ten-three as in Chinese."

      Andrew, the order between the single and ten-digits is reversed in different IE langueage, so it is irrelevant. E.g., in German "twenty-one" is "ein-und-zwanzig," etc.

      So, the only ones that stand out are eleven and twelve - and those, if anything, point to the still commonly used partial 12-(duodecimal)-system (a dozen, 5 dozen = 60 = three score (Grm.: Schock), 12 dozen = one "Gros", etc.)

      2,000 BC is way too late for IE "arrival" (if there ever was such a thing)in Europe - that's quite close to where I would put the proto-Germanic, proto-Italic, and proto-Celtic split - and everything at least East of the Rhine and north of (but including much of) the Alps already spoke IE for a long time.

      If LBK, as the currently limited mtDNA indicates, is of local Danubian origin, I have no problem with the idea that large areas of central and SE Europe already spoke a language at least very similar to proto-IE during that time - later including the late Bug-Dniester, when it changes to using LBK-type long houses form the subterranean stone houses. The isolation of this latter area for ~1,000 years or so may be the origin of the SATEM IE languages.

      We don't know if the non-IE languages in Italy before the arrival of Italic and Greek were related to Basque (seems at least likely) - but Etruscan is a late arriver, most likely from Anatolia.

      ReplyDelete
    17. "So I do think that Aurignacian and in general Aurignacoid industries most probably attest the colonization of West Eurasia by H. sapiens but certainty is lacking." 

      Very interesting thanks, I assumed that if chatelperronian by neanderthals was in doubt then aurignacian by modern humans should be sure. I see is more complex than that.

      "Nonsense. Basque would have been absorbed, annihilated. There was no room for two language families. There is no room even today most probably."

      Sorry I don't get it. Even excluding basque there are two different language families in Europe today. And basque was not alone before the roman expansion —rethian, etruscan iberian etc—. So I don't understand why you insist that there only could be one linguistic family.

      "Ayhow, I already posted a warning against IEPC freaks. Ozpa! Which in colloquial Basque means: "get out of here".
      I am simply not going to bother with such fringe pseudo-science, seriously."

      Didn't want to come as a troll. I mentioned Language Continuity blog cause I saw your comments there. And cited Otte's paper cause is strictly based on archaeological data, not linguistic.  Your blog, your rules so I'll don't continue with the matter. 

      I will not contest the last part as you obviously don't want to debate that.

      ReplyDelete
    18. "So I do think that Aurignacian and in general Aurignacoid industries most probably attest the colonization of West Eurasia by H. sapiens but certainty is lacking." 

      Very interesting thanks, I assumed that if chatelperronian by neanderthals was in doubt then aurignacian by modern humans should be sure. I see is more complex than that.

      "Nonsense. Basque would have been absorbed, annihilated. There was no room for two language families. There is no room even today most probably."

      Sorry I don't get it. Even excluding basque there are two different language families in Europe today. And basque was not alone before the roman expansion —rethian, etruscan iberian etc—. So I don't understand why you insist that there only could be one linguistic family.

      "Ayhow, I already posted a warning against IEPC freaks. Ozpa! Which in colloquial Basque means: "get out of here".
      I am simply not going to bother with such fringe pseudo-science, seriously."

      Didn't want to come as a troll. I mentioned Language Continuity blog cause I saw your comments there. And cited Otte's paper cause is strictly based on archaeological data, not linguistic.  Your blog, your rules so I'll don't continue with the matter. 

      I will not contest the last part as you obviously don't want to debate that.

      ReplyDelete
    19. @Eurologist: certainly Venneman argues for residual 20-base system in several European languages. The best known to me is French "soixant-dis" (60-10=70), "quatre-vingt" (4x20=80), etc. Which are identical to Basque numerals, just that in Basque we start at 30 (hogeitamar: 20+10, etc.)

      But I can't really discuss Germanic beyond English. Wikipedia lists Danish as the only Germanic example. Other Western IE cases are said to be in Celtic (again not anything I can discuss).

      However it is noticeable that the most common word for 20 in Brythonic (ugain, ugent) is an obvious cognate of Basque hogei. Actually I'd even consider tracking a proto-Vascoid root via the Breton ugain, which remainds a lot of "gain", which in Basque is summit, top (also gan, -gane often in toponymy) or "super-, over-", when used as preffix (gain-).

      Curiously, now that I am on it, related term Basque "gainditu", is to increase and might be related to IE gain/ganar. "Gain" has proto-Germanic roots but not clear PIE ones (Spanish "ganar" and French "gaignier" seem to be of Germanic origin).

      There are two pages of Basque dictionary beginning by gain-, so enough with this, I guess.

      "2,000 BC is way too late for IE "arrival" (if there ever was such a thing)in Europe - that's quite close to where I would put the proto-Germanic, proto-Italic, and proto-Celtic split - and everything at least East of the Rhine and north of (but including much of) the Alps already spoke IE for a long time".

      Agreed grosso modo. Though I'm not really sure if proto-Germanic is actually older, from Corded Ware (Single Burials) period maybe.

      Considering the data we have re. archaic European IE, for example Mycenaean Greek, which has a good amount of words that look more like Latin than classical or modern Greek, or the strange case of Lusitanian, which seems to be pre-Q Celtic or even pre-split Western IE maybe (again strange affinities with Latin in this and that word), I'd say that the Celtic (Q-Celtic) formation would not have happened until advanced Unrfield or even Hallstatt, what is more like 1000 BCE (roughly). I'd also say that Latin looks strangely conservative in the context of Western IE (of course classic Latin has been dead for almost 2000 years but still...)

      Very conjecturally, this might mean that the tribes speaking proto-Italic might have been less mixed (less creolized) than those speaking proto-Germanic and proto-Celtic. An often overlooked process that changes languages is creolization: when adults from a different background learn the language, unavoidably incorporating pronunciation, words and expressions, from their substrate, as well as possibly simplifying the grammar. A good example is the very simple English grammar but Romances also show that process in comparison to classic Latin, specially in the loss of declensions.

      The most conservative living IE seems to be Lithuanian, rather isolate since early IE expansion. In Germanic, at least in Norse, it seems to be Icelandic (again a remote geographically isolated language without any local substrate).

      ReplyDelete
    20. "If LBK, as the currently limited mtDNA indicates, is of local Danubian origin, I have no problem with the idea that large areas of central and SE Europe already spoke a language at least very similar to proto-IE"...

      I do have a problem with that. And it is that there is no known mechanism by which this conjectural PIE would have spread to the steppes, and from there to Southern Asia.

      That is a key weakness of Renfrew's model IMO. While we can see Kurgans advancing westward we do not see Danubian cultures advancing eastwards but to SW Ukraine at best. Even Corded Ware (which is Kurgan but anyhow) only spread to rather northernly latitudes, while Yamna-related cultures, independent from Corded Ware and such (or rather at its ultimate origins) dominated all the time the steppes... until Russia was formed in the Middle Ages (Slavic and related Varangian expansions).

      There's a technical impossibility for IE expanding from Central Europe to the steppes. And that really destroys the Neolithic expansion model, along with other problems like the known existance of many non-IE languages in West Asia and the Caucasus. The fact that Renfrew had to resort to Kurgans in order to explain Tocharian and Indo-Iranian, really says a lot about the limitations of his model.

      ReplyDelete
    21. @Alberto:

      "Didn't want to come as a troll. I mentioned Language Continuity blog cause I saw your comments there".

      I did post at Sanchís' blog but, if you noticed, all my posts questioned that IEPC model could make any sense. He got fed up of me and I got fed up of him. All he could say is "read my blog", what I had already done to a large extent at least without finding anything that could make sense.

      "And cited Otte's paper cause is strictly based on archaeological data, not linguistic".

      I did read Otte's paper in the past but didn't make much sense to me. In any case the main argument against IEPC is linguistic: it's too apparent a linguistic family to be that old. We should not be able to perceive so many and so clear evolutionary affinities if it was three times (or more) the usually accepted age.

      Of course, the main archaeological support, as far as I know, is very weak. It is highly Iberia-based and reverts all processes, so Iberian would have expanded from Catalonia and not the SE, Basque would have arrived as a most diffuse group of incineration practices, which are very limited in geography and have no clear relation to Urnfields (or anything I can imagine, except maybe the remote Chalcolithic Boléraz group), Celts would have been great sailors, which they are not known to have been, etc.

      There's no real model for Celtic expansion in IEPC, nor really for anything else. But who needs an expansion model if all languages were spoken locally since always... until historical times? Right?

      I call that: insulting the intelligence of people and manipulating their ethnic identity feelings in favor of a totally inconsistent ideological construct that has no basis either in archaeology nor linguistics.

      "Your blog, your rules so I'll don't continue with the matter".

      Yeah, better drop it because, while I am in fact very tolerant and won't probably class you as spam or start deleting your posts (unless you go too far), the IEPC model is a total rant. It reminds me of that guy who defends that Humankind originated in America, whom I had to eventually banish - not before long heated arguments, at my old blog and elsewhere.

      I'd rather not waste my time discussing what is obvious pseudo-science, really.

      ReplyDelete
    22. "I do have a problem with that. And it is that there is no known mechanism by which this conjectural PIE would have spread to the steppes, and from there to Southern Asia."

      Same as everyone else suggests. Except it would start from the Ukraine region SE-wards - just after IE acquisition.

      "There's a technical impossibility for IE expanding from Central Europe to the steppes."

      Not if you include the (late) Northern Ukraine region and Black Sea shores and view them as a late but very decisive (in the east) derivative.

      ReplyDelete
    23. I mean an archaeologically consistent cultural flow. There's nothing of that!

      In the Don-Dniepr region what we see is a locally evolved Neolithic (with whatever influences from West Asia or the Balcans but not a visible cultural flow) from local Epigravettian probably, it's called the Dniepr-Don culture, quite unoriginally. It was initially rather forager and only small-time farmer, though the apportion grows as time passes. It has different burial customs than Balcano-Danubian Neolithic (extended bodies, use of ochre - all very much Paleo) and a distinct rustic pottery with pointed bottoms (like Pitted Ware and the like, which are surely derived from Dniepr-Don).

      This culture persists without meaningful changes (other than the possible domestication of horse, now challenged by more ancient and clear evidence from Kazakhstan) until the Kurgan invasions: Seredny Stog II complex, which is actually an irregular mixture of Dniepr-Don and true Kurgan elements from beyond the Volga.

      We see then Kurgan elements pour into the Eastern Balkans and Central Europe but we do not see any Danubian cultural flow in all this time eastwards.

      The only exception would be a very late group in the Kiev region, which shows some Danubian influences on top of a very clear Dniepr-Don substrate, surely the last Dniepr-Don stand, then absorbed into Luboń (Kurgan) cultural area.

      There is no mechanism for an eastern expansion of Danubian cultural elements. It never happened: it's a pure fantasy. Non-steppary Europeans did not expand into the steppes before the Russian Empire did, at least not after Paleolithic times.

      "Not if you include the (late) Northern Ukraine region and Black Sea shores and view them as a late but very decisive (in the east) derivative".

      Can you point me to what culture (by name) are you thinking about? I understand that there is nothing like that.

      ReplyDelete
    24. @ eurologist

      "2,000 BC is way too late for IE "arrival" (if there ever was such a thing)in Europe - that's quite close to where I would put the proto-Germanic, proto-Italic, and proto-Celtic split - and everything at least East of the Rhine and north of (but including much of) the Alps already spoke IE for a long time."

      Why?

      Germanic expansion out of Denmark and Scandinavia largely dates after 750 BCE.

      The arrival of IE in Nordic areas is a plausible fit with either the early Nordic Bronze Age (1700 BCE) or more likely in my opinion given the cultural break in fit with the larger European scene with the Bronze Age generally predating distinctively IE cultural innovations, the late Nordic Bronze Age (1100 BCE). The late Nordic Bronze Age is when cremation begins to appear in Southern Scandinavia and many metal objects related to horses are first found.

      One very plausible read of what happened is that Bronze Age collapse weakened local cultures and spawned an IE diaspora, that spread across Europe and become the dominant culture in those regions. Germanic expansion, in turn, may have been driven by (per wikipedia cited above) "a small change in climate between 850 BC and 760 BC and a more radical one circa 650 BC brought in a deteriorating, wetter and colder climate (sometimes believed to have given rise to the legend of the Fimbulwinter)."

      The timeline of Celtic expansion is probably the best documented, and Maju is very within the mainstream in associating saying that: "I'd say that the Celtic (Q-Celtic) formation would not have happened until advanced Unrfield or even Hallstatt, what is more like 1000 BCE (roughly)."

      * As I note here, "In 753 BC Rome founded by Romulus according to Roman tradition. Rome was initially a small Indo-European language speaking city state dominated by the neighboring non-Indo-European speaking Estrucans of modern day Tuscany until around 500 BC when Rome assimilated the Estrucans into their state. By the 1st century CE Estrucans was a dead language. Ancient DNA evidence indicate the the genetic lineages associated with the Estrucans who received elite burials had vanished within a thousand years of Roman control. There were several small Italic language speaking states in Southern Italy around the same time which were probably founded around the same time and which were soon assimilated into Rome with their languages going extinct."

      Likewise, "1500 BCE- 1000 BCE The Balto-Slavic dialect ancestral to Proto-Slavic is estimated on archaeological and glottochronological criteria to have formed."

      To be clear, I'm not saying that IE languages arose in 2000 BCE. I am simply saying that prior to that point, from ca. 4000 BCE to 2000 BCE, that it is largely confined to Kurgan people of the Steppe and immediately adjacent areas, and the the main expansion beyonod the Steppe-Balkan-Aegean-Anatolian-South Asian area mostly takes off around 1300 BCE to 700 BCE.

      Linguistic dating to the extent that it is based on language change due to linguistic drift and doesn't reflect substrate influences and volatility in formative periods of a language is generally going to overestimate the age of a language. Archeological dating is more accurate and there are specific cultural signals that show up in that archeological data that fit the transition from pre-IE to IE culture.

      ReplyDelete
    25. "The arrival of IE in Nordic areas is a plausible fit with either the early Nordic Bronze Age (1700 BCE)"...

      My question in this matter is: is there a cultural break at the beginning of the Nordic Bronze Age or is it mostly continuity with Chalcolithic?

      It's a honest question because it is something I have not yet read about with some clarity. But my impression is that there were no further arrivals (of importance) at that moment, that proto-Germanic should be older than that (1300 years older: Single Burials). I may be wrong however.

      "Germanic expansion, in turn, may have been driven by"...

      ... at least partly Celtic weakness. As always, there's a push and a pull factor. The pull factor was the quasi-civilization forged in La Tène culture, with many polis (oppidae) arising everywhere and concentrating all wealth and power. That made the Germanic conquest a lot easier, as all they needed to do was to knock down (and plunder) those towns. That's at least what Krutas argues.

      "Ancient DNA evidence indicate the the genetic lineages associated with the Estrucans who received elite burials had vanished within a thousand years of Roman control".

      I don't think the known evidence says that, just that they are much less common than would be expected if these burials represented the mainstream Etruscan population and this population had continuity. Anyhow Romans did made some very harsh massacres in Etruria after they revolted, certainly suppressing Etruscan language at that point.

      "I am simply saying that prior to that point, from ca. 4000 BCE to 2000 BCE, that it is largely confined to Kurgan people of the Steppe and immediately adjacent areas, and the the main expansion beyonod the Steppe-Balkan-Aegean-Anatolian-South Asian area mostly takes off around 1300 BCE to 700 BCE".

      That's incorrect as far as I can tell. The Kurgan expansion to Central Europe is much older than that, from the late 3rd millennium BCE (and also to the Eastern Balkans).

      After a process of partial "Danubization", the Corded Ware expansion (just the culmination of earlier more modest expansions in Luboń and Globular Amphorae periods) really places all Central and Northern Europe within the Indoeuropean cultural and linguistic sphere.

      Then comes "the peace of the 1100 years" (Bell Beaker period and early/middle Bronze Age). And then the Urnfield expansion. Urnfield is not the IE expansion as such but just the expansion of a Western branch (Celts most clearly but probably also Italics and Illyrians).

      Meanwhile the Eastern IEs and surely also the less-well understood Balcanic branches were on their own.

      IMO the West-East IE split cannot be more recent than early Corded Ware and is likely quite older, with Western IE coalescing as the dialects of the "Western marches" since Seredny-Stog II, c. 3500 BCE.

      I'm uncertain on some details, specially regarding Eastern, steppary, groups but I suspect that the most direct, genuine, descendants of early IEs were steppe Indo-Iranians like the historical Scythians.

      ...

      ReplyDelete
    26. ...

      "Archeological dating is more accurate... "

      Necessarily. Same with genetics.

      "... and there are specific cultural signals that show up in that archeological data that fit the transition from pre-IE to IE culture".

      Not in the period you say. The correct timing for Kurgan expansion is, I understand, almost 2000 years earlier. And I am speaking based on archaeology.

      The first earliest IEs (not counting the various Balcanic branches, which have a different process) lived in East Germany c. 3000 years ago. The Baalberge culture, which has Kurgan burials and a town with clear concentration of "rural" wealth. As it is difficult to explain how Kurgan peoples skipped all Poland at that early moment, I conjecture they were called as mercenaries or something like that in the struggles that late West Danubians had among themselves. Their initial territory corresponds to that of a small, distinct, Danubian group of the Elbe, curiously the same one that Haak is always oversampling. Then they expanded to Poland and, for some time, Moravia.

      After losing Moravia tho the core Danubians of Baden culture (late Danubian renaissance), they split in two groups and it was the Polish group (around modern Warsaw) which was the most prominent since then (Luboń, Globular Amphorae and finally Corded Ware). This is the origin of Western Indoeuropean, as I understand it.

      ReplyDelete
    27. Maju,

      I agree with most of your dates. While I disagree about the linguistic impact of Kurgan, I (given time constraints) need to reserve a full discussion of that for later.

      Andrew, you seem to be - like unfortunately many people - following recent pseudo-science "Celtic" themes.

      I spend a few minutes and wrote up what I believe to be a reasonably valid, recent evaluation of actual evidence, here.

      ReplyDelete
    28. Hey, you finally started your own blog, Eurologist. Gratz! :)

      It's not that hard, right? I'll be your first follower (but you have to fix the follower gadget thing: just pick it from Blogger gadgets: it should work perfectly).

      "While I disagree about the linguistic impact of Kurgan, I (given time constraints) need to reserve a full discussion of that for later".

      Alright. My main point is that we do see Kurgan culture(s) groups advancing westwards and we do not see Danubian (LBK) culture(s) groups advancing eastwards at all. So it's really impossible that Danubian language(s) spread eastwards because there's no possible archaeologically-supported mechanism.

      Of course, one can always ponder how much linguistic (and even demic) influence this or that cultural flow had but, when we have no cultural flow whatsoever, we cannot expect any sort of linguistic nor demic flow.

      The same reasoning applies to a large extent to Neolithic demic replacement hypothesis, where there are several, and not just one, European Neolithic cultures, each one with their own dynamics.

      ReplyDelete
    29. Hmmm... now your gadget works. I got an error before but maybe it was a random one.

      ReplyDelete
    30. @ Maju : "Western IE coalescing as the dialects of the "Western marches" since Seredny-Stog II, c. 3500 BCE. "

      And indeed it seems to work well with the apparition of the Afanasevo culture in the eastern central Asian steppes. It's logical to spontaneously associate the Tarim's Tocharian language with this culture whose origin is from the north of the black sea (the haplogroups, both Y-DNA and mtDNA, also seem to link Xinjiang to the Altai/south Siberia).
      Since Tocharian is closer of the western branch (and of the Anatolian branch as well) than to the Indo-iranian branch, it fits well, I think.
      It's also the opinion of J. P. Mallory, I believe.

      "The first earliest IEs lived in East Germany c. 3000 years ago"

      Don't you mean 5,000 years ago (3,000 BC)?

      "I suspect that the most direct, genuine, descendants of early IEs were steppe Indo-Iranians like the historical Scythians"

      You mean culturally _the closest_ ones, right? Because the Scythian way of life appeared around 1,000-900 BC and is said to have appeared more likely in the eastern central Asian Steppes, near the Altai, If I'm not mistaken.

      ReplyDelete
    31. "Don't you mean 5,000 years ago (3,000 BC)?"

      Yes, that's exactly what I mean. My bad. :(

      "You mean culturally _the closest_ ones, right?"

      Culturally and maybe linguistically.

      "Because the Scythian way of life appeared around 1,000-900 BC and is said to have appeared more likely in the eastern central Asian Steppes, near the Altai, If I'm not mistaken".

      A bit unsure but I do not think so, as they are of Indo-Iranian language, not Tocharian. They should spawn from Andronovo in fact, in Kazakhstan and SW Siberia. Meanwhile Cimmerians would come maybe from the Srubna culture, which directly succeeds Yamna in Eastern Europe.

      ReplyDelete
    32. Maju : "A bit unsure but I do not think so, as they are of Indo-Iranian language, not Tocharian"

      Well that's what hints archeology apparently (the oldest culturally Scythian sites are in the north-east of the steppes - it's mentionned in Wikipedia but I also had already read it elsewhere before).

      As for the Tocharian language who knows what was its fate. If I understood correctly, first Afanasevo was "attacked" by populations from the east and soon after became part of the Andronovo group (all this occured around 1,700 BC) and I assume indo-iranian languages progressively became the norm in this region (there is Indo-iranian hydronymy in south Siberia, as far as the minusinsk region - sure it could still come from later scythian settlement).

      I do think the language of south Siberia was probably once "pre-tocharien" (Turkish öküz (ox) apparently related to Tocharian okso might be a lingering track of its presence in south Siberia) but I'm not sure it was still the case at the dawn of the Scythian era.

      "Culturally and maybe linguistically. "

      Interesting. So you're not of the opinion that there was a proto-indo-european language from which the wester IE languages and Tocharian conserved more and that the Satem languages were a later innovation?

      I'm not sure to understand well your view.
      Can you tell me a little more?

      If you consider that the original proto-indo-european language north of the black sea was maybe satem, then the south Siberian Tocharian hypothesis makes less sense, especially since this culture seems to be clearly derived from the culture of the black sea.

      Also the Anatolian branch which is apparently older than proto-indo-european language is classified Centum AFAIK.

      There seem to be some satem influence in Hittite but given the Mitanni case we can't exclude an Indo-iranian influence (related to the spread of chariotry and cavalry in this region).
      Mallory and others have pointed to the fact that the IE word for horse in this region seem to derive from the indo-iranian root, for instance (lycian "esbe", hittite "azu-wa", ugaritic "ssw" all resemble quite a lot the indo-iranian "asva").

      ReplyDelete
    33. "I spend a few minutes and wrote up what I believe to be a reasonably valid, recent evaluation of actual evidence, here."

      Not very impressed as it is not sourced, not dated and vague.

      ReplyDelete
    34. "it's mentioned in Wikipedia"...

      I do not see it. It mentions two, maybe conflicting, maybe complementary theories: one placing them in Central Asia and the other in East Europe. I am not able to make a clear distinction between Scythians, Sakas, Parthians, Persians and Medes (these all see nothing but Scythian offshoots)... and even not very clearly with Cimmerians either, though these may be more related to Thracians, depending who you read.

      "So you're not of the opinion that there was a proto-indo-european language from which the wester IE languages and Tocharian conserved more and that the Satem languages were a later innovation?"

      I do not know. Not everybody seems to agree that Tocharian is centum, the centum/satem divide is not the same as the Western/Eastern IE divide (leaving aside many smaller groups: Tocharian, Greek, Albania, Hittite...) and many argue that Satemization was a mere sound fashion. There are many possibilities in this and it's very possible we are contemplating a mere sprachbund areal influence in the satem issue, which anyhow is anything but central to IE linguistics.

      "I'm not sure to understand well your view. Can you tell me a little more? "

      All I can say is that Balto-Slavic is Western IE and is satem, so satemization (or centumization) is something not directly related to IE branching but an areal feature.

      It is even possible that satem arrived to Poland with the diffuse offshoot group of Catacombs that seems obscurely related to the genesis of Corded Ware or that for reasons of closer contact and lesser levels of admixture "Polish" IEs kept the original satem sound, centum being maybe a substrate influence.

      It's impossible to tell and the best is to leave the satem/centum issue somewhat aside.

      "Also the Anatolian branch which is apparently older than proto-indo-european language is classified Centum AFAIK".

      In Wikipedia Anatolian (Hittite) and Tocharian are not classified clearly by the centum/satem isogloss, at least not in the main map.

      In the other map Tocharian appears as centum but Anatolian is unclassified by all the various issoglosses. Balto-Slavic is consistently grouped in these issues with Eastern IE but we know that it is part of Western IE, so not all is a matter of simple sound shifts.

      ReplyDelete
    35. @ maju :

      "it's mentioned in Wikipedia" - "I do not see it"

      http://en.wikipedia.org/wiki/Scythians#Archaeology_2

      "From the 8th century BC to the 2nd century BC, archaeology records a split into two distinct settlement areas: the older in the Sayan-Altai area in Central Asia, and the younger in the North Pontic area in Eastern Europe."


      Maju about centum/satem : "which anyhow is anything but central to IE linguistics."

      It is when it possibly give important hints about the process and chronolgy of indo-europeanization and validate (or not) theories.
      How could Tocharian be something else than related to satem if satem was the norm, north of the black sea as you seem to suggest.
      I can't picture it.

      Once again, Afanasevo seems directly derived from the north of the black sea (ceramics, cultual objects, sepultures, economy, metallurgy (interestingly silver metallurgy as well) - even the physical type is the same, according to Russian and Chinese specialists, a robust "proto-europoid" type, opposed to the more gracile mediterranean type of the south-east of Europe or the mongoloid type in east Asia/Siberia).

      maju : many argue that Satemization was a mere sound fashion. There are many possibilities in this and it's very possible we are contemplating a mere sprachbund areal influence

      OK, but how does it invalidate my earlier observation about Tocharian/Afanasevo confirming the usual Kurgan model (satem occuring later from a 'k' basis)? I'm not sure to follow your chain of thoughts.

      ReplyDelete
    36. maju : "All I can say is that Balto-Slavic is Western IE and is satem, so satemization (or centumization) is something not directly related to IE branching but an areal feature."

      The absence of this areal feature in Tocharian * might tells us something on the chronology of the process though (coupled with what archeology has to tell us).

      * http://en.wikipedia.org/wiki/Tocharian_languages#Phonemes

      "Phonetically, Tocharian is a "Centum" Indo-European language" - (anyway 100 in Tocharian A : känt ; and Tocharian B : kante - but see below for a more thorough comment)

      maju : In Wikipedia Anatolian (Hittite) and Tocharian are not classified clearly by the centum/satem isogloss, at least not in the main map.

      I was not clear enough, by centum I was simplifying things to go fast.
      My point : the oldest "IE" languages were using a K not a S (logical, K->S is a simplification, it rarely goes the other way around). It is enough to assume which form spawns from the other one.

      Anyway :

      http://en.wikipedia.org/wiki/Centum-Satem_isogloss

      "Tocharian combined all rows into a single velar row by merging the palatovelars into the velars as in Centum and the labiovelars into the velars as in Satem, allowing two points of view: either Tocharian does not fit the model[5] or it is a Centum language.[3] The Proto-Anatolian language apparently did not undergo either the Satem or the Centum sound change, as the velar rows remain separate in Luwian.[6] The closely related Hittite is a Centum language, which it may have become secondarily, but the exact sequence is unclear."

      Even if we put Anatolian and even Tocharian out of the centum/satem model **, it's still work perfectly with the proposed pattern as they would be the two most ancient preserved IE languages.
      Satem would still look like a later change, the oldest Asian IE movement from the north of the black sea to south siberia sharing more with the west IE languages - all centum - (Common characteristics with tocharians are also found in Greek, Armenian, Phrygian and the Anatolian languages) than with Indo-iranian languages, thus seemingly dismissing either the existence of Satem or the coexistence of "centum-like" and satem dialect in the general area of north of the black sea (as satem would likely be in the east - where Tocharian would logically come from) at the time of the first IE migrations, like you seem to hypothesize.
      No?

      ** (let's not forget that the first tracks of Tocharian are 500 AD, about 4,000 yrs after the origin of Afanasevo the language had plenty of time to undergo some mutations if both this language and this culture were connected; as for Anatolian its first recorded tracks are several thousands after it separated from the ancestral IE homeland, logically)

      ReplyDelete
    37. I was looking in the other Archaeology section:

      http://en.wikipedia.org/wiki/Scythians#Archaeology

      Which says:

      "... two broad hypotheses. The first, more popularly supported, theory roughly follows Herodotus' (third) account, stating that the Scythians were an Iranic group who arrived from Inner Asia, i.e. from the area of Turkestan and western Siberia.[16][13][17]"

      "A second school of thought suggests an authochthonous development from within the Pontic steppe/ trans-Caucasian region. Followers of this theory argue that the Scythians emerged from local groups of the Timber Grave culture (broadly associated with the "Cimmerians")"...

      Each section seems to have been edited separately by different people in different times. The Central Asian theory anyhow, seems to have more sources, what is not necessarily better but at least suggests a extended backing.

      "How could Tocharian be something else than related to satem if satem was the norm"...

      The truth is that we know very little about Tocharian language(s). Also we do not know if satem was part of the original IE package or something that spread later by areal influence. We are talking of many (3-4) millennia, enough for many linguistic changes to happen, specially in languages in active and repeated expansion. A good example can be Latin, which soon was radically transformed by its own expansion (into Vulgar Latin, which is the true ancestor of Romances).

      At the moment I assume that centum is not part of any non-European language. Still it may require some explanation how it arrived to the Balcans or whatever... but not my job: linguists get paid for trying to explain and understand this kind of stuff.

      "Afanasevo seems directly derived from the north of the black sea"...

      Might be. But I am not aware of all this stuff (links?). If it is indeed original from the Black Sea rather than from East of the Volga, then its hypothetical affinity with Western IE languages (alleged centum) would be better explained.

      "a robust "proto-europoid" type"

      That sounds trivial because Central Asia was not yet Turkified, so we should expect Eastern Europeans and Central Asians to be more or less the same stock back then.

      "how does it invalidate my earlier observation about Tocharian/Afanasevo confirming the usual Kurgan model (satem occuring later from a 'k' basis)? I'm not sure to follow your chain of thoughts".

      You are putting all the emphasis on the satem/centum phenomenon. I instead tend to put it aside and follow the main philological West, East and other branches of IE instead. There is a broad consensus in this philology (i.e. East/West dichotomy disregarding the satem/centum issue, which overlaps but is not defining), even if the downstream structure and the place of Balcanic, Anatolian and Tocharian languages is not always agreed upon, though these are almost without doubt old distinct branches on their own right.

      ReplyDelete
    38. "My point : the oldest "IE" languages were using a K not a S (logical, K->S is a simplification, it rarely goes the other way around). It is enough to assume which form spawns from the other one".

      I do not have to agree with this. Earliest IE may have got a variety of dialects and we are not in position to discuss which was dominant where and when. I do not see clearly how K can become S, nor viceversa. They seem a most odd phonetic change to my mind really. (K>G, K>T, K>CH... all make some sense but K>S?) Maybe the original sound was something like CH, as in "cheese", which can evolve in both directions? It seems to demand a transition of that kind, because K<>S really seems most forced.

      Another possible transition would be via TH, as in "think".

      But in any case I feel it demands a substrate that alters the sound. For example Castilian C/Z (TH sound, not K!) becomes S in many Southern Iberian and Latin American dialects but the key seems to be the pre-Castilian substrate in Andalusia which removed the TH sound (or in some rarer cases the S into TH).

      It never changed in core Castile. So I'd say that centum is the innovation, if any, because I'd expect more phonetic conservatism in the core steppe area, not more shifts. Once the shift happened, maybe in Seredny Stog II, then it spread easily, as happened with the "mispronunciation" of Spanish C/Z into S, now much more widespread than the original sound.

      But well, let these complexities to professional linguists better. I do not think they alter the overall picture the least.

      ReplyDelete
    39. About Scythians : "Each section seems to have been edited separately by different people in different times. The Central Asian theory anyhow, seems to have more sources, what is not necessarily better but at least suggests a extended backing. "

      Maybe, I don't have a strong opinion on the subject. I had previously read the north-eastern hypothesis somewhere so when I came across it again in wikipedia I assumed it was the dominant view.
      I don't know enough on the subject to have certitudes about it. Any theory is fine with me, even though I admit the Scythian origin outside of the Altai would make more sense to me too for the same reason you gave (probability of a "Tocharian" substrate in south Siberia).

      About Afanasevo directly linked to the north of the black sea : "Might be. But I am not aware of all this stuff (links?)"

      I had read it in a Mallory book. I'll try to find an English version on Books.google.com later, if it's available.

      "There is a broad consensus in this philology (i.e. East/West dichotomy disregarding the satem/centum issue,"

      I know but anyway, AFAIK, the nature of the Tocharian language as a whole put it closer of the west IE+anatolian languages than to Indo-iranian apparently, so it would still be meaningful.

      "Maybe the original sound was something like CH, as in "cheese", which can evolve in both directions?"

      Yes, I admit it makes sense but given the omnipresence of the 'K' sounds everywhere in the fringes (west, east and Anatolia) and in the likely oldest IE languages, I can't help leaning towards the original 'k' hypothesis, personally.

      "They seem a most odd phonetic change to my mind really. (K>G, K>T, K>CH... all make some sense but K>S?)"

      Maybe so. Not sure about that. In French you have a similar phenomenon (partially). ca sounds "ka", co sounds "ko", cu sounds "ku", but ce is pronounced "se" and ci is pronounced "si" (while they originally were "ke" and "ki").
      And is it really stranger than G->Z in Russian (German gold, Russian zoloto; hittite gima (the original proto-IE word is hypothesized to start by a 'G' in the reconstruction I've seen), Russian zima; Ancient Greek Gnosis, Russian znanie (the words mean gold/winter/knowledge)) or even T->S (English water, German wasser; English white, German weiss, etc...) or W->G (Chinese wo -> japanese ga, Chinese wu, japanese go, etc...).

      "let these complexities to professional linguists better"

      Indeed, in the end, that's what we are reduced to.

      ReplyDelete
    40. "In French you have a similar phenomenon (partially). ca sounds "ka", co sounds "ko", cu sounds "ku", but ce is pronounced "se" and ci is pronounced "the nature of the Tocharian language as a whole put it closer of the west IE+anatolian languages than to Indo-iranian"

      Actually what you quoted above rather suggests that Tocharian had some centum and some satem elements so either it was a half-centumized satem or a half-satemized centum.

      Or maybe it was the original thing? It's generally considered the first IE to branch out, so maybe it does indicate that there was no such centum/satem dichotomy until later.

      Or was it from the intermediate zone... or who knows... we can really drink many beers while we speculate about this but I doubt we would ever arrive anywhere.

      "ca sounds "ka", co sounds "ko", cu sounds "ku", but ce is pronounced "se" and ci is pronounced "si" (while they originally were "ke" and "ki")".

      Just like in Spanish and, AFAIK every Romance, just that in Spanish it's not si/se but thi/the... except in the South and in America...

      In Italian it's K/CH etc. For some reason CE/CI were decided not to be KE/KI in Vulgar Latin but something else (yet this something else varies quite a bit).

      "And is it really stranger than G->Z in Russian"...

      Shouldn't this be the same as K/S? Part of the same satem/centum phonetic change?

      ReplyDelete
    41. "And is it really stranger than G->Z in Russian"...

      "Shouldn't this be the same as K/S? Part of the same satem/centum phonetic change?"


      Erm... I guess so. It was late :)

      ReplyDelete
    42. The really important points about Tocharian in my view to putting the puzzle of Indo-European languages together are:

      (1) There appears to be strong population and cultural continuity from ca. 1800 BC to 800 CE in the Tarim Basin, with a possible infusion of Indo-Iranian influence only late in that time period.

      (2) Tocharian is not an Indo-Iranian language and the physical type of the people and their culture also support its non-Indo-Iranian affiliation.

      (3) The case made by Mallory & Mair (2000) that the Tocharian languages were introduced to the Tarim and Turpan basins from the Afanasevo culture to their immediate north, is quite solid based on multiple lines of evidence, which puts a Proto-Tocharian language in Krasnoyarsk Krai at 3500 BCE - 2500 BCE.

      (4) Afanasevo culture and the Yamna culture (ca. 3500 BCE to 2200 BCE) are very similar in a variety of ways including burial practices, and essentially contemporaneous.

      (5) The Celtic-Tocharian cultural connections (e.g. plaids) are thin enough to support a conclusion that they have roots in relatively thin, Silk road trade connections rather than constituting a true common culture. The connection is too thin culturally or linguistically to support these contacts (or Indo-Iranian contacts) as a source of transition to an IE language from an earlier non-IE language.

      Thus, (1) the archeology supports a Tocharian branch split from the rest of IE around 3500 BCE, (2) this necessarily puts proto-IE no later than 3500 BCE, and (3) it substantiates the notion that IE languages were indeed a Kurgan innovation as opposed to a post-Kurgan linguistic shift of Kurgan peoples.

      Also, note that Mallory & Mair (2000) fail to reach a definitive conclusion about how Afanasevo culture arrived in Krasnoyarsk Krai, although the coincidence with the domestication of the horse is suggestive as a motivator for the expansion East. The source of Tocharian knowledge of irrigation agriculture is also not clear although there are some possible explanations.

      ReplyDelete
    43. I think I can agree with all you say here, Andrew.

      Specially I'd like to emphasize your conclusions:

      "(1) the archeology supports a Tocharian branch split from the rest of IE around 3500 BCE, (2) this necessarily puts proto-IE no later than 3500 BCE, and (3) it substantiates the notion that IE languages were indeed a Kurgan innovation as opposed to a post-Kurgan linguistic shift of Kurgan peoples".

      Yes, and it also says quite strongly (as Tocharian is in all phylogenies AFAIK the earliest language to branch out) that PIE did not expand anywhere (at least with continuity to historical times) earlier than that 3500 BCE date: that there was a single PIE language and ethnicity with that language (whatever minor dialectal variants) c. 4000 BCE, when the Kurgan cultural phenomenon was still restricted to the Samara valley.

      Tocharian language(s) and their origins at the remote Afanasevo culture are in fact a great calibration point for Indoeuropean linguistics. Thanks to them we can settle the matter with what I think is absolute certainty (there's more evidence anyhow but this evidence alone seems to be able to settle all the pointless discussions on IE origins in favor of the Kurgan model).

      ReplyDelete
    44. This comment has been removed by the author.

      ReplyDelete
    45. "The Kurgan expansion to Central Europe is much older than that, from the late 3rd millennium BCE (and also to the Eastern Balkans).

      After a process of partial "Danubization", the Corded Ware expansion (just the culmination of earlier more modest expansions in Luboń and Globular Amphorae periods) really places all Central and Northern Europe within the Indoeuropean cultural and linguistic sphere."

      My read is that Pit-Comb Ware culture was non-Indo-European and probably Uralic, and that Sredny Stog culture (ca. 4500 BCE-3500 BCE) is the culture in which IE languages come into being, probably as a hybrid of populations from the Corded Ware culture (or its predecessors) and the Cucuteni-Trypillian culture (ca. 5500 BCE and 2750 BCE), which was a direct cultural (and probably linguistic) continuation of the LBK. In this hybrid the women tended to come from the North and the men from the South.

      My take would be that Corded Ware (ca. ca. 2900 BCE – 2350 BCE) is really more a successor to the Pit-Comb Ware than to Sredny Stog, and hence would also have been Uralic. Since, Corded ware involved a "new culture and physical type" and "was clearly intrusive, and therefore represents one of the most impressive and revolutionary cultural changes attested by archeology" in its Western expanse, which provides a plausible mechanism for genetic discontinuity between LBK and current Europeans in the Northern LBK range (although probably not enough of a replacement to explain the demise of Y-DNA N1a relative to the decline of other parts of the LBK population genetic mix.

      Thus, Proto-IE is probably a Sredny Stog creole of Uralic and the language of the LBK. In this view, IE starts to seep into Central Europe beyond Ukraine only starting with the collapse of Cucuteni-Trypillian culture in the 3rd millenium.

      So, I will grant that there may have been 750 years or so prior to 2000 BCE, when Yamna branch IE is expanding beyond Urkaine and points East and making its way into Romania, Bulgaria, former Yugoslavia and maybe in Northern Greece and the far Southeastern Poland, and far Eastern Slovakia and Hungary, but given that Catacomb culture (ca. 2800-2200 BCE) is still pretty much within Ukraine, I'd suggest that the expansion of IE beyond Ukraine is biased towards the end of the 2750 BCE-2200 BCE range.

      "Urnfield is not the IE expansion as such but just the expansion of a Western branch (Celts most clearly but probably also Italics and Illyrians)."

      Fair enough. Urnfield expands IE into Western Europe. I would put Urnfield as the point of genesis of Italic and Celtic and Germanic branches of IE. The evidence regarding the Illyrians is too sketchy to make firm conclusions.

      ReplyDelete
    46. "My take would be that Corded Ware (ca. ca. 2900 BCE – 2350 BCE) is really more a successor to the Pit-Comb Ware than to Sredny Stog, and hence would also have been Uralic. Since, Corded ware involved a "new culture and physical type" and "was clearly intrusive, and therefore represents one of the most impressive and revolutionary cultural changes attested by archeology" in its Western expanse."

      I'll correct myself on this point. I still don't think that Corded ware is IE, but looking at it more closely the cultural trajectory seems Ertebølle culture => Funnelbeaker/ Globular Amphora culture => Corded Ware culture, with the last transition starting in Southern Poland but having origins in the East. And, the culture seems to have been intrusive in the Eastern part of its range, not the West.

      The would suggest an origin for Corded Ware not in Yamna and Sredny Stog, but in the Meaglithic peoples to the West. My working hypothesis is that the Megalithic peoples from whom Corded Ware probably evolved most directly probably had a language with a linguistic family relationship to the Vasconic (i.e. Basque) languages.

      I would, however, align the Dnieper-Donest culture with the Comb Cermaic as being probably Uralic, given the strong similarities of the two cultures.

      ReplyDelete
    47. Step by step:

      Pitted and Combed Ware:

      The map in that page is my creation (many years ago, see original map here)... except for that huge purple blob (Combed ware) which was added by some other editor on his/her best judgment, which was not too good probably.

      A big problem is that the big purple blob is not consistent with the chronology of the map, which refers to Middle Neolithic (c. 5000-4000 BCE), when neither Pitted Ware (senso lato) nor Combed Ware had begun their expansion westward. Hope this is clear for the sake of mutual understanding because that Clarifer clearly defenestrated my otherwise quite decent map.

      In my understanding, Combed Ware spread from the Volga Finnish area, while Pitted Ware represents an expansion from Dniepr-Don area via Belarus and the SE Baltic coasts (it has pottery very similar in design to that of Dniepr-Don and also the same type of extended burials with ochre, yet they are not apparently farmers, probably because of climate). Pitted and Combed may be related by contact but they are not the same probably.

      Anyhow, Uralic or not, I agree that Pitted Ware is pre-IE. Combed Ware is almost for sure Uralic.

      ReplyDelete
    48. Seredny-Stog II is not AFAIK related to Cucuteni-Tripolye at all. Cucuteni is a Danubian culture extending in Transylvania, both Moldavias and the SW hilly area of Ukraine. This is out of Seredny-Stog II area, which totally overlaps Dniepr-Don but has Kurgan elements in it, irregularly mixed with Dniepr-Don ones.

      Seredny Stog II is more a transitional complex than a culture properly. Yet it's central to understand early progression of IE groups into Central Europe and the Balkans.

      The classical interpretation is that Seredny-Stog II peoples gradually plundered the late Cucuteni area and that's why there are burned towns. The main Seredny-Stog II expansion vector seems to have been to the East Balcans and even the Maritza (East Hungary) area, where there are several isolated but rich kurgans (with platinum stained gold from the Ural mts. and animal head scepters). Their main incursion base may have been at the Danube Delta and they certainly dominated Wallachia for some time, gradually bringing the rich Balcano-Danubian civilization of Karanovo-Gumelnita (Bulgaria, Wallachia) to an end.

      However the resulting array of cultures is not clearly associated with Seredny-Stog II and each one shows its own affinities. A possible explanation is that the IE raiders were pushed to the mountainous area of Oltenia and Timisoara (but very unsure). The main resulting culture (Ezero, in Bulgaria mostly) shows some affinities with Dniepr-Don instead of Kurgans (extended burials with ochre again) they are sometimes claimed to be ancestral to the historical Thracians (unsure again).

      Whatever the case Seredny-Stog II was replaced c. 3000 BCE by the Catacomb culture, which is generally accepted to be a consolidation of Eastern IE influence in the Pontic area (though may have a Caucasus mixed influence, via Maikop, depending who you read).

      Catacomb culture is also interesting because of its apparent links to Afanasevo (cranial deformation) and later Corded Ware (first corded pottery and "battle axes" are from Catacomb c., it seems).

      ReplyDelete
    49. "My take would be that Corded Ware (ca. ca. 2900 BCE – 2350 BCE) is really more a successor to the Pit-Comb Ware than to Sredny Stog"...

      Can't agree.

      The origin of Corded Ware is complex: clearly its local roots are in the sequence Baalberge > Luboń > Globular Amphorae. Baalberge specially seems related (besides of the Kurgan elite and the Danubian substrate) to the wider Funnelbeaker complex, which in turn is related to Pitted Ware in ways not too clear (and not too linear).

      But to me at least it seems already a Kurgan and hence IE culture. However in the subsequent period all this Baalberge-derived group experienced mild Danubian influence from the South (Baden culture, which seems to have been the "great power" of the area for some 500 years - tribal league? kingdom?) and maybe its own substrate.

      There is some evidence of Catacomb culture presence in the Warsaw area just before Corded Ware appears out of Globular Amphora. Then Corded Ware expanded as we all know: to the Rhine by the West and to the North and NE as well.

      IMO Corded Ware is the consolidation (and one can even speculate as the moment of satemization of the eastern Corded Ware areas) of IE in Central-Northern Europe.

      And we must not forget that other groups, including the Western (East Germany) derivate of the original Baalberge and some other more or less indoeuropeanized Danubian and unclear groups (Deeply Impressed Pottery in Low Germany/Netherlands for instance) existed when Corded Ware expanded over them.

      So the Central-North IE genesis may be complex but all the time relates to the overall Kurgan expansion process, fist with Baalberge and later with Corded Ware. And even earlier, with Pitted Ware, with pre-IE but also Eastern European groups, which seems to have opened the way somehow for the Kurgan flows.

      "Thus, Proto-IE is probably a Sredny Stog creole of Uralic and the language of the LBK".

      Can't agree. PIE is the language spoken at Samara Valley before Kurgan expansion. Its likely affinity with Uralic or at least Finno-Ugric probably stems from coalescing in contact with each other (the Finno-Ugric urheimat was just north of the IE one, at the Volga-Urals region).

      Danubian culture (and hence language) never went beyond the Dniepr river and with all certainty never influenced the Dniepr-Don area, much less the Samara Valley.

      "In this view, IE starts to seep into Central Europe beyond Ukraine only starting with the collapse of Cucuteni-Trypillian culture in the 3rd millenium".

      It's a nonsense. Cucuteni was just another victim of IE expansion and in any case it did not influence them. That's a fantasy that holds no scrutiny.

      I don't understand because just a moment ago you were saying things that made perfect sense (the Tocharian-Afanasevo calibration point) and now you come with this? :(

      "... given that Catacomb culture (ca. 2800-2200 BCE) is still pretty much within Ukraine, I'd suggest that the expansion of IE beyond Ukraine is biased towards the end of the 2750 BCE-2200 BCE range".

      I do not consider Catacomb the beginning but the end of Kurgan expansion into Central Europe/Balcans. As said before, they trigger Corded Ware, without really being their origin probably.

      Careful with the dates: Baldia's dates usually are quite older than the ones I'm used to. And Baldia only deals with Central Europe in his site, so Catacomb dates should be either older for him or Corded Ware more recent for the rest.

      My usual reference is 3000 BCE for Catacomb and 2400 for Corded Ware (c. 2500 for the Catacomb arrival to Poland). So if Corded is corrected to 2900, then Catacomb should be also corrected to 3400 BCE, I guess. Even if not that much, Catacomb should be older than Corded Ware by some centuries, never younger.

      ReplyDelete
    50. "the cultural trajectory seems Ertebølle culture => Funnelbeaker/ Globular Amphora culture => Corded Ware culture, with the last transition starting in Southern Poland but having origins in the East".

      Not exactly what I say but similar enough to make some sense. IMO Funnelbeaker is a complex phenomenon and not a single/simple culture. Its origins are obscure and could be Ertebölle and/or Pitted Ware. But Baalberge is a culture and, while in Funnelbeaker, it is Kurgan (unlike the rest). It is Baalberge which eventually produces Globular Amphora (via Luboń) rather than the diffuse Funnelbeaker phenomenon that includes too many different groups.

      Also the transition to Corded Ware seems to happen at Cuyavia (Warsaw area) rather than in the more Danubian South. But maybe I' wrong in this extreme if some new evidence has been exposed.

      "And, the culture seems to have been intrusive in the Eastern part of its range, not the West".

      I'd say that Corded Ware is intrusive in all its range: in the West, in the East and in the North. It is a marked change even in the previous IE areas to some extent (trigger intrusion of Catacomb group in Cuyavia, altering the burial style - but always in Kurgan fashion).

      ReplyDelete
    51. Maju,

      Nice write-up. I still have a different view on which (additional) areas were highly IE-related initially, but need to do more work to prove my point (which is related to the question of how many significantly different language branches Eastern Europe could have supported ~7,000-8,000 years ago, and the inaccessibility of the region above the two great lakes after the ice age, until about then).

      Couple of quick comments:

      "In Italian it's K/CH etc. For some reason CE/CI were decided not to be KE/KI in Vulgar Latin but something else (yet this something else varies quite a bit)".

      Which makes me believe that in much of the other (than Roman) Italic languages, and the languages originally spoken in the north of Italy and in Raetia (outside of Etruscan), "c" before i and e was more of a fricative (similar to the hard German "ch"). Alemann/Swiss still preserve some of this, and German preserves the distinction of a hard "ch" after a, o, u - and a soft "ch" after e, i.

      Remember that Latin did not have letters that would approximate g, j, or the various "ch-" sounds - but that does not exclude that "c" may have stood for different sounds based on convention (especially, regionally).

      Finally, mutually unintelligible Italian regional dialects spoken today attests to the fact that even during Roman times there were huge local differences, and the Roman form of Latin/Italic never managed to completely impose itself. Incidentally, to me, Spanish (even if derived from a vulgar form) is much closer to Latin than to Italian.

      ReplyDelete
    52. "the inaccessibility of the region above the two great lakes after the ice age, until about then"

      It was not inaccessible at all. We have lots of evidence of people living in East Europe prior and after the LGM (I presume your "two great lakes" are the Black and Caspian Seas).

      There are two hypothesis on this: that there was more or less continuity since at least Gravettian era (Epigravettian documented since LGM) or that there was total "magic" replacement, never really explained properly. So I'm with the first one.

      In any case Dniepr-Don does not seem directly related to Balcano-Danubian Neolithic cultures: it's a different phenomenon, with different pottery style, different burial practices and more oriented to foraging, with agriculture being, specially at first, a complement.

      My hunch is that they spoke languages of the NE Caucasian family. But this is hard to prove.

      In any case the Pontic region was inhabited all the time since Kostenki, even if at times thinly so.

      "Remember that Latin did not have letters that would approximate g, j, or the various "ch-" sounds - but that does not exclude that "c" may have stood for different sounds based on convention (especially, regionally)".

      In Italian c before e/i is ch (oddly enough ch sounds k, as they use the H to neutralize the soft sound of ce/ci by writing these as che/chi). But well, crazy writing conventions.

      Ibero-Romance has ch sound (written tx in Catalan and Basque, where x represents the sh, /ʃ/, sound). Basque has a lengthy variety of such sibilant (fricative/africate) phonemes: X (sh), S, Z, TX (ch), TS and TZ. A diversity I have only found in Serbocroat so far (though it may be common in Slavic languages in general).

      If we consider the CH sound (/tʃ/) to be of Vascoid substrate (how realistic this is?), then the commonality of this sound in Atalntic languages could be explained. Notice that French does not have this sound (French ch sounds sh).

      Basque tends to alter some consonant sounds (n, l, t, d) after i and only there. But a different phonetic rule might have existed in some other presumably Vascoid language such as Ligurian or Iberian, which may explain the "softening" of ce/ci in Romances (and by extension English).

      What you say in this sense about Alemanic is interesting but not too clear to me.

      May it be a Celtic thing?

      "the Roman form of Latin/Italic never managed to completely impose itself".

      Clearly so. It was "pidginized" all the time, except surely for the educated elites.

      "Incidentally, to me, Spanish (even if derived from a vulgar form) is much closer to Latin than to Italian".

      That could be hard to explain. Spanish certainly has lots of loanwords from pre-IE, Germanic and specially Arabic but, apart of that, it's also considered the most innovative Ibero-Romance language. One of such innovations is replacing f by mute h (faba > haba, fab(u)lar > hablar), surely a Basque influence, where f is a foreign sound. In this sense Galician/Leonese, its closest relative, is clearly more conservative, as it lacks both Basque and Arabic influence.

      It also has sounds non-existant in Latin such as KH (written as j).

      Italian keeps for instance the plural in -i/-e (1st/2nd declensions), while all other Western Romances use the -s ending for that (Celtic or Germanic influence?) They can also pronounce the R soft even at the beginning of words (Roma, etc.)

      But all languages innovated to some extent and all began with Vulgar Latin where declensions had been conveniently abolished (replaced by prepositions).

      ReplyDelete
    53. Maju : "as Tocharian is in all phylogenies AFAIK the earliest language to branch out"

      Actually most people generally see the Anatolian languages as clearly the first to leave the ancestral IE homeland, Tocharian being second.
      They have several features to support this point of view.

      ReplyDelete
    54. Fair enough (even if that would be Indo-Hittite, rather than genuine Indo-European, depending of whom you read, right?)

      Anyhow, I suspect that Anatolian languages spread via the Caucasus from the Maikop culture (one of the earliest Kurgan offshoots, if not the earliest one), which in turn may be related to the origins of Catacomb culture (always depending who you read). The vehicle of this expansion would have been the related Kuro-Araxes culture, whose earliest dates are also from 3500 BCE, it seems.

      So again we are with a safe date of c. 4000 BCE for proto-IE being still a unified language in its Samara Valley urheimat, at the steppary confluence of Asia and Europe.

      The differences in glottochronology many be caused by variations of intensity in substrate influence, which should have been heavier for languages spoken in the linguistically rich Caucasus area. Alternatively the c. 200 years difference between Maikop and Afanasevo, as well as the maybe more intense steppary links of proto-Tocharians (via Scythians, Andronovo and such) in comparison with the rather isolated Anatolian group may also account for this.

      ReplyDelete
    55. Maju : Fair enough (even if that would be Indo-Hittite, rather than genuine Indo-European, depending of whom you read, right?)

      Definitely. The Anatolian languages have archaic features that make them anterior to the proto-indo-european language, according to many peoples.

      "I suspect that Anatolian languages spread via the Caucasus"

      Indeed, maybe it did.
      There is also still The possibility that the Anatolian language family reached Anatolia via the south-east Balkans. The first wave from the north of the black sea ends up in Bulgaria (and as such, quite close to Anatolia) around 4,400 BC which would fit well with the archaic features in the Anatolian languages, if we consider proto-IE around 3,500 BC.
      The indo-europeanized populations might have later entered Anatolia around 2,600 BC, which is the date of important destruction and population movements in the west part of Anatolia - according to the J. P. Mallory book I had read - (could this be the reason of the Etruscean migration to Italia? Aren't the dates in this timeframe?).
      2,600 BC would still allow for the quite visible differentiation in the different Anatolian languages, especially if they came to rule different population substrates with different languages (the first tracks of written anatolian languages are about 1,000 yrs latezr, I believe).

      As for the Samara culture * I think the most ancient date go as far as 5,500 BC.

      Also, maybe the process of Indo-europeanization of the north of the black sea had started earlier somehow (in this case pre-proto-IE, that is).

      * http://en.wikipedia.org/wiki/Samara_culture#Samara_culture_sites

      In addition to the name site mentioned above, other sites are Varfolomievka (on the Volga, actually part of the North Caspian culture) and Mykol'ske (on the Dnieper). Varfolomievka is as early as 5500 BC.

      Also interesting in this context :

      "The Samara period is not as well excavated or as well known as the other two. Gimbutas dated it to 5000 BC. The archaeological findings seem related to those of the Dnieper-Donets culture with this noteworthy exception: horses."

      http://en.wikipedia.org/wiki/Dnieper-Donets_culture

      "There are parallels with the contemporaneous Samara culture, and a larger horizon from the lower half of Dnieper to the mid-to-lower Volga has been drawn, particularly by the advocates of the Kurgan hypothesis as expounded by Marija Gimbutas. Dmitry Telegin (ru) assigns them to a broad cultural region that spanned the Vistula in Poland southeast to the Dnieper. Mallory includes this area within the limits of the Proto-Indo-Europeans. The precise role of this culture and its language to the derivation of the Pontic-Caspian cultures such as Sredny Stog and Yamna culture, is open to debate, though the display of recurrent traits points either to long-standing mutual contacts or underlying genetic relations.[2]"

      ReplyDelete
    56. Some parts of this discussion are very interesting and I thank you all for this, guys.

      "The first wave from the north of the black sea ends up in Bulgaria (and as such, quite close to Anatolia) around 4,400 BC".

      Which one? I hope you are not talking of Boian-Maritza, which is a Dabubian cultural flow into a Balcanic (Sesklo-related and Vinca-related) Neolithic area.

      If you mean Kurgan flows, my dates are from almost 1000 years later. Generically speaking about "north of the Black Sea", specially when the PIE area is more like north of Caspian Sea may be very misleading.

      "could this be the reason of the Etruscean migration to Italia? Aren't the dates in this timeframe?"

      No. Etruscan arrival to Italy is almost universally related to the proto-Vilanovan and Vilanovan cultures, whose timeline begins c. 1300 BCE and corresponds best with the destruction of Troy and the Sea Peoples' period.

      "The archaeological findings seem related to those of the Dnieper-Donets culture with this noteworthy exception: horses"...

      Many more things are different, specially burial practices. The core Kurgan culture (Samara, etc.) is characterized not just by the monumental mounds we call kurgans but also by the fact that burials are generally individual (unlike in Dniepr-Don, where collective burial is the main pattern, indicating family/clan identity or something like that), with animal sacrifices, weapons and other goods in the graves (in contrast with Dniepr-Don, that styles few or no burial goods). Also Dniepr-Don shows no signs of hierarchization, while in Samara basin these are common, specially since the Khvalinsk phase.

      There are some affinities (ochre, extended position of burials) but that's about it. The fact is that we do not know by the moment anything about early IEs before the Samara cultural phase. They may have been there for long or may have arrived a few centuries earlier. No deeper excavations have been performed so far AFAIK.

      ReplyDelete
    57. waggg : "The first wave from the north of the black sea ends up in Bulgaria (and as such, quite close to Anatolia) around 4,400 BC".

      maju : Which one? I hope you are not talking of Boian-Maritza, which is a Dabubian cultural flow into a Balcanic (Sesklo-related and Vinca-related) Neolithic area.

      I remember reading several times about a destructive population movement from the north of the black sea around 4,400 BC responsible for the destruction of Karanovo VI (located in Bulgaria) and I think it was even labelled as "the first kurgan wave" at least in one book about the Indo-europeans that I had read a long time ago.

      It was hinted that it could have had something to do with the Anatolian languages, IIRC.

      Does it ring a bell?

      ReplyDelete
    58. Ok, then it's a problem of different dates for a phenomenon (Cernavoda culture, loosely) that I am used to date to c. 1000 years later, in the second half of the 4th millennium.

      This wave IMO could be leading to Thracian (same as Albanian?) and maybe Greek too. But I'm not 100% sure.

      It's a problem with these datings that differ so much from each other, as I mentioned earlier re. Corded Ware, specially when you drink from unrelated sources which may be using totally different chronologies, as would be the case with all of Eastern European, proto-Anatolian and Central Asian chronologies.

      I'm quite sure that the Cernavoda I/II wave is roughly contemporary, or even slightly more recent than Seredny-Stog II, Maikop and Afanasevo. If you move Cernavoda a millennium deeper into the past, then the same should happen with all the other cultures, so Seredny-Stog II and Afanasevo would be from c. 4500 BCE, Samara would be from 6500 BCE etc.

      ReplyDelete
    59. Ok, so in your sources this Karanovo thing would rather be around 3,500 BC, if I understand correctly?

      I had read that 4,400 BC dating in an book at least more than 20 yrs IIRC, so maybe your dates are more accurate.

      ReplyDelete
    60. Karanovo VI or Karanovo-Gumelnita (Karanovo I/II/III/IV/V is used for other earlier Neolithic cultural phases) would be between c. 3500-3000 (Early Neolithic by pan-European chronology) with decadent persistence under the Cernavoda I/II hordes' pressure in the last centuries.

      It was probably a centralized kingdom and the oldest European state, older than dynastic Egypt (c. 3300-3100) and Troy (c. 3000) as well. However it lasted little and maybe it was its success (wealth) what pulled the IE hordes against it.

      Earlier Karanovo phases relate to Danubian (Karanovo IV/V, influenced and mostly indinstinct from Boian-Maritza), to Dimini-Vinĉa (Karanovo III/IV) and Sesklo/Balcanic Neolithic (Karanovo I/II). There was continuity through them anyhow, even if they adapted to various dominant cultural waves, as the very continuity of the naming site should suggest.

      "... so maybe your dates are more accurate".

      Unsure (because I also got them from a book, a very good professional book, I read many years ago - and I did not even bother annotating the source, as I did not anticipate these Internet discussions: I was just learning back then) but they are at least consistent with each other, the sequence of events should be pretty much ok.

      However I'd like to see a good pan-European late Prehistory site with well structured datings, references, maps, etc. With updated data, of course.

      ReplyDelete
    61. "The classical interpretation is that Seredny-Stog II peoples gradually plundered the late Cucuteni area and that's why there are burned towns."

      This is almost certainly not the correct interpretation. The towns were burned every 70-80 years on schedule, deliberately, in a way that wasn't marked with loss of human life. A cause related to poor fire safety in homes has also been largely ruled out.

      Since something similar is seen in early Mesa Verde peoples in Colorado who had similar types houses developed independently, there is probably something like "wall rot" that makes it beneficial to burn down a house of that kind of construction every couple of generations.

      ReplyDelete
    62. "PIE is the language spoken at Samara Valley before Kurgan expansion."

      How in the world can we know that?

      ReplyDelete
    63. "Etruscan arrival to Italy is almost universally related to the proto-Vilanovan and Vilanovan cultures, whose timeline begins c. 1300 BCE and corresponds best with the destruction of Troy and the Sea Peoples' period."

      Etruscan culture is a good example of a culture that would look IE but for the fact that we know otherwise. They were just over the frontier from Italic-Celtic expansion and managed to adapt rather than die (for about thirteen hundred years anyway).

      One can explain their survival later than any other non-IE language from France to the Ukraine and Turkey, in part, by their adoption of key IE technologies and cultural points, while retaining their language and some of their own culture.

      I think that there is a good argument for continuity betwen Etruscan and pre-Etruscan culture either in Tuscany or to the North of there linguistically and in the non-IE related elements of their culture.

      ReplyDelete
    64. I'm sorry but that interpretation I just can't believe it's correct. Why would they burn their own, so hardly elaborated towns, often with lots of wealth (pottery and such) in them and in many cases also with people inside them (this varies but it's common in certain periods, specially in the Chalcolithic one, when the IE raids should have happened).

      There are many hypothesis on this matter but the destruction by war should not be discarded and is, I think, the default one.

      I have no idea about Mesa Verde but I know of no living nor historical culture that destroys its own settlement, much less with wealth and people inside.

      ReplyDelete
    65. Andrew Oh-Willeke said : "I think that there is a good argument for continuity betwen Etruscan and pre-Etruscan culture either in Tuscany or to the North of there"

      Why exactly?
      The Rinaldone culture (Tuscany & Latium) is often interpreted as Indo-european (patriarchal/aristocratic society + animal husbandry & horse bones + warfare & at least one case of sati/suttee at "the tomb of the widow" (about 2,700 BC)).


      The Remedello culture is also often seen as IE.
      And apparently, some decorative elements seem to come from the east (the link about it on wikipedia is dead though) and there is the apparition of a different physical type in the population at this time (brachycephalous newcomers into a largely dolichocephalous population).

      ReplyDelete
    66. Regardless of possible Patriarchy in those Italian cultures, they are too early to be IE and are most likely derived from Cardium Pottery (via La Lagozza) with whatever Danubian influences.

      A confusion here is the matter of ritual axes in burials, which is not an specific IE practice. In fact, I believe the oldest cases are from Breton/West French Megalithism. The practice had a wide distribution in Central Europe and seems quite trans-cultural (religious, symbolic...)

      I am not aware of any particularly marked/disruptive/intrusive cultural flow into Italy in this period but it's likely that, in parallel with Megalithic influences from the West, other influences arrived from the East as well (from the Balkans in the south and from Central Europe in the North).

      I'd rather think of them as loosely "Liguric", pre-IE. Consider that they do pre-date Corded Ware by several centuries, so it's hard to explain any IE/Kurgan influence so early at such a distance (with Lengyel and Baden cultures as massive buffers in the Pannonian basin).

      ReplyDelete
    67. "I'd rather think of them as loosely "Liguric", pre-IE"

      Ok. The dates seemed pretty early to me too but there are peoples that see things differently.
      It's interesting to see how many and so different theories exist with the same data.
      Look at Jean Manco's opinion about this general region and Liguric by going there :

      http://www.buildinghistory.org/distantpast/peoplingeurope.shtml#indoeur

      and doing CTRL + F "The Stelae People"

      ReplyDelete
    68. I know Jean's opinion. She doesn't want to debate it, so I will not. Just to say that I disagree. I highly value her aDNA reference page but not the rest of her site, which is very much one-sided towards certain opinions.

      ReplyDelete

    Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

    Preliminary comment moderation is... ON (sorry, too many trolls).