Revenge of the herders

By Razib Khan | March 31, 2012 3:29 am

Let me make something explicit: I believe that the model outlined in First Farmers is too simple, and that extant patters of linguistic and genetic variation need to accept the likelihood of multiple population reorganizations across vast swaths of Eurasia within the last 10,000 years. The classic case in point are the Turks. Because of their exotic character vis-a-vis the populations which they displaced and assimilated we can peg rather easily their expansion. Between 0 and 1000 AD they began to make themselves felt across a broad expanse of Eurasia from the eastern fringes of Europe to the western fringes of China, and south toward the world of Islam. Between 1000 and 1800 the Turkic peoples took over much of Eurasia for various periods of time (e.g., the Ottomans, Safavids, and Mughals, were Turkic, while the Golden Horde which imposed the Tatar Yoke were mostly Turkic, not Mongol). It is notable to me that Turkic peoples contributed ~10 percent to the genetic ancestry of modern Anatolians. This is a significant achievement, because Anatolia has been a densely populated seat of agricultural civilization for almost the whole history of agriculture! In Central Asia Turks genetically admixed significantly, to the point of preponderance, with the Iranian substrate.

Why does this matter? Because if it hadn’t happened, and it hadn’t happened in the light of history, I doubt we’d believe it! The Turks were obscure tribes in Central Eurasia 2,000 years ago. There was no anticipating that somehow they would overturn what had been the Iranian world of western Inner Asia, and, that they would break through into the civilized societies of the periphery, to the point of taking them over from above, and assimilating them from below. I do not think the Turks are exceptional in this. It must have happened many times in the past. We just need to open our minds to the possibilities.

CATEGORIZED UNDER: Anthroplogy
MORE ABOUT: Turks
  • http://dienekes.blogspot.com Dienekes

    More important than the ability of the Turks to effect linguistic change in Anatolia is the inability of “herders” to effect linguistic change throughout most of the agricultural world at the periphery of the Eurasian steppe. China, the Indian subcontinent, the Iranian plateau, and Europe are all examples of densely populated agricultural regions where numerous waves of “herders” arrived, but ultimately failed to have lasting linguistic change.

    This is even more remarkable, given that the herders of historical times had two great advantages over prehistoric ones: the invention of mounted warfare during the Iron Age, and the use of writing which allowed the formation and administration of large political units.

    We do have examples of large-scale linguistic change in the agricultural regions of Eurasia during the last few thousand years, and its main agents were always farmers, not herders, and certainly not nomads: Romans, Slavs, Germans all spread their languages in Europe; all of them had societies principally based on farming. The same can also be said for the major linguistic changes that took place in East Eurasia, with the spread of the language and culture of the Chinese to the south being the prime example.

  • http://blogs.discovermagazine.com/gnxp Razib Khan

    #1, China, the Indian subcontinent, the Iranian plateau, and Europe are all examples of densely populated agricultural regions where numerous waves of “herders” arrived, but ultimately failed to have lasting linguistic change.

    depending on how you define ‘herder’ i think iran and india are cases where the aryans did effect change. europe is more complicated, but i lean toward the proposition that indo-european populations had an association with animal husbandry, though this may not have been distinctive to them vis-a-vis the locals.

  • http://blogs.discovermagazine.com/gnxp Razib Khan

    More important than the ability of the Turks to effect linguistic change in Anatolia is the inability of “herders” to effect linguistic change throughout most of the agricultural world at the periphery of the Eurasian steppe.

    also, i think this is the null. the changes in anatolia, and in hungary, should be surprising. even the shift in the oasis agricultural societies of transoxiana and the tarim too.

  • http://dienekes.blogspot.com Dienekes

    The arrival of Indo-European languages in Iran and India did not happen in the light of history, and is hotly contested. Personally, I derive the Indo-Iranians from the farmers of the BMAC, some of whom must have begun migrating after ecological change during the 2nd millennium BC took away their farmlands, with some of them shifting to a more mobile lifestyle during that time.

    But, the point is to examine the cases where nomads or herders arrived in the full light of history, so there is no doubt as to their success or failure to exert linguistic change. Iran has received Turkoman tribes, but Persian did not get replaced; India received Turkic and Iranian invaders but Indo-Aryan was not replaced; Europe received Iranian and Turkic invaders, and its languages were not replaced. Even Anatolia, the great success story saw language replacement being effected through a gradual process that was facilitated by a settled centralized state and bureaucracy ruling over an essentially agrarian population, and indeed the non-Turkic languages did not go extinct.

    We simply never encounter a complete replacement of native languages of settled farmer populations by those of pastoralists. If they failed to do so over the last three millennia in most of the agricultural world, whereas farmers _did_ effect substantial linguistic change over the same period, I don’t really see how an argument can be made that they succeeded in effecting not only linguistic change but near complete replacement during prehistory, especially since they lacked both the mounted cavalry and complex political organization of their Iron Age descendants.

  • Onur

    Even Anatolia, the great success story saw language replacement being effected through a gradual process that was facilitated by a settled centralized state and bureaucracy ruling over an essentially agrarian population [emphasis mine]

    I think that is the key. That also explains the replacement of non-Arabic languages by Arabic in most of the territories with a long agrarian past that were conquered by pastoral nomadic Arabs, and also explains the situation in Hungary.

  • http://blogs.discovermagazine.com/gnxp Razib Khan

    Iran has received Turkoman tribes, but Persian did not get replaced; India received Turkic and Iranian invaders but Indo-Aryan was not replaced

    this is misleading. about 1/3 of the citizens of iran speak turkic languages. much of northern iran has patches of turkic speech. azeri did replace the iranian dialect of the region in the light of history.

    Europe received Iranian and Turkic invaders, and its languages were not replaced.

    hungary. not iranian or turkic, but the magyars were their cultural heirs, and even later gave refuge to turks fleeing the mongols.

    . That also explains the replacement of non-Arabic languages by Arabic in most of the territories with a long agrarian past that were conquered by pastoral nomadic Arabs, and also explains the situation in Hungary.

    you sound stupid if you make analogy between hungary the ottoman state, and that of the magyar tribes. the kingdom of st. stephen was NOT centralized or bureaucratic. additionally, the spread of arabic in many regions (e.g., maghgreb) did not follow centralized states and bureaucracy, but cultural prestige as well as secondarily the migrations of the banu hilal seems to have been operative. in the levant and iraq arabic spread into regions of related languages, and where some arabs already lived.

  • Onur

    I used the word “centralized” in a looser sense. My point is that in all of the exemplified cases nomadic conquerors adopted a sedentary and more centralized government system, unlike traditional nomadic societies. Also, language changes were mostly gradual, relatively slow and far from perfect (as noted by Dienekes) in those examples.

  • http://dienekes.blogspot.com Dienekes

    this is misleading. about 1/3 of the citizens of iran speak turkic languages. much of northern iran has patches of turkic speech. azeri did replace the iranian dialect of the region in the light of history.

    By “replace”, I mean exactly that, “replace”, not co-exist. Of course Turkic languages have been added to the pre-existing languages of north Iran, but they co-exist with persian and kurdish. The realm of the Persians was overrun by Turkic peoples but the languages of the farmers proved resilient to replacement.

    If prehistoric herders from the Eurasian steppe replaced the languages of Europe, Anatolia, Iran, and India with their own, it is remarkable that they managed to do so almost completely, when their historical-era counterparts failed to do so. One could name a dozen historically-attested invasions of Europe from the steppe, and all that’s left of them is the conversion of Pannonia and scattered Turks and Tatars over Eastern Europe. So, I don’t really see how that which was not achieved by the numerous nomads of history could have been near-completely achieved by the pastoralists of prehistory.

    Even the case of Hungary argues against replacement, since the Magyars did linguistically convert a small slice of Europe -perhaps because of its terrain-, but the Iranic nuclei within the Croats, the Bulgar nuclei within the Bulgarians, the Avar nuclei within the Slavs, all of them did not manage to replace the languages of the farmer multitude.

    So, while we have evidence for steppe-derived peoples winning out in competition against the farmers in the language department, we have even more evidence for the opposite as well. And, we have the most evidence for farmers displacing the languages of other farmers or of hunter-gatherers.

  • Justtoinformyou
  • Wim Van Dijk

    Dienekes: “I derive the Indo-Iranians from the farmers of the BMAC, some of whom must have begun migrating after ecological change during the 2nd millennium BC took away their farmlands, with some of them shifting to a more mobile lifestyle during that time”

    — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

    But the early indo-iranian languages (Rgvedic Sankrit and Avestan) lack both agricultural and urban vocabulary. OTOH, pastoralism and horse vocabulary seem particularly important (and the content of the Rgveda and Avesta also rather point to a pastoralist society, originally).

    Besides, linguists see obvious (and numerous) proto-indo-iranian and indo-aryan stems in finno-ugric, so the BMAC doesn’t seem to fit at all as well as the andronovo cultures (and its likely offshoot, the gandhara grave culture) – that would fit not only geographically but also culturally.

    The scythians were indo-iranian speakers and they are clearly direct descendants of the andronovo culture, not of the BMAC.
    The BMAC had some input from the andronovo culture, not the reverse, if I remember correctly.

  • http://dienekes.blogspot.com Dienekes

    @10.

    The Scythians were not “Indo-Iranian speakers”, they were Iranic speakers who were first attested in the first half of the first millennium BC.

    I see absolutely no reason that the people of the “Andronovo culture” spoke the same language as the historical Scythians. After all, there was a shift from Iranic to Turkic languages across the Eurasian steppe in less than a 1,000 years during the historical period, so I see no reason to think that the historical Scythians spoke the same languages in unbroken continuity during the prehistoric period.

    In fact, population replacement on the steppe is all but certain, since the early steppe Caucasoids were uniformly R1a according to published ancient DNA results, but there is a plethora of Caucasoid haplogroups in extant steppe populations. It is certain that different groups of Caucasoids “went nomad”.

    Indeed, there appears to be evidence for the existence of admixture in the steppe, as there are two types of Caucasoid attested there, an early robust broad-faced Proto-Europeoid type, and a later gracile narrow-faced Mediterranean one.

    http://journals2.scholarsportal.info/details.xqy?uri=/15630110/v39i0001/147_ancsotasfta.xml

    “Nonmetric traits were studied on more than 80 crania from the Middle Bronze Age (Andronovo) burials in the Altai. Two alternative trait batteries were used: that introduced by T. Hanihara and H. Ishida, and that proposed by A.G. Kozintsev. Groups were compared using two multivariate methods, Correspondence Analysis and C.A.B. Smith’s Mean Measure of Divergence. Both trait batteries yielded similar results, attesting to the Southern Caucasoid (Mediterranean) affinities of the Andronovo population of the Altai. This agrees with the results of dental studies, but disagrees with the results of craniometric analysis. One of the possible explanations is admixture.”

    I am also not sure what the problem is with the attested contacts between Indo-Iranian and Finno-Ugric; it may very well be the case that pre-Indo-European steppe populations spoke some variety of Finno-Ugric; and, in any case, there was early contact between early Indo-Iranians and Finno-Ugrians in the fringes of the steppe.

    As for the Indo-Iranians, they do share common agricultural terminology; of course they must have become much less focused on agriculture after desiccation set in in their BMAC homeland, and their population shifted their subsistence strategy. It should be remembered that the BMAC population was huge, and these people didn’t just wait to die as their homeland became desert, they must have flowed in all directions outward: onto the steppe, the Iranian plateau, and South Asia being their eventual targets.

  • Wim Van Dijk

    “The Scythians were not “Indo-Iranian speakers”, they were Iranic speakers who were first attested in the first half of the first millennium BC.
    I see absolutely no reason that the people of the “Andronovo culture” spoke the same language as the historical Scythians.”

    Because there is a continuity between the andronovo cultures and the Scythians (archaeologically-speaking, so even if there were some admixture why would have the people that didn’t really affect the cultural matter would be seen as the major source of the language? It seems contradictory.
    Besides couldn’t these “mediterrean influence” in the gene pool coming from the Caucasus? Some of the results of Andronovo mtDNA have modern matches in the Caucasus (like the mtDNA U1a found several time in Lalueza-fox et al 2004, hg also found in the Altai today)) after all, andronovo is seen as spawning from more ancient cultures from the west, not the south.

    “it may very well be the case that pre-Indo-European steppe populations spoke some variety of Finno-Ugric”

    I can’t see any reason to think it. There is a real dichotomy between the hunter-gatherers of the Urals and the andronovo pastoralists of central Asia (no link is seen in archaeology AFAIK).

    And these warlike finno-ugric Andronovo pastoralists (culturally-derived from the Russian pontic cultures – do you claim that these cultures were also finno-ugric?) would have lost their own language because some agriculturalists would have decided to switch their ways and invite themselves on the Andronovo societies’ territories? These agriculturalists would have given their words for cow and milk (these are among the loanwords in F-U) to the fully pastoralist andronovo populations? That’s kind of weird.

    The simplest and most logical explanation is that Proto-indo-iranian speakers (then Indic, then Iranic) were living just south of the Finno-ugric-speaking population – that are universally understood as living then in the Urals – giving for instance some typical Indo-aryan cattle-related words to some Ugric populations as the latter came in contact with these new ways.

    Besides there are also some ancient indo-european (non Indo-iranian) influence in Finno-ugric (it’s mentionned in David Anthony’s book, Page. 93-97, he gives several examples).
    All this points to a close presence of the Indo-european world to the early Uralic world.

    I think stems in common in such different languages as Finnish and Tocharian – an allegedly very early IE language – (example: Tocharian kälk-, kalák- / Finnish kulkea, all meaning “to go”) are better explained with a proximity of both Uralic and Proto-IE urheimats.

    “As for the Indo-Iranians, they do share common agricultural terminology; of course they must have become much less focused on agriculture after desiccation set in in their BMAC homeland”

    The agricultural and urban vocabulary is basically absent (a few sparse agricultural terms exist (IIRC E. Kuz’mina told some andronovo cultures used some agriculture but nothing sophisticated, a very secondary thing), and are words such as “sickle”, or “to plough” – that happens to also mean “to dig” – really that convincing anyway?).
    They wouldn’t have lost so quickly all their relevant vocabulary as they moved southwards. Neither BMAC nor their neighbours of IVC should have been so devoid of this vocabulary, and so it’s difficult to understand why Indo-aryan should be so devoid of it.

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at http://www.razib.com

ADVERTISEMENT

See More

ADVERTISEMENT

RSS Razib’s Pinboard

Edifying books

Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »