22 October 2010

Economic Development and the Indo-Europeans

The story told here will . . . allow us to see the great majority of large-scale empires that have arisen throughout world history (and in both the “West” and much of the “East”, other than China) as arising from a shared cultural origin that goes much further back in time. Examples of such empires from ancient times will include the Roman, Athenian, Macedonian, Byzantine, Hittite, Mauryan, Carthaginian, Achaemenid, Sassanid, and Parthian empires; examples from the medieval period will include the Umayyad, Abassid, Sassanid, Carolingian, Danish, Mughal, Hapsburg, and German empires; and examples from more modern times will include the Portuguese, Spanish, Russian, Swedish, Dutch, British, French, Austro-Hungarian, German, Italian, and - most recently - American empires. These empires did not face the same difficulties in sustaining large-scale societies with the rule of law that many developing nations have, and features of their common origin may help to explain why.

From here by Robin Bradley Kar (University of Illinois College of Law).

A standard account of the history of Western Civilization looks at Western Civilization's evolution from Greco-Roman and Jewish culture to become prominent worldwide in a colonial era that starts a thousand years after the fall of the Roman empire.

The article, instead, makes the case for Indo-European cultural inheritances, rather than merely Greco-Roman and Hebrew cultural influences, as a foundation for a much wider collection of empires than those included in the typical "Western Civilization" story. Here is a taste of the argument from the article:

Comparative cultural studies have also revealed a surprising degree of formal and substantive similarities between the cultural descendants of the Proto-Indo-Europeans, not only at the level of language but also at the level of mythology, social structure, religion, ritual, music, and poetics—to name a few. . . .

For present purposes, two similarities will be of particular importance. The first is the tendency of many traditional Indo-European cultures to promote a specific division of society into three—and sometimes four—distinct classes, with one class tasked with war and ruling (the aristocratic class), a second tasked with performing various sacred and religious duties (the priestly, and often the judicial, class), a third tasked with carrying out a broad range of more common vocations such as crafts, trade, animal husbandry, and agriculture (the common class), and—in some cases—a fourth tasked with the so-called ―menial or ―servile tasks (the servant class). The Celtic and Indian branches of the Indo-European family are, for example, some of the most geographically distant and yet they both exhibit this precise four-part division. . . .

The second relevant similarity is the tendency of many traditional Indo-European cultures to rely on their spiritual class to act as the repositories of a complex set of oral traditions. In many traditional Celtic and Indian societies, for example, these persons—who were typically called the ―Druids in Celtic culture and the ―Brahmans in Hindu culture—were initiated into the priestly class by means of a highly rigorous apprenticeship. . . . For example, the Druids of some periods were tasked with memorizing the oral Laws of Fénechus, which formed the basis for the ancient Irish law system; and the Brahmans of certain periods were tasked with memorizing the oral Laws of Manu, which formed the basis for traditional Hindu law. Comparative cultural scholars have found striking resemblances between these oral traditions, and the earliest recorded version of the Laws of Manu is typically dated back to somewhere between 200 BC and 200 AD—though we can be quite sure that earlier traditions preceded this one. When reduced to writing, the Laws of Manu span a remarkable 12 chapters with 2684 provisions. Importantly, the members of the Indo-European spiritual classes, who were responsible for memorizing and transmitting oral traditions like these, were also often responsible for drawing on these oral traditions to advise the ruling classes on a host of spiritual, legal and political matters.

The article then makes the case for the Roman Catholic Church as the intellectual successor of this Indo-European traditional of a priestly class preserving a legal tradition. Here are some key threads of that argument (emphasis in the original):

In the ancient Roman, Greek, and Hittite legal traditions, one finds a parallel treatment of offenses committed by someone who is not considered a legal person – a slave, a child, a cow, or the like. In principle, several outcomes are possible; for example, declaring the offense to be a nonoffense (as is done with some juvenile offenses in the United States); providing restitution; turning the offender over to the person who suffered from the offense; and so on. What is interesting is that all three ancient traditions offer the same resolution – a choice is allowed between restitution and turning the offender over to the plaintiff. Thus, the structures of the legal codes are completely parallel, with not just one outcome being prescribed but a choice between two outcomes and, moreover, the same choices (out of several conceivable ones) being specified in all three traditions. Further, cognate vocabulary is used . . . . The structural parallels combined with the linguistic parallels thus permit us to infer a common origin for the legal practices being compared and even provide us with an idea of the technical language covering situations in Proto-Indo-European. . . .

[T]here is, in fact, considerable controversy over where exactly to locate the Proto Indo-European homeland. Still, there is broad consensus among experts that these people were primitive, nomadic pastoralists who originated somewhere outside of the Indian subcontinent, and who brought Indo-European culture to the Indian subcontinent sometime in or around 1500 BC. If some of the Western descendants of these nomads (such as the ancient Romans and Greeks) were subsequently able to develop rich cultural traditions, which were capable of supporting large scale human civilizations, and could form the basis for large parts of Western Civilization, then we really should pay tribute to their immense bursts of subsequent creativity. If this is what had happened, then it would also make sense to begin the story we tell ourselves about the earliest origins of Western Law and Western Civilization right here. But this is not—in my view—what happened. . . .

[M]y primary interest (at least in this project) is in the origins of Indo-European legal traditions and the cultural conditions that support them. As will become clear over the course of this project, this topic is both intimately related to and importantly separable from more traditional inquiries into the origins of the Indo European languages and peoples. To foreshadow, I will be arguing that certain early social phenomena in the Indus Valley gave rise to the cultural and legal traditions that have helped subsequent Indo-European groups transition to complex societies with the rule of law. Still, this view is logically consistent with the idea that even earlier Proto-Indo-European groups (or their ancestors) may have migrated into the Indus Valley from the West. . . . Still, [in] any such migrations . . . the groups in question would have lacked any traditions relevant to sustaining large-scale civilizations or the rule of law. These more specific Indo-European traditions developed in Indus Valley—on the present view—and it therefore here that we must locate the origins of significant aspects of Western Law and Western Civilization. . . .

[T]he Harappans also spoke a dialect of Proto-Indo-European—or so, at least, I will be arguing, thereby staking out a minority claim in the literature. The development of the early civilizations in the Indus Valley played a major role in the prehistoric coordination and expansion of the Proto-Indo-European language family as well, and, indeed, although some dialects of Proto-Indo-European were probably already spoken in a number of adjacent areas, the socio-cultural developments in the Indus Valley further stabilized these dialects and helped them to spread even further over several millennia. In the process of becoming one of the very first major world civilizations, the Harappans also developed a range of important cultural innovations that were specifically adapted to the maintenance of large scale human civilization.

The second half of the article sums up in a compact way, a fairly comprehensive pre-history of the world with a focus on the history of the major language families.

The argument expounded is basically a version of the Out of India theory of the Indo-European origins, moderated in a way that adopts the prevailing Kurgan theory of Indo-European origins, while positing a secondary layer of cultural evolution and nucleus of expansion for the Indus River Valley area.


The argument that many of the Indo-European societies have common cultural traditions in legal concepts that extend further than the traditional core of Western Civilization, is plausible and has merit.

The argument that these common cultural traditions were developed in the Indus River Valley, rather than received from there has significant barriers which it shares in common with the "Out of India" hypothesis of Indo-European language origins, despite being a distinct and more nuanced version of that hypothesis that contradicts the prevailing Kurgan hypothesis less strongly.

Supporting Evidence

There are facts that recommend the article's thesis that distinctively Indo-European cultural traits have roots in the Indus River Valley:

1. The demise of the Saravasti River system around 1900 BCE provides a powerful motive force for a previously statis oriented society to transition the an aggressive effort to seek a new homeland for its people, and helps explain the transition to the distinctive cultural marker of later Indo-European societies: cremation.

2. It explains why the Tocharians, who have linguistic commonality with other Europeans, but who split from other Indo-Europeans prior to 1900 BCE, have an Indo-European language and cultural ties to Kurgan people, but lack the distinctive crematory practices of other Indo-Europeans, and we not in an aggressive expansionist mode until much later in their history, when they were forced from their homeland in the Tarim Basin by invaders from the East.

3. There is a lack of evidence of a violent pastoral nomadic invasion of the Indus River Valley around 1900 BCE, and the Vedic texts both lack an "origins myth" and seem to refer to conditions that existed in the Indus River Valley far earlier than 1900 BCE.

4. It moots any coincidence in the fact that the first documented smelting of iron took place in Indo-Aryan territory around 1800 BCE and in Hittite Anatolia a couple of hundred years earlier.

5. It explains how some many terms for settled agriculture made their way into as society of nomadic pastoralist Indo-Europeans from the Kurgan culture.

6. Our limited linguistic knowledge of the Harappan language makes it impossible to rule out the possibility that the Harappans spoke an Indo-European language. The observations of Michael Witzel, who wrote in "Substrate Languages in Old Indo-Aryan," EJVS 5,1, Aug. 1999, 1-67, that there is not an apparant Dravidian substrate in the earliest Vedic Sanskrit, which could be consistent with an Indo-European language speaking Harappan society. Among the academic scholars who take this view are Shikaripura Ranganatha Rao.

Contradictory Evidence

There are also problems with approach.

1. The evidence of cremation in the Greek and Balkan areas precedes by a couple hundred years, the evidence of cremation in the Indus River Valley at Cemetery H. The evidence of iron smelting by the Hittites coincides with the appearance of cremation in Greece and the Balkan area, and precedes the appearance of iron in the Indus River Valley by that same couple hundred years amount of time.

The timing of the current finds would suggest that the timing of the dramatic evolution of Indo-European culture and shift to an expansionist mode in the article's hypothesis right, but that the expansion happened in the other direction.

2. The Vinca culture and its successor "house burning culture" complex, was a culture similar in scale and sophistication to that of the Indus River Valley civilization, and is a culture where ongoing contact and ultimate domination of the similarly peaceful Old European Neolithic culture by Kurgan peoples at a time of major cultural transformation is well documented in the archeological record.

Indeed, the similarity of the Vinca culture and the the Indus River Valley culture would help explain how the Indo-Aryans would able to take the reins so quickly in South Asia. They were replaying the conquests of their grandparents using the same methods.

3. The mtDNA haplotype R2 is common in the Indus River Valley in many of the same places where uniparental DNA types associated with Indo-Europeans are found, but is almost absent outside South Asia. If there had been a major secondary expansion of people out of the Indus River Valley around 1900 BCE to the West, we would expect mtDNA type R2 to be a constant companion of mtDNA haplotype R1a, which is normally seen as a strong marker of Indo-European expansions in South Asia, Anatolia and Europe. But, we don't.

This could be reconciled if there was strong population structure in the Indus River Valley, with mtDNA type R1a found in the North, and mtDNA type R2 found in the South at the time of Indo-European secondary expansion from the former Harappan empire. But, this isn't consistent with the archeological indications that the Harappan empire was one of the most unified ancient empires known to man from the very beginning.

The problem isn't limited to mtDNA R2. There is a large inventory of distinctively South Asian uniparental DNA haplotypes that collectively account for a large share of the South Asian population, including the population of the Indus River Valley area, which are almost completely absent from the rest of the world. But, there are uniparental DNA haplotypes found in both Europe and South Asia, and those haplotypes are found in the highest frequencies among high caste individuals who speak Indo-European languages. This is consistent with an Indo-European population invasion, and inconsistent with a major secondary Indo-European expansion out of India.

Alternately, Southern Europeans could have borrowed culturally from fellow Indo-European language speakers in the Indus River Valley without actual population replacement, a scenario that makes more sense with a horse riding plains people whose ruling classes would share a common heritage than it does in the case of people who have no cultural ties.

4. This hypothesis supposes that the Harappans acquired the Indo-European language from nomadic pastoralists, yet there is strong continuity in Indus River Valley civilization from its earliest Neolithic period formative period to the demise of Harappan Civilization.

One must be quite fussy in how one assumes that language transfer took place in this scenario. Adoption of substrate language and culture by a ruling class isn't unprecedented. Indeed, the case of this happening in the Greek part of the Roman Empire is part of the standard Western Civilization narrative. But, the ability of nomadic pastoralists to insert themselves into a ruling class so seamlessly would be notable.

The evidence of discontinuity associated with Kurgan society contact that is found in the Balkans at about the right time, isn't found in Harappan era in South Asia. In particular, there is little evidence of the use of domesticated horses at any point in Harappan society, until aobut 1900 BCE, yet any society impacted enough by Indo-European pastoralists to adopt their language would have made wide use of horses.

5. The climate change that lead to the demise of the Saravasti River in the Harappan area also coincides with major climate change in Southeast Europe and the Near East that probably led to the fall of dynasties in Mesopotamia and Egypt, and was believed to have precipitated the weakening of the Vinca and related civilizations in Southeast Europe that opened them up to Kurgan society domination.

6. The vocabulary of Proto-Indo-European includes core words for plants and animals that are found well to the North of Harappan territory. If there was a long soujourn of independent linguistic and cultural evolution in Harappan territory before a secondary expansion from that area, one would have expected many of these words to be lost in Sanskrit from disuse, and more names of distinctively South Asian plants and animals to have been preserved in some fashion in the Western Indo-European languages.

7. Divisions of society into a learned priestly caste (presumably, the proto-lawyers), a ruling caste, a warrior caste and commoners may be a relatively natural one in Bronze Age societies that can evolve independently. The emergence of distinctive ruling and priestly classes is found among non-Indo-European Egyptians, Sumerians, Minoans, Japanese, Incas, Mayans and Hungarians. Special status for warriors also seems to be present almost as soon as late Neolithic/Bronze Age socities and chiefdomships emerge, if not earlier.

In a scenario where distinctively Indo-European cultural traits are inherited from West to East, rather than in the reverse direction, as the article proposes, this societal structure may even have been borrowed by Hittites and Myceneans from predecessor Peligasian, Minoan, Hattic, Mesopotamian and Egyptian societies with whom they would have had contact.

8. During the fall of the Roman Empire, one of the most distinctive differences between the Indo-European language speaking Germanic invaders and the Indo-European language speaking Romans was the legal culture. The Roman legal culture was a recognizably "rational" legal culture with which modern Western legal systems have a direct evolutionary link. The Germanic approach to dispute resolution involved institutions like trial by ordeal and trial by combat that had been largely abandoned in classical Western cultures at least as far back as the Greek Golden Age. A hypothesis that posits a common Indo-European legal tradition does not comport with this observation.

9. Similarities between Hindu and Celtic legal cultures in the time period cited (200 BCE to 200 CE) do not show the direction of cultural transfer, simply their common origin.

10. The Harappan society, even in its late period, rather than being strongly stratified as it typical in Indo-European societies, as the author notes, was actually one of the most egalitarian large ancient empires. The material remains of that society do not provide strong evidence of the kind of caste system found in Vedic India. There is really no evidence of any kind of warrior class in Harappan society, and there is likewise not the evidence of a klepocratic ruling class found in other ancient cultural complexes of similar technological sophistication. If social class distinction was one of the hallmarks of the Indo-Europeans, then the Harappan society was an unlikely source for it. The priestly class doesn't emerge as important in Vedic society until a period after the early Rig Vedic period. Harappan religion, with its central role for the bull, also resembles the pre-Indo-European Minoan religious tradition. If Indo-Europeans arrived and dominated the society in its late period, we would expect more material evidence of social stratification.

11. There is similarly, little or no evidence of polytheism in even late Harappan society, yet that is one of the distinctive elements of Indo-European religion, the ancient Egyptian religion, and the Sumerian religion. The evidence appears to favor, instead, a Harappan society with a fertility cult similar to that of early Neolithic peoples in Old Europe and Anatolia. If Indo-Europeans arrived and dominated the society in its late period, we would expect temples to polytheistic gods to be in evidence.


The case for Indo-European being a relatively minor language until a secondary expansion around 4000 years ago is solid. But, the weight of the evidence seems to favor an expansion from somewhere in Southeast Europe that reached the Indus River Valley within a century or two more strongly than one out of the Indus River Valley to the West.

Finds of cremation or iron use or widespread horse use in the Indus River Valley a few hundred years earlier than current finds would suggest an opposite conclusion. But, the Indus River Valley has been relatively thoroughly studied in the period since this civilization was discovered in 1921, so the likelihood of new finds here that would fundamentally reverse that chronology seems unlikely. This suggests that pre-Vedic Harappan society probably did not speak an Indo-European language.

Footnote on the Harappan Language

If the Harappans didn't speak an Indo-European language, what was their language like?

The evidence that the Harappans spoke some language in the Dravidan language family prior to the Vedic era also looks increasingly weak. The archeological evidence seem to show only weak trade links between the Dravidians and the Harappans, who had strong trade links with Mesopotamia for much of their history. The crops used by the Harappans are Near Eastern crops that can into use within one or two thousand years of the earliest Neolithic era when they appeared in the Near East, while the crops used by the early Dravidians were Sahel African crops that appeared in South Asia a couple of thousand years later. The claim that there is a Dravidian substrate in the earliest Sanskrit writings has been seriously undermined by recent scholarship. The case for a linguistic and cultural link between the Dravidian languages and the Niger-Congo languages and culture in ways unlikely to be developed independently has been made quite strongly by multiple academic scholars. The age depth of the languages in the Dravidian language family, which remain reasonably similar to each other, also belies an age as old as that of the Harappan languages.

Michael Witzel's argument that Dravidian is not a substrate language in early Sanskrit is considerably stronger than his argument that there is an Austroasiatic substrate in early Sanskrit. Also, even if there is such a substrate, it does not necessarily follow that the substrate is Harappan in origin. South Asian Munda languages are on the far Western expanse of the Austroasiatic languages, which are believed to have expanded from a point of origin in coastal South China starting at a point in time roughly contemporary with that of the transition of the earlier Indus River Valley civilization with which it shows continuity into the actual Harappan civilization. The Harappans simply couldn't have encountered Austroasiatic language speaking peoples for the first several thousand years of the Indus River Valley civilization. And, agricultural traces of East Asian domesticates, as opposed to Near Eastern and local domesticates, do not emerge onto the scene as the Indus River Valley civilization evolved into full fledged Harappan civilization as you would expect if the Austroasiatic culture were to have enough of an influence to produce a language shift.

The case for the Harappan language having an Afro-Asiatic link (a language family that includes Semitic) also seems weak. Harappan language speakers were present in Sumeria before that region transitioned to the Semitic Akkadian language, which arrived in Sumeria at that point from the West. The sea trade routes between Sumeria and the Harappan in the pre-Akkadian era took place via the Persian Gulf and does not appear to have been within the range of Egyptian maritime trade in that era. Egyptian sea trade with India comes much later in its history. Egypt and the Indus River Valley received agricultural crops and domesticated animals with Near Eastern origins at about the same time, in opposite directions. And, if the Semitic languages originated to the West of Sumeria, then one would expect early Neolithic civilizations to the East of Sumeria to have Sumerian affinities, while early Neolithic civilizations to the West of Sumeria to have Semitic or at least Afro-Asiatic affinities. The Old European civilizations of Southeast Europe and the Anatolian civilizations that the Indus River Valley civilization appears to have the most similarities with did not speak Semitic languages, which have no known presence North of the Levant prior to the appearance on the historical scene of the Phonecians, at a point in time when the Harappan civilization had long since collapsed.

The only literary or proto-literary evidence we have of the Harappan language is in its seals, which show similarities with seal systems used by the Sumerians and the Vinca culture of Southeastern Old Europe, and perhaps also the records maintained in Linear A script by the Minoans, all of whom traded with the Sumerian to some extent, although the specific content of those early literary or proto-literary systems have not been deciphered or directly linked to each other.

Thus, in sum, it seems unlikely that the Harappans spoke a language that was Dravidian, Austroasiatic, Indo-European, or Afro-Asiatic. (No one seriously even proposed a relationship of Harappan to Tibeto-Burmese, Austronesian, Hmong, Papuan, Native American, Nilo-Saharan, Niger-Congo or Khoisan languages that is stronger than that of Harappan to some other extant language no in those language families.)

The unity of the Harappan civilization at all points in its history, its disunity with preceding hunter-gather cultures in the area, its long trade ties with Sumeria, the continuity of its civilization, and the origins of its crops all point to a likely origin of Harappan civilization as a monolingual population from the Mesopotamian or Anatolian part of the Fertile Cresent, who managed to endure without meaningful interruption for many thousands of years, until it collapsed for reasons including climate around 4000 years ago, give or take a couple hundred years.

My hypothesis is that the languages of Sumeria, Elam, Harappa, the Minoans, Limnean, the pre-Indo-European Greeks, and the Linear Pottery branch of the Old European Neolithic expansion of agriculture into Europe all spoke languages that were part of the single language family and shared a common cultural heritage. But, some combination of Indo-European languages, Semitic languages, Dravidian languages (in a couple of historic area colonizations in Western South Asia), Turkic languages (in Anatolia if they were spoken there), and Uralic languages (Hungary), in approximately that order of importance, have led to the extinction of every last one of the living languages of this macro-linguistic family, and Sumerian and Elam are the only two languages of this family in which we have any meaningful written texts.

If any living language is a member of that family, one of the language families spoken in the Caucuses would seem the most likely candidates for the closest living language relative of Harappan (although I would be at a loss to say whether NW Caucasian, NE Caucasian or Kartevelian languages were closer). But, I have been convinced to remain agnostic about the relationship of Basque, Etruscan, the languages spoken only in the Caucases, the pre-Indo-European languages of Anatolia and the Northern Taurus Mountains (like Hattic, Hurrian and Kassite), and the pre-Indo-European languages of Southern Mediterrean/Atlantic/Megalithic people in Old Europe.

There is an argument to be made that they were part of the same language family (perhaps splitting off at an earlier time depth) as the Sumerian language family. There is an argument to be made that some of these languages could have Afro-Asiatic connections. There is an argument that these languages formed one or more entirely separate language families. The roots of the languages as pre-Neolithic or post-Neolithic outside of the Fertile Cresent is unclear. But, those connections don't make much of a difference in resolving the question of what languages were spoken by the Harappans.

There is reasonably strong genetic evidence to suggest that the earliest agriculturalists of Western Europe were from a genetically distinct population (most strongly associated with mtDNA haplotype R1b) from that of the LBK agriculturalists of Central and Eastern Europe including Southeastern Europe (where R1a, N1a and other haplotypes are found). There also appears to be strong genetic evidence that they were distinct genetically from a population of Upper Paleolithic modern human hunter-gatherers in Europe whose mtDNA lineages showed a high frequency of U4 and U5 haplotypes - the closest genetic relatives of those Upper Paleolithic modern human hunter-gatherers today are probably the Uralic language speaking Estonians, although the strength of the genetic relationship and their linguistic relationship, if any, is not clear. There is reasonable dispute over whether the genetic matriline ancestors of the R1b populations of Europe arrived in the early Neolithic (probably starting around 6000 BCE), or earlier, in the period of time between the last glacial maximum (around 20000 years ago) and the Neolithic era in Europe, or perhaps even earlier than that. My instinct is to favor an early Neolithic origin.


Maju said...

Nonsense: self-aggrandizing Indo-European empty discourse. Too long to discuss, just to say that...

Neither IVC (Harappa) nor Vinca nor the "house burning culture" (Cucuteni-Tripolye?) have anything to do with IEs nor with each other. Vinca-Dimini was a relative backwater anyhow and the civilized role claimed in this text rather seems to correspond to Karanovo-Gumelnita, essentially of Early Balcanic and Danubian culture.

What languages did they speak? No idea but 99% sure they were not IE languages.

The various IE biological roots are not IE for the most part but pre-IE. Some important pre-IE elements have also penetrated the various modern/historical IE cultures causing their many differences.

andrew said...

I refer to R1a and R2 for example as mtDNA haplotypes, but they are actually Y-DNA haplotypes.