Then the king of Assyria came up throughout all the land, and went up to Samaria, and besieged it three years.
 In the ninth year of Hoshea the king of Assyria took Samaria, and carried Israel away into Assyria, and placed them in Halah and in Habor by the river of Gozan, and in the cities of the Medes.
 Therefore the LORD was very angry with Israel, and removed them out of his sight: there was none left but the tribe of Judah only.
Most Americans are aware of the term “Assyria,” if they are, through the Bible. The above quotation is of some interest because it alludes to the scattering of the ten northern tribes of Israel during their conquest and assimilation into the Neo-Assyrian Empire. Neo because the Assyrian polity, based around a cluster of cities in the upper Tigris valley in northern Mesopotamia, pre-dates what is described in the Hebrew Bible by nearly 1,000 years. During the first half of the first millennium before Christ they were arguably the most antique society with a coherent self-conception still flourishing aside from their Babylonian cousins to the south and the Egyptians (other groups like the Hittites who may have been rivals in antiquity had disappeared in the late Bronze Age). The period of the Neo-Assyrian Empire, in particular under Ashurbanipal, was arguably the apogee of the tradition of statecraft which matured during the long simmer of civilization after the invention of literacy and the end of the Bronze Age. The Neo-Assyrian Empire marked the transition from cuneiform to the alphabet, from chariots to cavalry. Assyria’s political evisceration by its vassals and enemies was inevitable, as a agricultural society on the Malthusian margin can squeeze only so much marginal product out of so many for so long. Once social and cultural capital is gone, there’s a “run on the bank,” so to speak.
But the Assyrians are still with us! Baghdad Raids on Alcohol Sellers Stir Fears:
Eight men carrying handguns and steel pipes raided a Christian nongovernmental organization here on Thursday night, grabbing computers, cellphones and documents, and threatening the people inside, according to members of the group.
“They came in and said, ‘You are criminals. This is not your country. Leave immediately,’ ” said Sharif Aso, a board member of the organization, the Ashurbanipal Cultural Association. “They said, ‘This is an Islamic state.’ ”
The intruders wore civilian clothes, said Mr. Aso and others at the organization, but their arrival was preceded by three police vehicles that blocked off the street. He said the men stole his ring and bashed him on the leg with a pistol.
First, a little etymology. It turns out that the term Syrian likely has a root in Assyria. That term itself deriving from Assur, the primary god and city of ancient Assyria. After the conquest and dismemberment of the Assyrian Empire the core Semitic lands between the Mediterranean and the Zagros mountains became the cultural domain of the Syrian people. That is, those who spoke one of the Syrian dialects. Politically Assyria never arose independently again after its conquest by the Persians. Despite the dialect continuum, and deep roots in the Assyrian Empire and Near Eastern polities preceding it, for nearly one thousand years the eastern and western segments of the Syriac domains were divided by politically, and to some extent culturally, between the Classical Greco-Roman spheres and the Iranian orbit. People of Syrian origin became prominent in Roman life, such as the emperor Elagabalus and the writer Lucian. In the east, under Persian rule, Assyrians such as Mani were also culturally and socially prominent, though marginalized politically by the dominant Zoroastrian Persian ruling caste. The division between east and west was also evident among the Jews in Late Antiquity; ergo, the two Talmuds.
The coming of Islam changed this dynamic: the eastern and western Syrian world were reunited into one political and cultural order. Even though there always existed connections across the Roman-Persian frontier (which in any case periodically shifted), it is notable that the ancient historical divisions persist down to the present day among those who consider themselves the descendants of the (As)Syrians of that era: the Middle Eastern Christians. The Christians of Syria and Lebanon divide between those who are aligned with the Syrian Orthodox Church, or Christians affiliated with Eastern Orthodoxy and Roman Catholicism. In Iraq the majority of Christians come from a different stream, the ancient Church of the East which grew out of the Christian communities of pre-Islamic Iran and Iraq. Today the majority of Iraqi Christians are in communion with the Pope of Rome, while the Assyrian community of the Church of the East is predominantly found abroad (this is due to 20th century politics). But whatever the current configuration, it remains true that to this day these churches can root their lineage back to the Roman and Sassanid period.
And Syriac in the form of neo-Aramaic remains a living language in the Middle East among some Christians. In Syria it is almost extinct, but substantial numbers of Christians in the east still speak it. This is one reason that there is some debate as to whether “Arab Christians” are Arab at all. Ignoring the reality that whole Arab tribes were known to have been Christian even before Islam, it is probably correct to assume that almost all Arab Christians are Arabicized Aramaic or Coptic speakers. In the The Rise of Western Christendom Peter Brown claims that the conversion to Islam by subjects of the Arabs in the Fertile Crescent increased in pace only after the shift from Syriac and Greek to Arabic. In other words, Arab Christians were far more common than Syriac Muslims.
Even though the majority of the population of the core Middle Eastern nation is descended from the peoples of antiquity, they now consider themselves by and large Arab. The Arabs were also present in antiquity, and are mentioned early on as a group on the margins of the ancient world (and sometimes at the center). But it seems implausible that the antique Arabs had the demographic heft to overrun so many peoples across the Fertile Crescent, let along Egypt. Though the Semitic populations of the Middle East now generally have an Arab self-identification in keeping with their dominant language, some among the Christians dissent. For speakers of neo-Aramaic in Iraq this makes total sense; but Arabic speaking Lebanese Maronites also object to an Arab identity (though this gains some traction due to the common bilingualism of Maronites in French and Arabic). But even if most of the Christians of the Arab Middle East are no longer non-Arabs by speech, they preserve a direct link with the ancient pre-Arab Middle East in their liturgy. In the Fertile Crescent this would be a variant of Syriac, but in Egypt it would be Coptic, the language which descends from ancient Egyptian.
There are obviously many in the Middle East who take pride in their pre-Islamic past. Saddam Hussein liked to fashion himself a latter day Nebuchadnezzar II and Hammurabi, while the government of Egypt is a lavish funder of Egyptology. But the Christians seem particularly attached to the pre-Islamic past, because their religion is a tie back to antiquity, and its broad outlines were formed then. This has a bit of an ironic aspect, because in Late Antiquity the Christian Church was a powerful force in the destruction of the indigenous religious traditions of Syria and Egypt. In Syria it seems that a non-Christian culture and society made it down to the Islamic period around the city of Haran, showing up in history as the Sabians. This was probably just a coincidence of geography, as the forced conversion which Justinian the Great imposed on the non-Abrahamic minorities (and to a lesser extent on the Jews and Samaritans as well) in the 6th century was unfeasible so close to the border with the Sassanid Empire. Unfortunately the textual records from Persia are not so good. We don’t know how the Semitic population shifted religious identity from non-Christian to Christian (or Jewish), particularly in an environment where the political elites were not adherents to an Abrahamic religion (though if someone can post a literature reference I’d be very curious).
However it happened, what we do know that is that by the early Islamic centuries the Aramaic speaking populations of the Fertile Crescent were instrumental in being channels for the wisdom of the Classical Age. Many of the Syrians were trilingual, in their own language, as well as Greek and Arabic. For an overview of what transpired between then and now to the Christians of the Middle Eastern Orient, read my review of The Lost History of Christianity. Suffice it to say, by the year 1900 Westerners who were reacquainting themselves with Oriental Christianity observed that they had lost much of its cultural vitality, and been subject to involution. Over a thousand years of Muslim rule and domination meant that the Christians of the Middle East had been ground down into total marginality; to such an extent that Western Orientalists had to “re-discover” them.
This marginality was an end consequence of the dhimmi system to which they’d been subjected to, a system that Christians had imposed upon Jews and Samaritans earlier. They were allowed to persist and exist, but only marginally tolerated. Debilities and indignities were their lot. One famous component of the modus vivendi between Muslim polities and the non-Muslims whom they dominate is that one can defect to Islam, but defection from Islam is not tolerated. The involution of dhimmis then is simply not cultural, it is genetic. By and large the cosmopolitan welter of the great Islamic Empires would have passed the dhimmis by. Eastern Christians then may given us an excellent window into the impact of the Arab conquests on the genomes of the peoples of the Middle East. For example, how much of the Sub-Saharan genetic load in modern Egyptians is post-Roman, and how much pre-Roman? A comparison of Copts to Muslims would establish this. It has clear political implications in the United States, where Afrocentrism is rooted in part on the presupposition that ancient Egypt was a black civilization.
But this post is not about Egypt. Rather, let’s go back to the Assyrians and the Middle East. I wrote up the historical introduction for perspective. But this is about genes. Nature on The rise of the genome bloggers:
David Wesolowski, a 31-year-old Australian who runs the Eurogenes ancestry project (http://bga101.blogspot.com), also focuses on understudied populations. “It’s a response, in a way, to the lack of formal work that’s been done in certain areas, so we’re doing it ourselves,” he says. Wesolowski and a colleague have drilled into the population history of people living in Iran and eastern Turkey who identify as descendants of ancient Assyrians, and who sent their DNA for analysis. Preliminary findings suggest their ancestors may have once mixed with local Jewish populations, and Wesolowski plans to submit these results to a peer-reviewed journal.
A few weeks ago Paul Givargidze, David’s colleague mentioned above, informed me that it didn’t look like the article would be published in the near future due to time constraints. But with all the energy invested Paul wanted something to come out of the project, so he forwarded me a link to a set of files, and suggested that if I found it of interest I could blog about. Here’s the link:
Additionally, Paul informed me that the background of the Assyrian samples were Jacobite (Syrian Orthodox?), Church of the East, and Chaldean. The latter two are the same for our purposes; the the separation of the Chaldean Church from the main body of the Church of the East is a feature of the past 500 years. The Jacobites though presumably are from Syria, though I know that there were some Jacobites in the Assyrian lands as well. In any case, the key is this: these populations have been isolated from others since the rise of Islam 1,400 years ago. They give us an insight into the genomic landscape of the Late Antique Levant and Mesopotamia.
The slide show below has what I believe are the most pertinent figures (I’ve reedited them a bit). The first two are ADMIXTURE plots. So they’re showing you the breakdowns by population/individual for K ancestral quantum (8 and 10) respectively. The rest are MDS which relate individuals within populations on a two-dimensional surface.
[zenphotopress album=248 sort=sort_order number=7]
Some of the populations should be familiar. They’re from the same set as a Jewish genetics paper from last spring. And that’s why you see a diverse set of Jewish groups too. One thing to keep in mind is that the patterns you observe are partly conditional on the inputs. Remember that the “Near Eastern” constrained data sets aren’t simply geographical zooms from the “West Eurasian” set. Rather, the spatial relationships reoriented themselves as the underlying data set from which they emerge are changed.
In regards to the Jews, there are three obvious groups. An Ashkenazi + Sephardi cluster, a Mizrachi cluster, and finally, a, Yemeni cluster. There are also other Jewish groups which don’t fit neatly into this typology. The Jews of India, the Beta Israel of Ethiopia, etc. But since this is focused on Middle Eastern populations, you’re looking at these three groups. The Yemeni cluster is straightforward: this looks like a classic Judaizing population. In other words, the historical records which suggest Jewish convert kings of Yemen were likely accompanied by mass conversions of populations. Or, perhaps there were mass conversions of segments of the population which prompted the conversion of a kingdom to Judaism, or the rise of a Jewish noble to power.
With that taken care of, it’s time to move on the Ashkenazi + Sephardi vs. Mizrachi cluster. We already saw this prefigured last spring: it looks as if Jews under Roman and Persian rule respectively parted ways genetically nearly 2,000 years ago! This is a strange finding, in particular since some of the Sephardi samples are from Syria. But this is a somewhat deceptive division, as much of the Sephardic Jewish community in the Middle East dates to the Ottoman years, as Iberian Jews fled the increasingly intolerant Catholic monarchies of the peninsulas. Though the indigenous Jewish often preserved their own customs (e.g., Romaniotes), by and large they were absorbed by the arriviste Sephardim. The results from the Syrian Jews imply that either these newcomers were very numerous, or, they were very fecund vis-a-vis the native population of Jews. Where the Mediterranean touched it seems that a common Jewish genetic-cultural pool existed. But what about where it did not? For that, one needs to move east, to the land of the Assyrians.
The Mizrachi Jews of the Middle East are a different tradition from the Sephardim. Not only are they different, but these “Oriental” Jews have also been relatively isolated from outside influences. Their closest cultural analogs are probably the Oriental Christians amongst whom they lived before the rise of Islam. I believe that the MDS to the left illustrates exactly what Paul Givargidze and David Wesolowski were suggesting was noteworthy: Assyrian Christians cluster with Mizrachi Jews. It seems as if Iraqi Jews are of equal distance from Assyrians and Iranian Jews. Overall, the three communities, along with Georgian Jews, form a distinct cluster. And and this is the reason I went to great lengths to outline the historical background which set the stage for the world of the Assyrian Christians who came under the rule of Islam in the 7th century.
One plausible explanation of why modern Assyrians are so close to Mizrachi Jews is that the Assyrians and Mizrachi Jews derive from the ancient Semitic populations which have long between resident in the Near East; the Assyrians of antiquity and the Hebrews of antiquity. There is probably some truth to this, but I think it’s a more complicated picture. First, we have plenty of records of Assyrian population movements, enforced from on high. Even if the extent of this was exaggerated, it is likely that this sort of forced transplantation was instrumental in the crystallization of an Aramaic creole which became the lingua franca of the Near East. Other Semitic languages were marginalized, from Akkaddian to Hebrew. But with the linguistic unity likely came a level of fluidity between the fuzzy sets which bounded the communities which we perceive so clearly later in history.
Judaism as we understand it today, or “Orthodox Judaism,” is a product of the religion of the Pharisees, and the tradition which matured with the Babylonian Talmud. The Judaism of the period of the Hebrew kingdoms was no doubt very different, and even that of the earlier Roman period was more variegated than we understand today. For most of history, or the history we record, Jews have lived under relatively brutal religious monopolies in the form of Christianity and Islam. Their community was limited and constrained. But outside of these contexts Jews could be quite different in how they behaved. For example, the two Jewish rebellions under the Romans or the efficiency of the Jewish subordinates who served the Persian Zoroastrians in the Levant after its conquest in the early 7th century. Just as people left Judaism, no doubt others were assimilated into the Jewish community. The Jewish religious texts provide plenty of evidence of this. And even after the Islamic conquest dhimmis were free to convert from one religion to another so long as Islam was not part of the picture.
Azar Gat has convinced me that we moderns to underplay nationalism in antiquity in War in Human Civilization. But just as modern national identities exhibit some fluidity, they no doubt did in antiquity. Jews and other Aramaic speakers in the Fertile Crescent shared a common language. During the Roman period Jews were not distinguished by being a particularly urban community vis-a-vis gentiles. The connection between Mizrachi Jews and Assyrians probably has to do with them coming out of the same broad North Semitic continuum of peoples.
The question I have is if David found any haplotype blocks connecting the Assyrians and Mizrachi Jews, and compared them in relation to the Sephardi + Ashkenazi cluster. If the demographic separation of Assyrians and Mizrachi was very recent, there may not be much of a “Jewish” distinctive signature. On the other hand, if Jews as a whole share lots of identical-by-descent regions of the genome not shared with Assyrians, then it is deeper than I’m positing here. The clustering of the Assyrians with the Mizrachis could be just an artifact because these two groups haven’t been admixed with other non-Semitic groups, as the European Jews have.
Image credit: gdcgraphics, Karin Bar