Tuesday, March 26, 2013

Gather the Bones, 16, 3: Cold Iron

"I climb to the ridge of the Pei-mang Hills
And look down on the city of Lo-yang.
In Lo-yang how still it is!
Palaces and houses all burnt to ashes.
Walls and fences all broken and gaping,
Thorns and brambles shooting up to the sky.
I do not see the old men;
I only see the new young men.
I turn aside, for the straight road is lost;
The fields are overgrown and will never be ploughed
I have been away such a long time
That I do not know which path is which.
How sad and ugly the empty moors are!
A thousand miles without the smoke of a chimney.
I think of our life together all those years;
My heart is tied with sorrow and I cannot speak."

-Cao Zhi [Ch'ao Shih], transl. A. Wales, web-hosted by Gwilym Williams

Outer fortifications of Tongwangcheng, from Borbala Obrusanzky's "Tongwancheng," hosted at www.transoxiana.org
How a Google search for "Ruins of Louyang" leads me to a university poet in residence and pictures of a "southern Hun city" taken by a member of the "Research Group on Hungarian Ancestral History" I leave to the high priests of the search algorithm. I would speculate, but there's enough of that around here already. 

Specifically, I want to explore the idea of iron (and the Iron Age) as an exogenous technological economic input, and take some steps to draw together what seems to me to be related developments.

Louyang lies on the north bank of the Lou river, which rises in the mountains of eastern Shaanxi Province and flows east to join the Yellow River just south of Louyang in Henan Province. Wikipedia, in one of those not-yet-called-out "weasel word" phrasings, says that it is "said" to be the geographical centre of China, and, as such, ritually important from Neolithic times. Again, it is "said" to have been taken as capital, under the name of Zhenxun by Tai Kang, third king of Xia. I use the scare quotes because Chinese archaeologists rest claims for the historicity of the Xia Dynasty on the Erlitou and Erligang Cultures. Both are named for very impressive ruins that are near to Louyang, but even  40 kilometers can be a pretty significant distance when you have to walk it. More to the point, actual Louyang was taken as capital by the Eastern Han, who might have had reason to see it talked up. 

Setting ritual geography aside, I will point to economic/strategic: towns on the Lou river of Henan intercept the roads that bypass the Yellow River rapids at Hukou and provide access to Shaanxi Province, including the fertile province of the Wei River, which contains Xi'an, capital of the Western Zhou, Qin, Western Han and Western Tang dynasties.  I will wave wildly in the direction of more knowledgeable people than I in defence of the proposition that this alternation of Eastern/Western; Xi'an/Louyang is an alternation of the relationship between the Central Plains of China and the "near steppe" of the Ordos Basin. I will not wave at them in first summarising an argument wherein China is constantly seeking Inner Eurasian nomad cavalry and this desire draws essential Chinese civilisation into a close and corrupting relationship with pastoral barbarism. The question is simply one of seasonal variation in the price of fodder. If it goes high enough, as it does in the Central Plains, in India, in Egypt, on the Mediterranean littoral, buyers import mature horses instead of supporting brood mares and foals for several years. Scholarship gets this. (1, 2, 3). 

In fact, the weird collection of references here that I've been trotting out for as much as 20 years here illustrates the problem. The "zeroth" volume of the Cambridge History of China is a freaking Cambridge History. It's the scholarly consensus, the textbook of choice, and yet its conclusions are given a quick pat on the head and sent to stand in the corner. Of Egypt I would not speak except that I was reading just this week about how its "choice" to go with Mamluk mercenaries instead of feudal service must have cultural roots. 

Because it doesn't. It's the fucking summer rains, guys. That's all it is.

Speaking of the controversial chapters of The Cambridge History of Ancient China, remember this discussion? The claim tried out here is that Sumerian, at least in the form that it comes down to us, was shaped by the prior existence of texts, the so-called "proto-cuneiform." Grammatical features and vocabulary was picked out of the working languages of the southern alluvium that would support the easiest reading of texts that were not, themselves, originally intended to represent a spoken language at all. It turns out that actual experts have been saying this, or something like it, for a long time, and the argument in some sense goes back to Halevy, that is, to the losing side.

That said, I am working, very tenuously, from analogy here. And the analogy here is pretty clear. In a series of papers and monographs culminating with 1995's Outline of Classic Chinese Grammar, Edwin Pulleyblank proved that Old Chinese, the form of the language used from the first oracle bone inscriptions to the Han Dynasty, had Indo-European roots. 

Now, I use "proved" here in the technical sense of "I can't find people who sound like they know what they're talking about who disagree with Pulleyblank." The argument is even reproduced in the Cambridge history. That being said, I notice a very distinct absence of Pulleyblank in the Wikipedia article on the subject. This is, to  put it gently, frustrating. If Pulleyblank is a crank who deserves to be shunned,* someone should tell us poor historians.

But, for now, take Pulleyblank seriously. Now I want to visit the fabled Ruins of Yin, at Anyang, another prefecture of Henan Province. In contrast to the doubts that hang over the identification of Louyang with Zhenxun, we know perfectly well that Anyang was Yin, because the name of Wu Ding, twenty-second King of Shang, according to the Grand Historian, is up there with Ashurbanipal in terms of his devotion to the "heavenly writings," the oracle, omen and teratological texts that scholars and kings at the dawn of the Iron Age were gathering together in the hopes of finding an empirical science of prophecy. Following ancient local tradition, Wu Ding's scapulimancers and plastromancers made thousands of prophecies. Following not-nearly-so-ancient tradition, they then inscribed those predictions on the burnt ox shoulders and tortoise shells with which they made these predictions. The language of these inscriptions is Old Chinese, and unless you believe some extremely dubious claims, they are the first, and remarkably copious texts, in Old Chinese.

So between 1200 and about 250, the texts, while remaining readable, lost the "semantic components" that Pulleyblank thinks represent the inflection that forms an Indo-European grammar, and gained a new one. The texts continued to be read, but the original language behind it was lost, and a new one substituted (or invented.) I think that's uncontroversial, but I could be wrong, on account of not understanding one word in three of what the experts are saying. I know that I never fail to point that out, but I also never fail to find it ironically amusing, so there you go. 

Now, if Pulleyblank is right, the language is Indo-European, and it is not an unreasonable inference that Indo-European scribes are writing it. I say that it is not unreasonable because excavations at Anyang (about which you can read from the multiple copies of Chang on Shang Civilisation at any half-decent library, but the copyright of which Google Books still protects) have turned up a multitude of chariot burials. Chariots are an odd enough technology that they point in the direction of foreign experts coming from the west in themselves. The idea that the Shang might also have used western experts to write out records and prophecies is. . . .

Well, let's face it. It's hugely controversial, for mainly nationalist reasons. The more completely autochthonous Chinese civilisation was in the deep past, the more awesome it is today! Stands to reason. If I ever have to argue this with an eager young nationalist, I will point out that I am arguing for the same mechanism spreading "Indo-European" west in to Europe, and given that the language family took in Europe but not China, this is in itself evidence of the strength of indigenous Chinese traditions. 

I doubt that the argument will wash. But I will stick to my guns on this at least: technology is spreading with horses. Horses are the things that the Central Plains cannot produce for themselves. They must look westward, towards the Ordos Basin and the Kangxi Corridor and beyond, probably towards the Dzungarian Gate and southern Siberia. 

Why is the geography important? Do much research on the subject of the early Silk Road and you're going to be hit over the head with the claim that some Caucasoid mummies prove that an Indo-European speaking population, possibly ancestral to the presumptively indigenous Uighurs, lived in the Xinjiang-Uighur Autonomous Area in the Early Bronze Age. I won't bother to reproduce the evidence, which to my mind is missing more than a few links, just point out that the mummies are from the eastern Tarim, and that if this Bronze Age Silk Road went where water and grazing was most easilyl found, it crossed the mountains at Urumchi, circled the edges of the Dzungarian Basin, and entered Kazakhstan via the Gate. 

The paucity of early archaeological finds in the western Tarim Basin is then explained by the late spread of underwater irrigation canals, the qanat or foggia. So if the timing of the earliest qanats in Inner Eurasia and of foggia in the Sahara is Iron Age, and iron represents a tool material that would have made it possible to cut them, why not argue that they're related?

I have already talked about the axe, forest clearance, and arboriculture in Europe, and tentatively plumped for the Early Riders argument, that whenever horses were first domesticated, the appearance of state-sponsored horse cavalry followed the Iron Age because iron tools are necessary to the largescale use of horses. 

Now I am proposing that the desert trade routes that linked centres of qanat agriculture were an epiphenomena of the iron age. But it doesn't stop there, because there is archaeological evidence that the high steppe was reorganised with widespread horseriding, that the "immemorial pastoral lifestyle" of Inner Eurasia was, in fact, an economic revolution that began about 800AD and terminated about 400. I am gesturing here, because I am just about out of writing time, to di Cosmo's book, already linked to but not cited by name above. And, finally, to the idea that Indo-European is a clue to a social transformation that began with bronze and ended with iron. 

Even in places where the evidence for it is either lost, or hallucinated by a crank. (But a crank who taught at my alma mater, and that's got to count for something.) 

*And in a classic version of argument from opposite-of-authority, Mallory and Mair agree with the Pulleyblank thesis.  


  1. Some speculation about Pulleyblank - http://redd.it/1b3r8a

  2. Pulleyblank is certainly no crank - very much to the contrary - but I don't think he's saying what you think he says. I don't have convenient access to his papers on this, but his 1996 paper "Early Contacts between Indo-Europeans and Chinese" apparently discusses, as the title suggests, ancient Indo-European loanwords into Chinese, many of them technological (http://books.google.fr/books?id=5FzANyya1BEC&lpg=PA218&ots=s17AKLy81p&dq=%22Chinese%20and%20Indo-Europeans%22%20%22%22Early%20Contacts%20between%20Indo-Europeans%20and%20Chinese%22&pg=PA96#v=onepage&q=pulleyblank&f=false). That would be the period of interest for this post, and that would amount to a very different claim than "Old Chinese... had Indo-European roots", unless by "roots" you simply mean "some words".

    As for the 1960s paper, UBC's summary suggests that they tentatively advanced the idea of some much earlier connection between Indo-European, Northwest Caucasian, and Sino-Tibetan (http://www.asia.ubc.ca/people1/edwin-pulleyblank/). But - if that's valid at all, and the evidence sounds pretty slim - that would have to have happened long before Chinese civilisation was a glimmer in anyone's eye, and certainly millennia before they started writing. This would be a claim that Old Chinese and Indo-European had a very remote common origin, which again is not what you're saying above.

    But maybe you're alluding to a different paper that I've missed?

  3. Claims of Indo-European loan words in early Chinese are long-established, can be motivated by the whole chariot thing, and, arguably, by the claim that Shang and Zhou "ancestor worship" aligns with emergent Near Eastern ritual practice that we can see in retrospective as the symposium. That is, hanging out in mortuary settings giving cult to dead ancestors with ritual vessels, including tripod wine servers.

    The paper reprinted in the Cambridge volume, "Number Zero of the Cambridge History of China" goes much further, and seems pretty clear to me --with the usual Ralph Wigginish "What's a morphology?" caveat.

    William G. Boltz, The author summarising, or gesturing at, Pulleyblank in support of wandering further into the weeds, thinks that you can recover Indo-European morphology in Early Chinese. You can see this indirectly referenced in the Wikipedia article, which notes the controversial claim that you can extract words ending with "s."

    In Botlz's chapter, IIRC, and that's a huge caveat as it turns out, these are claimed to be read back as, uhm, conjugations, I guess, so showing that the early script embodies an inflected grammar. That doesn't demonstrate an Indo-European association in its own right, but the word final "s" does.

    I, personally, wouldn't go so far as to say that "Early Chinese" was Indo-European. What I would say is that the script in which the Anyang oracles were written was inspired by an Indo-European script.

    But it was a script, not a spoken language. It was a script, and my skepticism is inspired in the first place by sitting in multilingual lunchrooms where grammar (morphology?) is lost long before signifying vocabulary.

    What's fascinating is the way in which, later, a new tonal morphology was grafted onto the old in order to read, and thus pronounce, the high prestige texts.

    Here, again, is a pdf purporting to show that tones in Chinese consistently replace the same word ending consonants, that, in other words, tone replaces inflections that can be identified as ones found in Indo-European languages.

    On the other hand, I feel myself being led much further into the fever swamps than the lifeline to Pulleyblank allowed me to get . . . .


  4. Which IE script would that be, given the time and place?

    A causative in s is found in Indo-European and Old Chinese, yes - also in Afro-Asiatic, Niger-Congo, Kartvelian, Haida, Tlingit, and Burushaski. S-causatives are just too common worldwide to count as helpful evidence for contact or common origin, except maybe at some insanely early period. On top of that, reflexes of it also seem to be shared with the rest of Sino-Tibetan. (http://crlao.ehess.fr/docannexe.php?id=967)

    The conversion of consonants (or consonant properties) to tone is not some strange property of Chinese - it's pretty much a universal process, one of the only ways known for tone to emerge. It even happened in Punjabi, and it's very conspicuous indeed in Tibetan. It can lead to affixal inflections turning into mere tone differences, but that's kind of incidental.

  5. the scenario I'm imagining here is more along the lines of independent invention on the basis of influence. The reinventin of writing, usually in the form of syllabaries, is widely attested.

    To take a de minimis approach to language ranges, the Andronovo archaeological horizon impinges on both the Bactrian-Margian Archaeological Complex and the Afanasevo, which extends (logically enough) into Dzungaria.

    Andronovo appears freely in theorising about early Indo-Iranian in Mallory and Mair. I have little regard for the two of them as scholars: the crucial point is that the range is right, and if Burkert's prophets and divine singers can spread Middle Eastern cultural influences to Greece, I see no reason why they can't do it eastward bound. That distance would attenuate the influence is a hypothesis that seems baked into the evidence at hand.

    Actual scholarship I just found: http://www.csen.org/BAR%20Book/BAR.%20Part%2001.TofC.html