Thursday, January 31, 2013

Gather the Bones, XVI: Shiftless Spaniards, the Eanna of Uruk, and My Terrible Grasp of Logic, I

Just a jeux d'esprit today.

Some countries are rich, and some countries are poor. For some reason, economic historians who live in rich countries like to blame the people who live in poor countries. For example, although the reasons change with the season, it's always been pretty clear why Latin America's economic development lags behind North America's.

It's because the people down there are, well, you know. I'm sorry, you didn't quite catch that? I'll repeat myself: "you know." No, really. You know. Look, if you really need me to spell this out for you, get me drunk down in the quiet corner of the Arbutus Club.

Just maybe make sure that no waitress you actually like is working that shift.

Okay, enough of being arch. I have a late Landes in my collection that shows the persistence of this kind of thinking in some places. David Landes' mind, anyway. The fact that I picked it up off a pile of garbage on the curb the week after the August-moving-to-Toronto-to-go-to-graduate-school-disillusionment-can-wait-for-later festival here in Kitsilano is probably a more accurate reflection of just how widely the view that Third World poverty is down to Third Worlders being [you couldn't quite catch me mutter this word]ly defective in some way. That's not the point.

The point is that in the great expansion of the Age of Reconnaissance, the peoples of the north Atlantic basin expanded in all directions. The received view is that in North America, they encountered virgin forest, recently miraculously depopulated of a people who gardened and hunted in a [cue the third parenthetical culture war in six short paragraphs over choice of adjective here] relationship with nature. In Africa they encountered proto-states in Mali and Congo that, like the would-be colonisation drive itself, failed to thrive until the Nineteenth Century. In East and South Asia they encountered states in crisis that restored themselves before, again, falling partially or completely to a second wave of colonialism much later.

And in Latin America, they encountered vast and populous empires that fell easily and at once under Spanish hegemony and yet which still failed to thrive, and have continued to fair to thrive up to the present, at least compared with the supposed settler states of North America. My point in my snide parentheses is that we're less astonished by this than we should be because we have been too ready to embrace theories about how societies become wealthy that are psychologically satisfying rather than logically defensible.

There's another way to take this, however. Things have changed a lot in the days since people had to genuinely fear famine, and there's not many economic writers I trust to have a finger-tip's feel of the issues at stake in such an economy. Good thing that I have a copy of an incredibly digressive and (for its time learned) book by one such author to type from!

"Even the Peruvians, the more civilised [of the pre-Columbian states encountered by the conquistadors], though they made use of gold and silver as ornaments, had no coined money of any kind. Their whole commerce was carried on by barter, and there was accordingly scarce any division of labour among them. Those who cultivated the ground were obliged to build their own houses, to make their own household furniture, their own clothes, shoes and instruments of agriculture. The few artificers among them are said to have been all maintained by the sovereign, the nobles, and their priests, and were probably their servants or slaves. All of the ancient arts of Mexico and Peru never furnished on single manufacture to Euorpe. The Spanish armies, though they scarce ever exceeded five hundred men, and frequently did not amount to half that number, found almost every-where great difficulty in procuring subsistence. The famines which they are said to have occasioned almost wherever they went, in countries too which at the same time are represented as very populous and well cultivated, sufficiently demonstrate that the story of this populousness and high cultivation is in a great measure fabulous." (Smith, Wealth of Nations, 1: 226).

Adam Smith goes on to tell us that although Spanish institutions are defective, cheapness of land makes up for much. He quotes the Scottish business traveller Frezier and "Ulloa" on the population of Lima in 1713 and 1740: twenty-five thousand have given way to more than 50. The authorities cited are a little slight, but, Smith concludes, the growth rate of Peru is "scarcely less" than that of the English colonies. For Smith, there is no mystery here. Latin America is growing at about the same rate from about the same level as the English colonies. a fuller account would bear down on the "about" and inquire into whether another century of colonialism in the southern continent counts as a "head start" or the reverse, although the last question would opened up an inquiry into the specifics of how North America might have begun its entry into the Atlantic world a century or more before the Mayflower, and wouldn't that be awkward.

But then what about that Inca Empire? (No Aztecs today: I've quite enough to do in this post without going back to Aztlan.) Empires and colonialism go together. You know when the first person ever sneered about how the natives were shiftless and lazy, and let their women do all the work?

The Uruk Expansion: From the Hacinebi Project at Northwestern.
The outline story is that people began building villages on the Levantine flanks already in the Natufian, during the last Ice Age. From the beginning of the Holocene about 9000 down to 6000 there are increasing signs of agricultural settlements, and from 6000 to 3200BC a movement towards towns. The latter marks a key dividing line, because 3200BC is roughly the date at which proto-cuneiform documents begin to be found in the Eanna Precinct of the abandoned city of Uruk in southern Iraq. My ancient copy of Nissen introduced me to the standard post-processual typology of the archaic state with what still strikes me as an inescapable datum: Uruk has by far the largest surveyed area of any site of its era. Setting speculative stabs at population aside, we can say from the get go that this southern Iraqi city was the biggest around. Arguably, it is the largest and oldest city in the world, and more than a century of excavation has given science a pretty good idea of what Uruk's material culture looks like.

So the map above is pretty well rooted in the Uruk data. Starting tentatively in 3800BC, and to an increasing extent down to 3200BC, the sites marked by dots and black triangles begin showing Uruk's material culture, culminating with entire towns, or enclaves within settlements, built in imitation of Uruk's architecture. "Uruk's political landscape is replicating itself," to adopt foggily theoretical language. 

So what's going on with this "Uruk Expansion?" Take that one to your Google searchbar and see, because I don't trust myself to summarise the very complicated answer that I think we see. This is one for the sociology of academia --or of Google, maybe? The reason that I am being indirect is that the intuitive answer, the one t that has already occurred to you, is robustly resisted in the articles that will top your search list, and, oddly, there is as of this hour no Wikipedia article to summarise the state of the art. 

Okay, I lied. Or, rather, I have my tongue slightly in my cheek as I give my own short summary. In the latter half of the fourth millennium BC, complex but resource-poor societies on the southern alluvium established interaction spheres with resource-rich areas.

My short summary is less than seriously offered because I am parodying a perceived academese effort to avoid my previously implied intuition: we are looking at the archaeologically recovered map of humanity's first colonial empire. Academics do not want to call Uruk an empire, probably because it seems to them an overly bold claim. 

There are good epistemic reasons for being skeptical and cautious. It is just that I have a scientific suggestion: look at the Incas.  The archaeology of the ancient Near East has an excellent comparative example to test the thesis of an Uruk empire against. There is one prehistoric geographically expansive empire that we know existed.

The Incas show that empires can exist without writing, without (much) urbanisation, without money, without a professional military or many other things. We actually have the interesting preliminary result that, as one modern writer points out, it would be very difficult to recover the Inca empire from its architecture. Outside the core area around Cuzco, the well-known and pervasive Inca hegemony had not yet worked an identifiable change in the political landscape of its subjects. The fact that the Uruk expansion did means that its hegemony was more pervasive than Incan. We can construct models in which that hegemony was exercised other than through imperial might, but in light of the comparative data, the "Uruk Empire" is the most parsimonious explanation of the Uruk Expansion.

Tah dah! a scientifically useful contribution. On a blog. On the Internet. Sorry, got to go: the academy is calling me. LOL.  

On a more serious, but still completely not-serious note, I could take this completely the other way and take the Uruk Expansion and bring it to bear on the Incas. This one is lame; we're taking the less known case and using it to illuminate the better known one. Or performing an exercise in circularity. Whatever: the point is that if the Uruk Empire is an Empire, than the Incan Empire could represent a similarily "archaic" state of economic development, empire notwithstanding. The fundamental anomaly, an empire that is not poised for more rapid economic development to modernity than a supposedly virgin forest is no anomaly at all. The specific conditions of Peru in 1450, like the specific conditions of Iraq in 3800BC, call forth an imperial mode of organisation from an economy that might not necessarily all that different from the one prevailing in, say, the Rhineland or along the Ohio.   

Now that's a bold claim. The site-typological, ie site size area evidence is that Iraq/Peru is "more advanced" than Rhineland/Ohio. The towns are bigger. The material culture is more impressive. The "Tech Level" is higher. Rhineland/Ohio better invest in some educational infrastructure, or Iraq/Peru will launch their Alpha Centauri spaceship first!

This is an interesting point, and I think that it goes back to the problem we have in taking the "Uruk Empire" seriously in the first place. We're used to thinking of southern Iraq at a certain point as "Sumeria," a land of competing city states and literature. Especially literature. In part, this is linguistic essentialism, again. People began writing in Sumerian, probably shortly after the proto-cuneiform stage. Therefore, the people who lived in Sumeria were Sumerian, and city states were just the way that the Sumerians did things. No empires for them: that was for the Akkadians of Babylonia and Assyria to the north.

Along the years, people have tried to argue around this, but the most common argument is also the most problematic to me. Sumerian --in cuneiform-- is a very strange language: relentlessly monosyllabic and agglutinative, except when it's not.  People, and by people I mean historical linguists, have been looking at the exceptions and trying to detect a "substratum" for years. That is, evidence that thousands of years before 3200BC, or maybe thousands of years after, a historian in a stuff map drew a great arrow on a map the ancestors of the "black-haired folk" undertook a great folk migration down onto the alluvium and created a land of culturally and linguistically homogenous Sumerian speakers by exterminating all the lesser folk. Or enslaving them, or killing the men and marrying the widows. (Because that would work!)

Just recently, Gordon Whittaker has pointed out that ethnic homogeneity has never existed in all the later history of southern Iraq. He might have pointed to Richard White's summary of the mixed-ethnicity villages of the Middle Ground of the Ohio in the same breath, although he didn't. The point is that the region was unlikely to have been monolithically Sumerian speaking at any point in its  past, and that anomalous words in Sumerian could have been borrowed from other languages.

Then he goes on to posit an Indo-European (of all things) substratum. Sigh. One step forward, two steps back. Here's Whittaker, in an Internet-accessible pdf (what an odd thing for a scholar to do!), and, perhaps because we're talking historical lingustics (get it? talking?) a demolition job that the author felt had to be published in blog comments.

Okay, because I'm a masochist and enjoy having my ignorance exposed: let me propose an alternative. Let's say that Uruk was the capital of an empire that included many population/linguistic groups. Let's say that, at the height of the empire, administrator-priests at the Eanna begin to write administrative documents in a symbolic pseudo-language. Let's say that at some point they begin to enunciate them. Wouldn't they use the most efficient language for enunciating them? The weird thing about Sumerian, compared with, say, Indo-European, is that it is so rigorously agglutinative that the roots don't change. You don't have to have a way of showing inflection change because there is no inflection change. It needs only one sign for one word, whereas  "to be," to take an example that me-no-understand-linguistics can understand, requires many.

Long story short: what if "Sumerian civilisation" emerges from the needs of empire?

*The Eanna of Uruk, original photo from ..somewhere, but hosted at a woo-woo blog.


  1. Maybe to historians back in the 19th century, substratum automatically implied VIKING CRUSH SLAVE - but that's not the general meaning of the term. When people stop speaking one language and start speaking another, the way they speak the new language is influenced by the way they used to speak. Usually, they keep a few words, but also what looks to the original speakers like a funny accent and some odd ways of putting things. Natually, this is often associated with empire-building: Irish English has a prominent Gaelic substratum, not because the Gaels were wiped out, but because they weren't. But it can also work the other way around: Eastern Aramaic seems to have an Akkadian substratum, for example.

    If Uruk was the capital of such an empire, then it would hardly be surprising if other people in the region started adopting its language.
    It would be a good deal more surprising if its administrator-priests decided to adopt a less prestigious language just because it was easier to read in. Granted, having unchanging stems makes it easier to sound out; but that didn't seem to stop the Egyptians.

  2. The vocabulary that Whittaker identifies as adopted into Sumerian from Euphratic shows that Euphratic was spoken by people living in city states on the southern alluvium, and includes placenames, which, yes, I tend to see as being employed in the VIKING CRUSH SLAVE paradigm, for whatever that says about me.

    The chronology and Inca heuristic leave plenty of room for this. The Uruk Empire begins in Hacinebi Late Chacolithic B1 and rapidly develops in B2. Assuming that the half-millennium chronology c. 3800BC-- c. 3300BC survives technical critique the fall of this archaic state predates the first evidence of Sumerian by another half-millennium, and the first use of proto-cuneiform, too. It's more the era of the invention of the cylinder seal than of writing.

    The Incans, meanwhile, used internal ethnic deportation to secure their empire, and their relationship with prestigious precursor states was ambiguous. The traditional account of the fall of Chimu has the last emperor and his smiths deported to Cuzco as a group, in large part because of the prestige of the latter. After a half-millennium of empire, there is no reason to exclude a priori the presence of any constituent population of this hypothetical empire from any region of it. Were the Sumerians originally colonised people of the Tauran Piedmont? Euphratics? Semites? It strikes me, subject to linguistic correction, that no scenario is excluded over so long a time as 500 years.

    For me, the intriguing questions are, first, the economic logic of empire, and, second, the possibility of a recursive relationship between the language chosen to vocalise pseudo-cuneiform and the grammar of that language. Pulleyblank's much-ignored work on Early Chinese proposes a positive feedback cycle between writing and grammar and along the way demolishes one particular identification of language and ethnicity in the most complete way that I can imagine. Why not here? (More on that later.)

  3. It's rather plausible that there is a relationship between the structure of a language and the ease of coming up with a writing system for it. For the initial developers of writing to pick which language they want to try to write specifically on the basis of its structure seems to require implausible levels of foresight. But I could imagine different people in different cities all trying to do more or less the same thing, and the Sumerian speakers among them succeeding faster because of the structure of their language - and hence ending up setting the standards for the newly emerging profession, and adding to their language's prestige while they're at it.

    That said, Sumer is a none-too-large, very flat plain with rivers running through it, within which, given their boat skills, getting from A to B would have been faster than just about anywhere else in the contemporary Middle East. Those are not propitious circumstances for linguistic diversity. Comparing to cases like pre-Columbian Louisiana, one can imagine it with three or four main spoken languages (Sumerian, Semitic, Euphratic?), but we're not talking the Caucasus or PNG here. Contemporary Sumer is basically monolingual, except for rare survivals of Aramaic or spill-overs of Persian.

    As for that Euphratic paper (which, incidentally, never refers to Euphratic as a "substratum"): For most of the supposed loans he gives no reason to reject the possibility that they came from Sumerian into early Indo-European, rather than vice versa. A few of the comparisons, such as "herdsman" and "wide", are impressive enough to suggest he's onto something - but not enough of them to justify the rather sweeping conclusions he draws about Euphratic speakers' lifestyle. The idea that an Indo-European language at this period had feminine gender goes against the general consensus, and doesn't seem to be well-supported by the examples he gives. Of the two Akkadian prepositions he cites, one in fact has South Arabian cognates, so an IE origin is rather far-fetched. But it would take a Sumerologist to give an informed critique.