Wednesday, 29 August 2012

South America: A Post

Travelling on the bus/train to and from work, I tend to spend my time reading.  This is probably the most productive portion of my reading day; I have forty minutes or so each way in which to make progress on whatever book I happen to be reading, and I tend to read about fifty or sixty pages each trip depending on the book.  I want to read at other times, sat on the sofa or maybe even in the garden, but I don't maximise usage of time as well as when I'm on the way to work.  Paradoxically, this means that I read more when I'm working than when I'm not.

Anyway, I've recently been reading about pre-Hispanic South America.  While I am far from an expert on it, and speak none of the local languages very well, I have in my collection a number of excellent books on the continent, including great overviews of the archaeology of the region - Ancient South America by Karen Olsen Bruhns, The Incas and their Ancestors by Michael Moseley, The Ancient Civilizations of Peru by J. Alden Mason (very old and out-of-date, but still quite good) - as well as some specific studies of particular regions or aspects of life - Of Summits and Sacrifice by Thomas Besom, Signs of the Inka Khipu by Gary Urton, Lizot's Tales of the Yanomami.

I've also got quite a few scattered journal articles here and there, including several by Urton on Inka astronomy and astrology.  And, of course, several of Levi-Strauss's articles primarily concern South American topics.  So I've read relatively widely on the place, and that level of reading would, elsewhere, allow me a general understanding sufficient to satisfy my curiosity for a while.

But here's the thing: South America is confusing as hell.  There are at least a dozen known language families from the continent, and they are widely and very bewilderingly scattered.  All classifications of South American languages allow for unclassified and unclassifiable languages and language families, adding to the confusion (The Atlas of Ancient America [Coe, Snow, & Benson 1986] has both 'other language area' and 'unclassified or unknown language area', along with a list of a dozen known families, some of which are controversial).  Those families that we do know about are spread bizarrely across the region.  Arawakan, spoken in the Greater Antilles when Columbus arrived, could also be found as far south as the Andes (including part of the Inka empire, Tawantinsuyu, called Antisuyu) as well as in Venezuela and the Mato Grosso in Brazil.  Cariban, which gave the name to the Caribbean, shows a similar diversity in distribution, from the Lesser Antilles (where it overlapped with Arawakan as a more recent invader from mainland South America) through Venezuela to Brazil.

Tupi-Guaranian, Macro-Ge - these language families are spread across the continent east of the Andes in confusing ways throughout the swamps, forests, and highlands, and they correlate with the archaeological data in equally confusing ways.  (Guarani, by the way, is one of the national languages of Paraguay, spoken there alongside Spanish and found not only among the native communities but also among European-Paraguayans.  I think that's pretty cool.)  Other language families are peculiarly confined to particular segments of forest or swamp.

A language map of Afro-Eurasia shows the dominance of only a few language families (Indo-European, Niger-Congo, Afro-Asiatic, Sino-Tibetan, etc) with very few isolates, and we can make some sense of the history and prehistory of the supercontinent almost on that basis alone.  But South America is nothing like this.  Only in the Andes is there a coherence and lack of bizarre geographic spread, with Quechua and Aymara, or languages related to these, found across most of the region formerly dominated by Tawantinsuyu.

The archaeological situation is also troubling.  This is partly because there are few language families to correlate with archaeological cultures, making their spread difficult to understand.  It's also because no South American group developed an indigenous writing system.  The khipu is arguably different - but only arguably, and even the most charitable interpretations of it see it as a device for recording data in a currently non-decipherable way.  There was no written tradition on the entire continent before the arrival of the Spanish in the sixteenth century, despite what you may have seen in Indiana Jones and the Kingdom of the Crystal Skull.

Lacking any sort of decipherable written language in which to look for answers, archaeologists inevitably disagree in their interpretations of the archaeological data.  Wari (or Huari), a Middle Horizon* archaeological culture of the highlands, is a typical example.  Moseley considers the Wari material to be evidence of the largely non-militaristic expansion of a religious cult.  Bruhns believes instead that Wari was a state that expanded militarily, rather than a cult.  The archaeological data are silent on the issue; certainly Wari seems to have represented a single expanding group of people and their religious beliefs, and certainly some violence occurred in Wari settlements and in confrontations with other groups.  But what else is new?

There is evidence of Wari-constructed roads, and Pikillaqta, a Wari site in the southern Cuzco valley, seems designed to control traffic, which is consistent with a state.  Wari people also seem to have made use of the khipu, and they used bronze - some of the first use of tin rather than arsenical bronze on the continent.  It seems to be a prototype of Tawantinsuyu.  But as Tawantinsuyu also seems to have included a fair amount of religious motivation in its expansion, it is hard to draw the line between cult and state, just as it would be for archaeologists studying Arab expansionism/Islam in ancient Afro-Eurasia.

Wari is a well-studied and relatively well-known archaeological culture from the best-studied and best-known part of the continent, the Cordillera.  Wari artifacts are found from Ecuador to Bolivia, the latter being at the time the heartland of the Tiwanaku civilization (I have posted about the Tiwanaku civilization before).  It is, with the exception of Tawantinsuyu, one of the most well-understood social phenomena in pre-Hispanic South America, and yet good, learned anthropologists and archaeologists disagree about its most fundamental facets.

Studies of the rest of the continent lag behind even this.  The Tupi-Guarani family, for instance, may have spread from the Atlantic coast of Brazil in the middle of the first millennium CE, as indicated by a pottery style emerging there around 500-700 CE, which has been found among early-contact and present-day Tupian-speaking people and which correlates with the general geographic spread of Tupi-Guarani speakers.  There is also, according to Bruhns, evidence of post-Hispanic migrations of Tupian peoples, often groups of people led by a prophet, showing that such large-scale migrations are far from impossible.  But the identification of the material as Tupi-Guarani is not as certain as the identification of the Yamnaya horizon material around the Black Sea with speakers of Proto-Indo-European, due to the scattered nature of the Tupi-Guarani language family, the limited archaeological evidence, and the peculiarities of the climate and terrain of lowland South America that make exploration and excavation very difficult indeed.

A number of things can be said in general about the continent, though.  One is that dualism is extraordinarily prevalent.  Dualism in this context is a tendency to categorise phenomena in pairs - day/night, black/white, etc - and it can also be used to describe certain sorts of sociological arrangements.  Found across South America are moieties; 'moiety' comes from the French for 'half', and in the terminology of kinship it means a pair of exogamous descent groups that marry one another.  A man from moiety A can't marry a woman from moiety A; he has to marry someone from moiety B, and from no other group than this.  A man from moiety B has to marry a woman from moiety A, often out of preference a cousin of his.  There is much more to it than this, but fundamentally it is a prescriptive**, eternal, dualistic alliance between two groups of people.

It also correlates with, but doesn't appear to determine, a view of the universe divided into two, such that people with a strongly dualistic social structure often classify things into pairs.  You might oppose 'night' to 'day', and 'women' to 'men', for instance - and then see in these oppositions a similarity of form, meaning that you then correlate all of the classifications together, such that male, day, activity, hunting, the sun, and the colour red (for instance) become associated with one half of the cosmos while female, night, passivity, agriculture, the moon, and the colour black become associated with the other half.  These kinds of correlations of dualisms are found almost wherever a strong tradition of dualism is, which is, frankly, a very puzzling fact about people.***

Social-symbolic dualism is so pervasive in South America that a strongly dualist tradition has been considered part of the culture of the first migrants into South America 12,000 years or so ago (Moseley suggests this, for instance).  This sort of arrangement is famously found in the lowlands (Levi-Strauss devoted much of his work to studying dualistic social and symbolic arrangements from the Americas, including especially Brazil), and also in the highlands, where the entirety of Tawantinsuyu was divided into hanan (upper) and hurin (lower) classifications, and where things as diverse as plants, animals, colours, and cosmological principles were divided neatly into two opposed categories, at least for Inka purposes (possibly not shared by the other ethnic and cultural groups over whom the Inkas were lords).

The Inka empire was dualistic in a profound sense.  But it was also divided into four parts centred on Cuzco (Qosqo in Quechua).  Tawantinsuyu means 'the united four parts', from tawa, 'four', plus -nti, 'a numbered group or set' and suyu, a part or quarter.  In the north was Chinchaysuyu, to the south was Collasuyu, with Cuntisuyu occupying the southwest and Antisuyu occupying the northeast respectively, the latter two segments being much smaller than the former two (here is a map showing this more clearly than I can describe it, with Chinchaysuyu rendered as 'Chinchansuyu').  These were also classified dualistically, however, such that hanan (upper) Cuzco, Chinchaysuyu, and Antisuyu were classified together as the hanan part of the empire, and hurin (lower) Cuzco, Collasuyu, and Cuntisuyu were the hurin part of the empire.

There was also a hanan and a hurin part of each of the quarters, and within these subdivisions were villages further divided into hanan and hurin sections - intermarrying moieties.  It is also possible that the empire was ruled by a pair of kings: a diarchy, not a monarchy (which fascinates me, as my principal interests are in Indonesia and the Pacific - native states throughout these regions were diarchies, including many of those of Timor, Tonga, and Hawai'i).  The evidence is incomplete on this point, however.

Dualism is often recursive.  You may have heard people jabbering on about yin and yang in a Chinese context, talking about how there's a bit of yin in the yang, and a bit of yang in the yin, and that there's a bit of yang in the yin-part of the yang, and a bit of yin in the yang-part of the yin, and so on to infinity.  That's crazy New Age gibberish most of the time, but it also gets at a key issue in the study of dualism, one often overlooked.  Dualism allows for a nested hierarchy of classifications.  It is consequently quite a useful way of conceptualising and managing a country or empire if you don't have a writing system or a bureaucracy reliant on it.  You can think 'the hanan village in the hurin part of hurin Chinchaysuyu' instead of making exhaustive lists of the names of villages, lists that a pre-literate society can only record mentally.  That may partly explain why social/symbolic dualism is prevalent in non-literate societies throughout the world (as well as some literate ones), but the heritage of the first South American population may also have had something to do with it.

Dualism is also very important in understanding the khipu, and Gary Urton's book argues essentially that khipukamayoq in Tawantinsuyu - the bureaucrats of the Inka empire who used khipu, knotted cords, to record information - relied on a dualistic classification of the universe to record information in the cords.  Cords can be plied in two directions, S-ply or Z-ply, and they can be attached to the main cords using one of two knots.  The colours used to dye the cords can be classified as either dark or light, and two different kinds of materials can be found in the construction of the cords on individual khipu - llama wool or cotton, for instance.  Dualism pervades the system.

The khipu seems to have a device for recording information dualistically, so that a person aware of the classifications of the specific part of Tawantinsuyu under discussion would be able to work out which segment of which village of which part of which quarter were being recorded.  A Z-ply cord could mean hanan Chinchaysuyu to the khipukamayoq recording the information, where an S-ply cord could mean hurin Chinchaysuyu.  The khipu would then be a means of recording not language, but information about parts of the empire divided into paired sections.  This fits a dualistic view of society and the cosmos, and would work well to record information - especially mathematical information, whose recording on the khipu is well-understood and much less controversial than Urton's theory - for bureaucratic and religious purposes in a dualistically-conceived multi-ethnic empire.

Well, that's just a theory.  But it's fascinating.  I read Urton's book at Oxford, and the possibilities of the theory literally kept me awake thinking about what it could mean for what we know about the ways in which humans can record information, create bureaucracies, govern empires, and recite epics and sagas, all without needing to record speech or anything like what we conventionally call 'language'.

In any case, South America is probably the most mysterious and least understood continent on earth in terms of its ancient human population.  That seems to be generally true.  But the things that are known about it are incredible and exciting.  The disagreements archaeologists and linguists have about ancient South America are a lot like the disagreements they once had about other parts of the earth, including the Mayan world and ancient Polynesia, and it is fascinating, and pretty darn wonderful, to watch as information comes in and new ideas are created in order to understand it.  South American archaeology isn't my main area of interest - as I said, I don't speak any of the local languages well, and I haven't devoted my academic life to studying it - but it provides for me one of the greatest pleasures in the world: the pleasure of finding things out.

*First millennium CE.

**'Prescriptive' means that the behaviour is prescribed - it is mandatory, and there are punishments for transgressions - rather than simply preferred.  Lots of people have wanted to blur the line between the two, and there are good reasons for doing so, but here the distinction seems reasonable.

*** Dualism is an annoying topic because the people who dominated studies of it until recently - Claude Levi-Strauss and Louis Dumont - were resistant to integrating it into broader cognitive science.  They treated it as separate to the rest of the functioning of the brain or, in Levi-Strauss's case, saw it as the dominant factor in all of human thought.  This is not the case.
Fundamentally, what Levi-Strauss and Dumont were suggesting was no different in principle to cognitive science - it was just more restricted and much less useful than it, and focussed on a smaller set of more esoteric subjects.  I think that if you have no problem with Dumont, you should have no problem with cognitive science - but this seldom seems to hold true.  It's sad.

No comments:

Post a Comment

You can post anonymously if you really want to, but I would appreciate it if you could provide some means of identifying who you are, if only for the purpose of knowing who has written what.