The Paradox of (Inka) History

A Review by Daniel Lord Smail

Inka history 

Some years ago I went to the South Tyrol Museum of Archaeology in Bolzano, in Italy’s Alto Adige, to gaze upon Ötzi. Better known as the Iceman, Ötzi was an early Bronze Age traveler and homicide victim whose well-preserved body was accidentally discovered in 1991 as it emerged from a melting glacier. But however interesting it was to see the blackened body of the poor fellow in deep-freeze storage, more compelling to me were the interpretive exhibits displaying all the equipment that Ötzi had with him when he was assassinated, from his boots and leggings to his backpack. A goodly amount of his stuff was held together by string and cord made from the bast fibers of lime wood. At one interpretive station I sat down to learn how to make Ötzi-rope. It’s an addictive experience. If you’ve done it you’ll nod knowingly as I describe how the twist of the strands of fiber has to turn against the rotation of the braid, because if you don’t, the thing just unravels. It is nearly impossible to explain this in words. It’s a kind of skill that is first learned in the fingers and then remembered by the fingers.

Bear this in mind when you sit down to read Gary Urton’s new book; it’s helpful to have some cord in hand when you do so. I used a nice, round shoelace. Are you curious to know what an S-twist is and why is it different from the Z-twist? Just do it; you’ll figure it out. One helpful diagram has pictures of all the major knots, including the long knots that sign the digits from one to nine. The first one I tried was a real mess, but in time, I got it right. This experience, however fleeting, was central to my appreciation of the book. It made me feel what it was like to be a khipukamayuq, a maker of khipus.

Khipus, consisting of knots on cords, were the devices used by Inkas to keep their accounts. Urton proposes to write a new kind of Inka history from these sources. As Urton would be the first to acknowledge, Inka History in Knots is not itself a history. At the heart of the book lies a series of chapters, some of them filled with dense mathematics-based reasoning, that bring the reader up-to-date on some of the main lines of khipu studies. The chapters are bookended by a proposal to rethink the kind of Inka history that we ought to be writing. Inka history may never fit the linear narrative we typically associate with accounts of the past. But the problem here doesn’t necessarily lie with the primary sources available from Tawantinsuyu. It may lie instead with the narrative box into which we have been trying to package those sources.

History, it is often said, is written by the victors. In the case of the Inkas, much of that history has been based on Spanish accounts, meaning that we see Inka civilization through the eyes of the conquerors. Generations of scholars have expressed their misgivings about the need to give credence to such sources. In the same way that archaeologists constantly remind themselves that processes of site formation skew the available evidence, historians are skeptical about their own evidence. They approach any given document with the assumption that its author was perfectly capable of being mistaken or of lying. Those who deal with documents left by colonial powers or other dominant elites, in contexts where power relations are so nakedly present, are especially attentive to the fictions of the written record.

Where available, the findings of historical archaeology sometimes provide a check against the tissue of falsehoods that suffuses every document. As Urton points out, however, archaeology is not a wholly satisfying alternative. Archaeology and archaeoscience are constantly generating new information in the form of artifacts, sites, isotopes, and so on. Yet even with the best intentions in the world, no archaeologist can avoid putting words in someone else’s mouth. Archaeologists, unwittingly, find themselves in the position of the conqueror who tells an Inka history without Inka voices. As any social anthropologist would point out, of course, we cannot assume that those voices will get us any closer to the “truth.” But at least it would give the Inkas a say in the rendering of their own history.

This is the great paradox of history: how does the historian allow subjects to speak for themselves when any act of history writing is an act of ventriloquism? The problems exposed by the paradox are clear enough in a context where a literate civilization has conquered and absorbed a non-literate civilization, very nearly destroying the entirety of that civilization’s historical record in the process, and has then set about to write the lost history of the latter. But the problem suffuses every history, even those that lie inside the conventional boundaries of the nation state. What right does a medieval European social historian like me have to speak for the medieval peasants or artisan when, so often, I have to rely on sources written by and for the dominant classes?

Social historians have been sensitive to this problem since the days of E. P. Thompson, the great historian of the English working class. One of the many duties arising from this awareness has been to search for documents that might allow the members of a subaltern population to speak for themselves. In the case of modern history, these kinds of sources are readily available. It is true that they get harder to find the further you go back in time. But even in contexts where subaltern-authored texts are rare in relation to other historical and archaeological sources, they nonetheless act like a vital leaven in a bread dough, bringing the whole thing to life.

Among the world’s great civilizations, the Inkas are unusual insofar as they did not produce their own texts. More accurately, they did not keep texts based on alphabetical characters that lend themselves to narration. This suggests that the Inkas can never speak themselves. Or does it? Urton offers an elegant solution to the great paradox of writing Inka history, arguing that where the Inka past is concerned, we have been deploying the wrong historical model. In effect, we have been trying to pound the round peg of Inka sources into the square hole of linear, narrative history. One of the most subtle and pervasive legacies of Spanish hegemony may in fact be the linear form of history-writing itself, since this form cannot fit the evidence from Tawantinsuyu. If the khipus were allowed not only to speak for themselves, but also to determine the form of history to which they spoke, what would that history look like?

This question brings us to the different kinds of khipus whose analysis occupies the central chapters of the book. These range from khipus bearing census-type data to accounts of labor services and inventories of beans and chiles. The world’s collections currently hold 923 khipus of which many date from the period of Inka hegemony. There is every hope and expectation that more will continue to be discovered. At present, khipus cannot speak to the linear forms of narrative history favored by civilizations that possess alphabetical characters because the so-called narrative khipus have yet to be deciphered. But as Urton demonstrates, we are beginning to develop the tools needed to read the data-bearing, non-narrative khipus that constitute the bulk of the archive.

Although the non-narrative khipus don’t lend themselves to narrative histories, Urton points out that they easily fit into a different kind of history writing, notably the social-scientific approach to history, based on census data, tax records, records of land tenure, and other serial sources, that was pioneered by the French Annales school in the 20th century. Where alphabetical writing is about narrative, Urton argues, cord-making is about atemporal structures. If I understand him correctly, moreover, he suggests that because Inkas “thought” with khipus rather than letters, they were led by the very form of their archival system into a worldview that did not bracket the past as a distinct place. Among other things, Inkas were prone to a view in which the matter of the world is organized into binary sets, a framing amply in evidence in the khipus themselves.  

There is an obvious objection to Urton’s proposal: Inka history has not exactly been emancipated from European historiography if one Eurocentric historical model has simply been substituted for another. But this is beside the point, for Urton isn’t proposing that historians and archaeologists of the Inka world should absorb the Annales paradigm in some naive and uncritical way. Rather, his point is to suggest that scholars writing Inka history should be as attentive to the form of the history as they are to the contents. Truly original Inka histories may come of this. And thanks to this marvelous book, I, for one, shall be eagerly awaiting the results.


Daniel Lord Smail is Frank B. Baird, Jr., Professor of History at Harvard University. He writes about medieval European social history and the deep history of humankind.