Knot-Words or Not Words

Frank Salomon, The Cord Keepers: Khipus and Cultural Life in a Peruvian Village. Duke University Press, 2004, 331 pp.

The Khipu Enigma

I felt a pang the other day when I read about multi-spectral imaging techniques that are recovering inscriptions on papyrus fragments over two thousand years old. Rescued from an ancient trash heap south of Cairo, this collection of 400,000 bits of trash has sat in Oxford for over a century, painstakingly curated, and yet the vast majority burned and discolored beyond legibility. With the new technique, a by-product of satellite technology, words from the distant past are coming into focus—seventy lines of a lost play by Sophocles, thirty lines from a seventh century B.C. poet named Antilochos. We can look forward to new (to us) poems by Pindar, Sappho and Euripides, writings by early Christians—and of course the detritus of everyday Greek-Egyptian life, like tax returns and horoscopes.

So why should this interesting news send a pang through my heart? Because I study the Andes, home to a great civilization whose trash heaps and tombs yield up no ancient words. As far as words are concerned, we have to make do with Spanish accounts of post-Conquest Inka society, and with the very rare colonial era documents written by native Andeans in their own languages, like the Quechua manuscript written by an anonymous author in Huarochirí around 1608. Pre-Columbian Andean inscribed no words on paper, stone or wood. We can cull no words from the artifacts they left behind them.

What they did leave behind is amazing quantities of marvelous cloth. Fiber arts have great time depth ( at least five millennia) in Andean civilization and must have been deeply meaningful, as they still are in parts of the rural Andes today. Not surprisingly, fiber was the medium in which Andeans developed techniques for recording information. By the Middle Horizon (c.1000AD) they were keeping records on complex sets of knotted strings. When Spaniards arrived in the early sixteenth century the Inkas were using these string devices to administer an extremely far-flung empire, stretching from southern Colombia to northern Chile. Spanish chroniclers were impressed with the amount, complexity, variety and reliability of information stored in these khipu s (the word means "knot" in Quechua). Early-on the colonial regime allowed evidence from khipus (though not the khipus themselves) to be entered in court records and even required that native authorities keep tribute accounts on khipus. By late sixteenth century, however, Spanish inquisitors had caught on to another aspect of these string devices: A very un-Christian type of sacredness inhered in them; khipus recorded a complex system of shrines housing ancestral mummies, and some were even interred with them. Thus, khipus were outlawed as idolatrous; and khipu-users, if they persisted, did so at great risk. Standard histories assume that khipu-use died out at this point, and that by early seventeenth century khipus were a things of the past.

Unfortunately the colonizing Spanish never figured out how the khipus actually worked; apparently the system was too foreign to learn. The Inka and pre-Inka khipus left to us—mostly looted and totally without provenance—are silent in our hands. No one has ever turned up a Rosetta Stone equivalent (as, for example, a khipu attached to a written account of its contents).

Thus the Andean khipu remains a enigma. In its knotted strings we find a sophisticated communication device, developed in relative isolation from the rest of the world and apparently premised on principles radically different from those used in other forms of writing. Inka khipus are of a mind-boggling size and complexity that clearly goes beyond a "knot-on-the-finger" type of mnemonic device. "Decoding" them would not only to transform our understanding of one of the world's great (and least understood) empires, but could provide new perspectives on human intelligence as well. A new book by Frank...