More thoughts on the khipu code.

Abstract

Charles C. Mann's article on new attempts to decode Incan knotted strings or khipu (“Cracking the khipu code,” News Focus, 13 June, p. [1650][1]) notes that the predominant view is that these are not a written language but simply mnemonic devices or, at most, accounting tables. He rightly observes that a successful translation would be absolute proof of a language. Failing that, it would appear that the majority would hold to a null hypothesis of no written language. I suggest that the contrary is a more parsimonious hypothesis and that the burden of proof lies with those who deny an Incan written language. Most people are familiar with the old parlor game where participants sit in a circle; one receives an oral message to be whispered to one's neighbor, who repeats the process with his neighbor, until the message has passed around the circle. Input and output are usually so different as to provoke much laughter. Over 50 years ago, I directed a rather large field research project for the military, recording and following oral communications both to equals and up and down the chain of command. The demands of military precision did not improve communication accuracy over that of the parlor game. Given the nature and inefficiency of human oral communicative behavior, it is inconceivable that the Incas could manage the largest empire in the world without written communication. If indeed they did so, then they would have had to develop behavioral systems that were far beyond the capabilities of modern behavioral knowledge and organizational management. Such cultural capabilities would be a matter of enormous fundamental research interest and would certainly be attractive to corporate benefactors anxious to learn major organizational secrets that could affect business and government effectiveness. [1]: /lookup/doi/10.1126/science.300.5626.1650