Deciphering the Andean Knotted-String Records: A Project Aimed at Creating and Querying a Khipu Database

Harvard University, Cambridge MA

Investigators

Abstract

The objective of the two-year project to be undertaken by Dr. Gary Urton at Harvard University with the support of a research grant from the National Science Foundation is to create and query a database composed of information from study of surviving samples of Inka khipu. The khipu (Quechua: "knot"), were the principal devices used for record-keeping in the Inka empire. There are some 600 surviving samples of these devices in museum and private collections around the world. These artifacts, which are composed of dyed and knotted cotton and camelid fibers, were used by Inka administrators to record both statistical data and narrative information. Although khipu have been studied by researchers for several decades, these ancient knotted texts remain largely undeciphered. Working with a database manager, Urton will coordinate efforts to design and build procedures for entering data recorded by earlier researchers, as well as new data on khipu construction recorded from recently completed and on-going museum studies by P.I. Data entry will proceed throughout the '02-'03 academic year. In Spring, '03, work will begin on two additional components of the project. First, procedures will be designed to query the khipu database, searching for patterns in numbers, colors, construction techniques, etc., that might help researchers to understand how information is organized in the khipu and, therefore, how one might theorize the nature and procedures of encoding and decoding (or reading) these ancient records. Secondly, work will be undertaken on the design and construction of supplementary electronic databases drawn from two types of textual sources that may help in querying the khipu database; these sources will be, first, a text written in colonial times in the Quechua language (probably the Huarochiri Manuscript), and secondly, material from documents transcribed by Dr. Urton from archival research in Seville, Spain, containing Spanish transcriptions of khipu readings that were given by native recordkeepers in the Andes in early colonial times. Data storage and querying procedures will be developed that will allow researchers to cross-reference and query the khipu database from information contained in the databases of Quechua textual and ethnohistorical materials. The khipu database will be posted on the WWW, along with instructions for how researchers can request queries of the database. The objective of the project is to attempt -- if not the decipherment of the khipu -- at least a significant advance in our understanding of this remarkable indigenous South American recording device.