Tying the Archive in Knots, or: Dying to Get into the Archive in Ancient Peru

Abstract

Record keeping in the Inka Empire of the Andes of ancient Peru was based on the knotted-string recording device, the khipu (or quipu; Quechua: ‘knot'). Khipus were produced and consulted by Inka administrators for a variety of purposes, including the recording of censuses, tribute data, as well as life histories and genealogies of the Inka nobility. Cord-keepers were organized in a hierarchical arrangement of officials, from local khipukamayuqs (‘knot makers/organizers'), to higher-level officials who staffed provincial administrative centers, to state cord-keepers in the capital, Cusco. The khipu-keepers stored collections of khipus in regional centers and in Cusco where they could be consulted on a variety of matters of interest to the state. This study looks first at the way information was recorded on the knotted-cord records. This is followed by an overview of what we know to date about archival collections of khipus, including a close study of a colonial era khipu archive from the Santa Valley, on the north-central coast of Peru. Of particular note is the fact that many khipus were stored in burial chambers with ancestral mummies, a situation that left these records accessible to descendants of the ancestors, who visited the burial chambers where they paid tribute to the mummies and consulted the knot records.

Notes

The project website of the Khipu Database (KDB) at Harvard University is at: http://khipukamayuq.fas.harvard.edu/

Locke, ‘The Ancient Quipu’.

Urton, 2005.

At the time of writing, we have information in the KDB on construction materials for 40% (i.e., 15,616 cords) of a total of 38,831 cords. Of the 15,616 cords, 85% (13, 297 cords) are made of cotton, 9% (1,357 cords) are made of camelid, and 6% (962 cords) are of sheep's wool, vegetal fibre, hair, or other materials. Cotton appears to have been the material of choice; however, it is important to note that every khipu recovered to date comes from a cotton-producing region, including the coastal deserts of Peru and northern Chile, or from the Utcubamba (‘cotton plain’) River of the northern Peruvian highlands (Chachapoyas). If camelid khipus were more common in the highlands, which was home to the huge herds of llamas, alpacas, and vicuñas tended by the subjects of the Inkas, we might expect that, were conditions of preservation to have been more favourable in the highlands, there would now be a much higher percentage of camelid fibre khipus in our corpus.

Conklin, 2002:60.

Arellano, 1999.

Locke, 1923; Ascher and Ascher, 1997; Julien, 1988; Pereyra S., 2006; Urton, 2003.

Guaman Poma de Ayala (1980 [1615]).

Guaman Poma, 1980:330 [359].

Guaman Poma, 1980:332 [360].

Guaman Poma, 1980:336 [364].

Guaman Poma, 1980:331 [361] .

Julien, 1988.

Murra, 1980.

Pärssinen, 1992.

Cieza de León, 1967 [1553; Cap. XII]:36-37; translation by Urton.

Garcilaso 1966 [1609]:275.

Palaima in Fissore, 1994:358.

Urton and Brezine, nd.

Villacorta O., 2005.

Urton and Brezine, 2005.

Urton and Brezine, 2007.

von Dassow, 2005.

Radicati, La ‘Seriación’ como posible Clave para Descifrar los Quipus Extranumerales (‘Seriation as a Possible Key to the Decipherment of Extra-numeral Khipus’), 2006.

Radicati, La ‘Seriación’ 2006.

Radicati, 2006:158-9.

While such an arrangement of colour differentiated sets of cords – known as colour ‘banding’ – is found in many other khipus in the corpus of extant samples, such similarity in structure and colour is unusual in a set for which there exists at least a claim that they were recovered together from the same tomb (another such example, but with secure provenance, is the Puruchuco archive; see Mackey, 1970, and Urton and Brezine, 2005).

Zevallos Quiñones, 1991.

Brokaw, 2010; Urton, 2002.

Salomon, 2004.

González Holguín, 1952.

Lerche, 1999.

see von Hagen and Guillen, 1998; Muscutt, 1998.

Urton, 2001.

see Doyle, 1988; Salomon, 1995.

Isbell, 1997.

Doyle, 1988:110.

Doyle, 1988:68; see Mills, 1997, on colonial attempts to destroy the mallkis, as objects of idolatry.

Doyle, 1988:61, 117, 135-137.

Withers, 2002:305-6.