The quipu or khipu is both ordinary and mysterious. Made from cotton or wool knotted cords, it was the backbone of the bureaucratic and centralised Inca Empire, used to record amounts of goods and numbers of people. Computations were decimal, the highest knot standing for one, the next for 10, then 100, 1000 and so on. Each ruler, governor, commander and village chief had a quipucamayoc, the quipu-teller. The sons of the nobility studied to read a quipu in their training college, as it was always necessary to know the truth of the quipu-tellers’ numbers. Used for keeping records, the quipu was also a mnemonic device. Different dye colours, lengths of cords, types of knots, weaving direction and number of plies all changed meaning. The size varied from a few cords to more than a thousand. No-one nowadays can read a quipu.

Recent research, particularly from that of Gary Urton, as well as Marcia and Robert Ascher, suggests that quipus were used for narrative as well as statistical information.¹ Perhaps place names, stories of battles, genealogies were ‘read’ by the quipu-tellers. Quipus were used by the Incans for census taking, for astronomical and calendrical information—very important in an agricultural economy—and to account for loads of maize, flocks of llamas and bottles of chicha or corn beer due as tribute. They must have been universally legible, as the road messengers transported them around the empire.

Spanish colonisers witnessed quipus being used, although the native system was soon replaced by written records. In 1533, the first year of the occupation, Hernando Pizarro, brother of the conqueror Francisco Pizarro, told of how he and his soldiers removed firewood, llamas, corn and chicha from an Inca storehouse, ‘and the native accountants recorded the transaction on a knotted-string recording device’, untying some knots on 
the deposit section, and retying them in another section.² 
The Spanish soon suppressed the quipu as part of the idolatrous Inca religion, and now only about 750 are known to survive.

Quipus are associated mainly with the Incas, as they were the ruling government when the Europeans took over, and their centralised authority was noted by the newcomers, as was their control over the production and distribution of goods. Like many other elements of their society, however, the Incas adopted earlier inventions and discoveries, and the use of quipus seems to have been widespread in Andean civilisations. The Huari highland culture of 600 to 1000 AD certainly used quipus, although perhaps not in precisely the same manner as the Incans.

The Amano quipu is composed of 169 separate cords in 20 groups in five colours: white, brown, blue, cream and red. The Larco quipu consists of 100 principal cords and seven subsidiary ones, 
in green, brown cream and white. The cords’ placement above and below the central keeping cord was also significant for different meanings.

Christine Dixon

1. For example, see Jeffrey Quilter and Gary Urton (eds), Narrative threads: Accounting and recounting in Andean khipu, Austin: University of Texas Press 2002.

2. Gary Urton, ‘An overview of Spanish colonial commentary on Andean knotted-string records’, in Quilter and Urton, p. 6.

The quipu or khipu is both ordinary and mysterious. Made from cotton or wool knotted cords, it was the backbone of the bureaucratic and centralised Inca Empire, used to record amounts of goods and numbers of people. Computations were decimal, the highest knot standing for one, the next for 10, then 100, 1000 and so on. Each ruler, governor, commander and village chief had a quipucamayoc, the quipu-teller. The sons of the nobility studied to read a quipu in their training college, as it was always necessary to know the truth of the quipu-tellers’ numbers. Used for keeping records, the quipu was also a mnemonic device. Different dye colours, lengths of cords, types of knots, weaving direction and number of plies all changed meaning. The size varied from a few cords to more than a thousand. No-one nowadays can read a quipu.

Recent research, particularly from that of Gary Urton, as well as Marcia and Robert Ascher, suggests that quipus were used for narrative as well as statistical information.¹ Perhaps place names, stories of battles, genealogies were ‘read’ by the quipu-tellers. Quipus were used by the Incans for census taking, for astronomical and calendrical information—very important in an agricultural economy—and to account for loads of maize, flocks of llamas and bottles of chicha or corn beer due as tribute. They must have been universally legible, as the road messengers transported them around the empire.

Spanish colonisers witnessed quipus being used, although the native system was soon replaced by written records. In 1533, the first year of the occupation, Hernando Pizarro, brother of the conqueror Francisco Pizarro, told of how he and his soldiers removed firewood, llamas, corn and chicha from an Inca storehouse, ‘and the native accountants recorded the transaction on a knotted-string recording device’, untying some knots on 
the deposit section, and retying them in another section.² 
The Spanish soon suppressed the quipu as part of the idolatrous Inca religion, and now only about 750 are known to survive.

Quipus are associated mainly with the Incas, as they were the ruling government when the Europeans took over, and their centralised authority was noted by the newcomers, as was their control over the production and distribution of goods. Like many other elements of their society, however, the Incas adopted earlier inventions and discoveries, and the use of quipus seems to have been widespread in Andean civilisations. The Huari highland culture of 600 to 1000 AD certainly used quipus, although perhaps not in precisely the same manner as the Incans.

The Amano quipu is composed of 169 separate cords in 20 groups in five colours: white, brown, blue, cream and red. The Larco quipu consists of 100 principal cords and seven subsidiary ones, 
in green, brown cream and white. The cords’ placement above and below the central keeping cord was also significant for different meanings.

Christine Dixon

1. For example, see Jeffrey Quilter and Gary Urton (eds), Narrative threads: Accounting and recounting in Andean khipu, Austin: University of Texas Press 2002.

2. Gary Urton, ‘An overview of Spanish colonial commentary on Andean knotted-string records’, in Quilter and Urton, p. 6.