Archive/ The Vault/The Inca Khipu
The Vault · Texts & Tablets
The Inca Khipu: An Empire Tied in Knots
In the archive
- Held
- About 1,000 or more in museums worldwide (Lima, Berlin, New York, Harvard and elsewhere)
- Date evidence
- Mostly c. 1400–1600 CE; earlier Wari examples c. 600–1000 CE
- Era
- Era VII · The High Medieval
- In the archive
- The Inca
In 1583, the Third Council of Lima, the Catholic bishops of Spanish Peru, ordered that the Indigenous peoples' khipus be taken and burned, because they were used to record "idolatry" and to confess sins in the old way. The khipu was the knotted cord of the Andes: a main cord from which hung dozens or hundreds of coloured pendant cords, knotted in patterns. With them the Inca, who ruled the largest state in the Americas, counted people, llamas, tribute and time. The Spanish chroniclers said that the khipu specialists, the khipukamayuq, could also read out histories, laws and songs from them. Whether that is true is still unknown.
Read a khipu
An illustrative khipu built on the rules Leland Locke worked out in 1912. Touch or hover over a cord to read its number, then the top cord.
Cord 1: 23 · Cord 2: 141 · Cord 3: 8 · Cord 4: 1 · Cord 5: 60 · Cord 6: 205 · top cord: 438
Study the object
Images of khipus belong to their holding museums; the Open Khipu Repository publishes data on many of them.
The holding institution's own viewers
Paste any IIIF manifest into the viewer to study published images; it needs JavaScript.
The numbers
In 1912 the American historian Leland Locke showed how the numbers worked. Each pendant cord holds a number in base ten: clusters of single knots mark hundreds and tens, and the units position uses a long knot of 2 to 9 turns, or a figure-eight knot for one. An empty position means zero. Some khipus have a top cord recording the sum of a group of pendants, and it was these sums that proved the knots were numbers. Later work by Marcia and Robert Ascher (1981) documented how khipus were structured, with colour, cord order and subsidiary cords carrying further information.
The interactive khipu above follows Locke's rules. It is illustrative, not a copy of a surviving khipu.
Words in knots?
Many khipus do not follow the numerical rules. That, with the chroniclers' reports, suggests that some khipus recorded words, names or narrative.
- Gary Urton (2003) proposed that khipus encode information in binary choices: cord material, direction of spin, knot direction.
- In 2016, a Harvard student, Manuel Medrano, working with Urton, matched six khipus to a Spanish census of the 1670s, and found that cord features might mark names.
- In 2017 Sabine Hyland studied two khipus kept by the village of Collata in Peru, said by villagers to be letters written in a rebellion. They use 95 distinct combinations of colour and fibre, enough for a logosyllabic system, and she proposed that the final cords spell family names.
None of these is an accepted decipherment. The question of narrative khipus remains open.
Symbology and sacred encoding
Knots in a sacred landscape. Around Cusco, the Inca organised the valley's shrines (huacas) along some 41 lines, the ceques, radiating from the Temple of the Sun. The anthropologist R. T. Zuidema compared the ceque system to a khipu laid on the landscape; the comparison is influential but interpretive.
Colour. Cord colours were meaningful, possibly marking categories such as types of tribute or social groups. What each colour meant is mostly unknown.
Religious use and suppression. Chroniclers report khipus used to record sacrifices, the calendar of festivals, and confessions. Hence the Church's decree of 1583. Some Andean villages, however, kept khipus in ritual use into modern times.
Connections
The Inca (ch44). Compare the Dresden Codex (V31) and the Codex Mendoza (V51), a record of Aztec tribute, for other record systems of the Americas, and the Phaistos Disc (V56) for an undeciphered text.
What's well supported by the evidence
- Khipus used by the Inca (and earlier by the Wari) to record numbers in a decimal system, as shown by Locke in 1912.
- Chroniclers' reports that khipus also recorded history and other information, and that the Church ordered them destroyed in 1583.
What is not supported
- Any claimed full decipherment of narrative khipus.
What's genuinely open
- Whether some khipus recorded language and, if so, how.
- How far the Collata and census studies can be extended to the rest of the roughly 1,000 surviving khipus.
Sources
The khipus
- "Quipu", https://en.wikipedia.org/wiki/Quipu
- Harvard Khipu Database Project, https://khipukamayuq.fas.harvard.edu/
- L. L. Locke, "The Ancient Quipu, a Peruvian Knot Record", American Anthropologist 14.2 (1912)
- M. Ascher and R. Ascher, Code of the Quipu (University of Michigan Press, 1981)
Decipherment
- G. Urton, Signs of the Inka Khipu (University of Texas Press, 2003)
- S. Hyland, "Writing with Twisted Cords: The Inscriptive Capacity of Andean Khipus", Current Anthropology 58.3 (2017), https://www.journals.uchicago.edu/doi/10.1086/691682
- Harvard Gazette, "Student finds clue to deciphering khipus" (2017), https://news.harvard.edu/gazette/story/2017/01/in-search-of-khipus-meaning/
- R. T. Zuidema, The Ceque System of Cuzco (Brill, 1964)
From The Divine Archives · https://getconexto.com/vault/inca-khipu.html