Quipu and Record Keeping

The quipu (also spelled khipu or quipo, from the Quechua word khipu meaning “knot”) was a knotted-string recording system used throughout the Inca Empire and in many of the peoples it incorporated. A quipu consisted of a primary cord, from which hung a series of pendant cords, each of which carried subsidiary cords. The pendant cords were marked with knots tied at specific positions, and the knots were combined in a system of positional notation to record numbers. The color of the cords, the direction of the knot twists, and the arrangement of the subsidiary cords encoded additional categories of information, including the type of record, the subject, the source, and — as recent research has suggested — possibly narratives, names, and historical information. Quipus were used to record censuses, tribute lists, military rosters, storehouse inventories, and historical narratives, and they were an essential instrument of the Inca provincial administration. Although the Spanish conquest of the 1530s and the subsequent suppression of indigenous record-keeping practices led to a long period in which the quipu was treated as a primitive mnemonic device, recent archaeological and historical research has demonstrated that the quipu was a complex semiotic system that recorded not only numbers but also categories of information that could be read by a trained interpreter.

Origins and Pre-Inca Antecedents

The knotted-string recording system of the Andes has a long pre-Inca history. A number of knotted-cord objects have been recovered from Wari sites, including the well-known find from the site of Conchopata, and these have been interpreted as recording devices used in Wari administration. The Inca inherited and expanded the knotted-string tradition. The chronicler Garcilaso de la Vega, in his Comentarios Reales de los Incas (1609), wrote that the quipu was a “royal invention” of the Inca state, and that the first quipus were introduced by the ninth Sapa Inca, Pachacuti, in the fifteenth century. The claim is supported by the dramatic increase in the number and complexity of quipus in the archaeological record of the late fifteenth and early sixteenth centuries, and by the fact that the most elaborate surviving quipus — those from the Inca period — are found in sites associated with the imperial administration, such as the great tambos of the Qhapaq Ñan and the provincial capitals.

The quipu was embedded in a broader set of cultural practices for storing and transmitting information. The scholar Tom Zuidema has argued that the quipu system is best understood as one element of a “semiotic complex” that also included the ceque system of Cusco, the ritual calendar, and the quipuy (the act of remembering) tradition.

How the Quipu Works

A quipu consists of a primary cord, often about 50 centimeters to 1 meter in length, from which hang pendant cords. The pendant cords may themselves carry subsidiary cords, which in turn may carry further subsidiaries, producing a hierarchical structure. The pendant cords are marked with knots tied at specific positions along their length. The simplest type of knot is a single overhand knot, but the Inca used a variety of knot types: the single knot (the simplest), the long knot (a single knot with one or more turns of the cord above the knot, the number of turns indicating the value of the knot), and the figure-eight knot (a complex knot used for special purposes). The position of the knot on the cord, the type of knot, and the color of the cord all encode information.

The numerical system of the quipu is a positional decimal system. The pendant cord is divided into a series of levels, with the lowest value at the bottom of the cord and the highest at the top. The position closest to the primary cord represents the highest decimal place, and each lower position represents a power of ten less. The system allows the recording of numbers of any size, although in practice the largest numbers recorded on surviving quipus are in the hundreds of thousands. The system is, in principle, a positional decimal system, but it differs from modern positional systems in several ways: the orientation of the knots may encode sign or other categorical information, and the colors of the cords and the structure of the subsidiaries encode additional categories.

The operation of the quipu was a complex technical skill. The quipucamayocs — the keepers of the quipus — were a distinct class of officials who served at the court of the Sapa Inca, in the regional capitals, and in the provinces. They were trained from youth in the interpretation of quipus and were responsible for recording the affairs of the empire. Garcilaso de la Vega, in his Comentarios Reales, wrote that the quipucamayocs were organized in a hierarchy parallel to the administrative hierarchy: there were quipucamayocs for each household unit, each district, each province, and the empire as a whole, and each level reported to the next. The chronicler also reported that the quipucamayocs used a system of color coding that assigned specific colors to specific categories — gold, silver, maize, wool, soldiers, llamas — and that the cords of the quipu were arranged in a specific order depending on the type of record.

The materials of the quipu were as varied as the uses. The primary cord and the pendant cords were typically made of cotton, llama, alpaca, or vicuña wool, dyed in a range of colors. Cotton quipus were used for general administrative records, while llama, alpaca, and vicuña quipus were reserved for special purposes, particularly the recording of llama herds and the production of state cloth. The most elaborate quipus — those associated with the imperial administration and with the cult of the ancestors — were made of fine alpaca or vicuña wool, with carefully chosen colors and elaborate subsidiary structures. The variety of materials and the sophistication of the construction suggest that the quipu was a highly developed technology, comparable in complexity to the writing systems of the Old World.

What Quipus Recorded

The Inca used quipus to record a remarkable range of information. The most common use was the recording of numerical data: census counts, tribute lists, llama herd inventories, storehouse contents, military rosters, and the output of state mines and agricultural estates. The Spanish chroniclers described the use of quipus for many of these purposes: Pedro Cieza de León recorded that the curacas used quipus to keep track of tribute obligations, and Bernabé Cobo reported that the state granaries (qollqa) along the Qhapaq Ñan were inventoried on quipus. The quipus recorded the quantities of maize, potatoes, quinoa, dried meat, cloth, and other goods stored at each tambo, and the records were updated regularly by the quipucamayocs.

The quipus also recorded categorical information, narrative information, and historical information. The color of the cords, the direction of the knot twists, and the arrangement of the subsidiary cords encoded the type of record, the subject, the source, and the date. The chronicler Felipe Guaman Poma de Ayala included in his Nueva Corónica y Buen Gobierno (c. 1615) a detailed discussion of the various kinds of quipus and their uses, including quipus for population counts, llama herds, maize harvests, storehouse inventories, the genealogy of the Inca, the history of the empire, and the rituals of the cult of the sun. The scholar Sabine Hyland, in her 2017 study of the San Juan de Collata quipu, has argued that the quipus recorded not only numerical data but also narratives, including the names of people and places, and that the system was a form of writing in a broad sense.

The use of quipus extended across the empire. Surviving quipus have been found from northern Ecuador to northern Chile, and the system was used by both Inca and non-Inca populations to record transactions, contracts, and obligations. The quipus of the Lupaca, a Colla people of the Lake Titicaca basin, were sufficiently elaborate that the Spanish chroniclers described them in detail, and the Lupaca quipucamayocs continued to use the system for local administration well into the colonial period. The quipus of the Kuna, a Chibchan-speaking people of Panama and northern Colombia, are unrelated to the Inca quipus but show similar characteristics, and they have been used as a comparative case for the study of the Inca system.

The Quipucamayocs and the Imperial Administration

The quipucamayocs — the keepers of the quipus — were a distinct class of officials who served at the court of the Sapa Inca, in the regional capitals, and in the provinces. They were trained from youth in the interpretation of quipus and were responsible for recording the affairs of the empire. The chronicler Garcilaso de la Vega described the quipucamayocs as a hereditary class of officials who were organized in a hierarchy parallel to the administrative hierarchy: there were quipucamayocs for each household unit, each district, each province, and the empire as a whole, and each level reported to the next. The chronicler Martín de Murúa included a detailed description of the training of the quipucamayocs, who were taught from a young age to read, interpret, and tie quipus.

The quipucamayocs were an essential instrument of the Inca administration. They recorded the tribute lists of each province, the output of state lands, the size of llama herds, the contents of state granaries, the recruitment of soldiers, and the genealogies of the Inca and the local nobility. The quipucamayocs were also responsible for the rebinding of the quipus at the death of a Sapa Inca or the appointment of a new governor: the chronicler Bernabé Cobo described the rebinding as a major administrative and ceremonial act, in which the quipucamayocs gathered the quipus of the previous administration, prepared new cords, and tied them into new quipus in the presence of the new ruler and the high officials. The scholar Gary Urton, in his influential 2003 study Signs of the Inka Khipu, has argued that the quipus recorded information in seven different ways: the knot type, the knot position, the cord color, the cord twist direction, the cord type (cotton vs. animal fiber), the subsidiary cord structure, and the order of the pendant cords. According to Urton, the quipus were a “binary” or “binomial” system in the sense that each piece of information could be encoded in one of two possible values, and the system was thus capable of recording a vast range of categorical and numerical information.

Decipherment and Modern Research

The decipherment of the quipus has been one of the most active areas of Inca studies since the late nineteenth century. The North American scholar Leland Locke, in a series of articles in the 1920s, demonstrated that the quipus used a positional decimal system and laid the foundation for the modern study of the numerical structure of the quipus. The modern era of quipu studies began in the 1970s and 1980s, with the pioneering work of John V. Murra and the recovery of a small number of quipus from archaeologically and historically datable contexts in the central and north-central Andes.

The most famous of these are the quipus of San Juan de Collata, found in a burial context in the highlands east of Lima in 1982 by the anthropologist Frank Salomon, and the quipus of Laguna de los Cóndores, found in a sealed burial chamber in the cloud forest of the Chachapoyas region in 1997 by a team led by Federico Kauffmann Doig. The San Juan de Collata quipu has been associated with a colonial-era document that records a dispute over land and water rights, and the comparison of the quipu with the Spanish record has provided direct evidence of the kinds of information the quipus recorded. The Laguna de los Cóndores quipus, which are exceptionally well preserved, include subsidiary structures not found in other quipus, and their study has expanded the known range of the system.

The field has been transformed in recent years by the application of digital imaging and computational analysis. The Khipu Database Project at Harvard University, led by Gary Urton, has compiled a digital database of all known quipus, and the project has made it possible to compare quipus from different regions and contexts. The most ambitious recent claim, advanced by Sabine Hyland in her 2017 study of the San Juan de Collata quipu, is that the quipus recorded not only numerical and categorical data but also narratives, including the names of people and places, and that the system was a form of writing in a broad sense. The claim is supported by the comparison of the San Juan de Collata quipu with a Spanish document that records a dispute over land and water rights, and the analysis of the quipu shows that the structure of the pendant cords and the colors of the cords encode information that corresponds to the names of individuals, places, and transactions in the Spanish record. The full extent of what the quipus recorded is still being established, and the decipherment of the quipus remains an active area of research.

Decline and Survival

The quipu system was suppressed by the Spanish in the colonial period, but it did not disappear immediately. The colonial records contain many references to the use of quipus for local administration, and the first Viceroy of Peru, Blasco Núñez Vela, attempted to suppress the quipus in 1544. The Synod of Lima in 1552 and the Council of Lima in 1567–1568 prohibited the use of quipus in the recording of Christian rituals, marriages, and other sacramental matters. The use of quipus for secular administration continued for a longer period, but it gradually declined as the Spanish introduced written records and as the indigenous population was assimilated to the colonial administration.

The quipu has experienced a revival in the modern Andean context, both as a subject of academic study and as a cultural symbol. The Peruvian anthropologist Luis E. Valcárcel, one of the founders of indigenismo in Peru, included the quipu in his account of Inca civilization, and the quipu has been adopted as a symbol of indigenous Andean identity by political and cultural movements. The quipu is also the subject of a growing body of contemporary Andean art, in which artists use the forms and motifs of the quipu to explore themes of memory, identity, and resistance. The quipu has thus come to occupy a place in the cultural memory of the Andes comparable to the place of the Rosetta Stone in the cultural memory of Egypt: a symbol of a recording system that, if it is ever fully read, could open up a world of the past.