Recent Discoveries in Quipu Decipherment (Urton, Hyland)
The decipherment of the quipus — the knotted-string recording devices of the Inca Empire — has been one of the most important problems in Andean studies since the late nineteenth century. For most of the modern period, scholars treated the quipus as a primarily numerical system that recorded census data, tribute lists, and storehouse inventories, and the possibility of a deeper meaning in the quipus was largely ignored. The pioneering work of the Harvard anthropologist Gary Urton from the 1990s onward and the anthropological linguist Sabine Hyland from the early 2000s onward has transformed this picture. The combined archaeological, ethnographic, and linguistic evidence now suggests that the quipus encoded not only numerical information but also categorical, narrative, and possibly phonetic information, and that the system functioned as a form of semiotic recording whose full extent is still being established.
The Pre-Urton State of the Field
The early modern study of quipus was dominated by the question of whether the quipus were a form of writing. The Spanish chroniclers of the sixteenth and seventeenth centuries, including Pedro Cieza de León, Garcilaso de la Vega, and Felipe Guaman Poma de Ayala, recorded that the quipus recorded not only numbers but also names, places, and events, and they sometimes described the quipucamayocs (keepers of the quipus) as reciting historical narratives from the quipus. The nineteenth-century scholars who took up the question, including the German naturalist Alexander von Humboldt and the British archaeologist Clements Markham, were divided on the issue. By the 1960s the dominant view was that the quipus were a numerical system that recorded information in a way that was not comparable to European writing.
The discovery of a small number of quipus from the Peruvian central coast in the late twentieth century, particularly the Puruchuco quipus recovered by the archaeologist William Isbell in the 1980s, allowed scholars to begin to correlate the knot patterns of specific quipus with Spanish written records of the same events. The studies of the Puruchuco quipus by Isbell and later by the anthropologist Frank Salomon demonstrated that the quipus could record the same information as the Spanish documents, and they established the validity of the quipu system as a record-keeping technology.
Urton’s Categorical Encoding Hypothesis
The Harvard anthropologist Gary Urton, in a series of studies beginning in the 1990s, transformed the field by proposing that the quipus encoded not only numerical information but also a system of categorical information that allowed the quipus to record information that European writing could not easily capture. Urton’s hypothesis, developed in his 2003 book Signs of the Inka Khipu and elaborated in subsequent publications, was that the quipus used a system of binary coding in which the various attributes of the cords (color, direction of twist, direction of attachment, and so on) encoded a series of binary oppositions that, taken together, specified a unique identifier for the object or person being recorded.
The hypothesis was based on a statistical analysis of the surviving quipus, particularly a corpus of 21 quipus from the Puruchuco estate. Urton and his collaborator Carrie Brezine demonstrated that the distribution of cord attributes in the Puruchuco quipus was non-random, and that the attributes were organized in a hierarchical structure that resembled the structure of a written document. The analysis showed that the colors of the cords encoded categorical information (for example, the color red was associated with one set of records and yellow with another), and that the direction of the cord twist (S-twist vs. Z-twist) and the direction in which the subsidiary cords branched off the pendants encoded additional categories of information.
Urton’s work has led to the establishment of the Khipu Database Project at Harvard, a digital repository of information on all known quipus. The database has become an essential resource for quipu research.
Hyland’s Narrative Quipus
The anthropologist and linguistic scholar Sabine Hyland has developed a complementary line of research that has produced striking evidence for the existence of narrative quipus, that is, quipus that may have encoded not only numerical and categorical information but also stories, names, and possibly phonetic content. Hyland’s research has focused on the San Juan de Collata quipus, a set of quipus that were recovered in 1982 in a highland community near Cusco and that were, by Hyland’s account, in active ritual use in the late twentieth century.
The San Juan de Collata quipus, as described in Hyland’s 2017 book, are unusual in several respects. First, they include quipus with a small number of pendant cords (often fewer than ten), which is too few to record the kind of numerical data that the colonial quipus typically recorded. Second, the pendant cords of the San Juan de Collata quipus are arranged in a way that suggests a narrative structure, with the order of the pendants corresponding to the order of events in a story. Third, the quipus were used in the late twentieth century in the context of a community ritual that involved the recitation of a historical narrative about the origin of the community and the role of the kuraqkuna (lords) in its foundation.
Hyland’s interpretation is that the San Juan de Collata quipus encoded a record of the community’s history, including the names of the kuraqkuna and the dates of important events. The interpretation is based on a combination of ethnographic work (interviews with the quipucamayocs of the community), linguistic analysis (comparison of the recorded names with the historical records of the community), and the structural analysis of the quipus themselves. The work has been controversial, and the debate is ongoing.
Other Important Contributions
The anthropologist Frank Salomon of the University of Wisconsin has studied the colonial-era quipus of the northern highlands of Peru, demonstrating that the colonial quipus recorded information that is consistent with the colonial written records, and that the quipus were used as official documents in the colonial legal system.
The archaeologist Jeffrey Splitstoser of George Washington University has studied the Laguna de los Cóndores quipus, the largest collection of pre-Columbian quipus ever found in a single archaeological context. The Laguna de los Cóndores quipus, which were discovered in 1997 in a cliff tomb in the Chachapoyas region of northern Peru, are associated with a pre-Columbian burial and have been dated to the late fifteenth or early sixteenth century. The quipus are unusually elaborate, with hundreds of pendant cords, and they include features that suggest a complex semiotic system.
Other researchers, including the linguist Julio Calvo-Pérez at the University of San Marcos, have studied Ocucaje-style quipus from the Ica Valley of the south coast of Peru. Calvo-Pérez’s work has proposed that some of these quipus, like the San Juan de Collata quipus, may have encoded information in a more narrative form, though the interpretation remains contested.
Where the Field Stands
The work of the past few decades has substantially reshaped the study of the Inca quipus. Researchers have shown that, in addition to recording numerical data in a positional decimal system, the quipus encoded categorical information through color, twist direction, attachment direction, and subsidiary cord structure, and that some quipus may also have encoded names, places, and other narrative elements — though the latter claims remain contested. The full extent of what the quipus recorded is still being established, and the topic remains an active area of research rather than a closed problem. Future progress depends in part on the conservation of the surviving quipus and on the continued collaboration between Peruvian and international scholars, both archaeologists and linguists.