Back to top

Narrative Threads

Narrative Threads
Accounting and Recounting in Andean Khipu

In this benchmark book, twelve international scholars tackle the most vexed question in khipu studies: how did the Inkas record and transmit narrative records by means of knotted strings?

Series: Joe R. and Teresa Lozano Long Endowment in Latin American and Latino Art and Culture

August 2002
This is a print-on-demand title. Expedited shipping is not available.
391 pages | 6 x 9 | 66 b&w figures, 14 tables, 14 color illus. (cloth edition only) |

The Inka Empire stretched over much of the length and breadth of the South American Andes, encompassed elaborately planned cities linked by a complex network of roads and messengers, and created astonishing works of architecture and artistry and a compelling mythology—all without the aid of a graphic writing system. Instead, the Inkas' records consisted of devices made of knotted and dyed strings—called khipu—on which they recorded information pertaining to the organization and history of their empire. Despite more than a century of research on these remarkable devices, the khipu remain largely undeciphered.

In this benchmark book, twelve international scholars tackle the most vexed question in khipu studies: how did the Inkas record and transmit narrative records by means of knotted strings? The authors approach the problem from a variety of angles. Several essays mine Spanish colonial sources for details about the kinds of narrative encoded in the khipu. Others look at the uses to which khipu were put before and after the Conquest, as well as their current use in some contemporary Andean communities. Still others analyze the formal characteristics of khipu and seek to explain how they encode various kinds of numerical and narrative data.

  • Acknowledgments
  • Preface (Jeffrey Quilter
  • Part One. Background for the Study of Khipu and Quechua Narratives
    • 1. An Overview of Spanish Colonial Commentary on Andean Knotted-String Records (Gary Urton)
    • 2. Spinning a Yarn: Landscape, Memory, and Discourse Structure in Quechua Narratives (Rosaleen Howard)
  • Part Two. Structure and Information in the Khipu
    • 3. A Khipu Information String Theory (William J Conklin)
    • 4. Reading Khipu: Labels, Structure, and Format (Marcia Ascher)
    • 5. Inka Writing (Robert Ascher)
  • Part Three. Interpreting Chroniclers' Accounts of Khipu
    • 6. String Registries: Native Accounting and Memory According to the Colonial Sources (Carlos Sempat Assadourian)
    • 7. Woven Words: The Royal Khipu of Blas Valera (Sabine P. Hyland)
    • 8. Recording Signs in Narrative-Accounting Khipu (Gary Urton)
    • 9. Yncap Cimin Quipococ's Knots (Jeffrey Quilter)
  • Part Four. Colonial Uses and Transformations of the Khipu
    • 10. "Without Deceit or Lies": Variable Chinu Readings during a Sixteenth-Century Tribute-Restitution Trial (Tristan Platt)
    • 11. Pérez Bocanegra's Ritual formulario: Khipu Knots and Confession (Regina Harrison)
  • Part Five. Contemporary Khipu Traditions
    • 12. Patrimonial Khipu in a Modern Peruvian Village: An Introduction to the "Quipocamayos" of Tupicocha, Huarochirí (Frank Salomon)
    • 13. The Continuing Khipu Traditions: Principles and Practices (Carol Mackey)
  • Contributors
  • Index

Jeffrey Quilter is Director of Pre-Columbian Studies at Dumbarton Oaks in Washington, D.C. Gary Urton is Charles A. Dana Professor of Anthropology at Colgate University, as well as a MacArthur Fellow (2001-2005).


Browse the book with Google Preview »

In 1922, the founder of modern Mesoamerican iconographic research, Eduard Seler, died in Berlin. Although a number of important advances in the study of Maya hieroglyphic writing had already been accomplished, the field was still in its infancy. A year later, in 1923, L. Leland Locke published The Ancient Quipu or Peruvian Knot Record, in which he presented the basic understanding of khipu maintained to the present. Despite an early precociousness, however, the study of khipu has lagged far behind the decipherment of Maya glyphs. This seems all the more remarkable because Mayanists were handicapped for decades by Sir J. Eric Thompson, who, though he was only just entering Cambridge University in 1922, later actively discouraged acceptance of the glyphs as true writing. Though Locke did not have as much influence over Andean studies as a whole as did Thompson over Mayanist studies, he was insistent that khipu only recorded numbers. A principal goal of this book is to question that dictum.

Given that Maya hieroglyphs and Andean khipu are two of the very few elaborated record-keeping systems of the ancient New World, why have investigations in one advanced so rapidly in comparison to the other? The answer to this question is not straightforward. One simple factor is that there are far more Mesoamericanists than Andeanists. The countries of Middle America have been easier to reach from Europe and the United States than those of the distant Andes. Thus, more and a greater variety of scholars from afar have traveled to the lands of the Maya, creating a fairly large international scholarly community.

It may also be argued that initial conditions for scholarly research have kept Mesoamerican and Andean studies following early established directions. Although Seler had wide-ranging interests in all of the Americas, he specialized in Mesoamerica. His legacy, as well as those of other Mesoamericanist scholarly pioneers, engendered subsequent generations of students who continued working in the same areas and on the same problems as their predecessors while training the next generation of scholars, who did the same.

Another factor that has helped to encourage Maya studies is the attraction Maya art and architecture have had for Westerners. Even though we now know that much of the tropical forest was cut down to support Maya centers, the discovery of ruins "lost" within dark jungles appealed to nineteenth-century romantic sensibilities. When the forest was cut back, ruins exposed, and tombs opened, Maya art was found to be mysterious but not entirely unapproachable. Here were depicted human forms bedecked in elaborate costumes performing bizarre rituals, or monkey-headed figures swaggering with strange accouterments. Although debate raged as to whether these were gods or kings, the art was original, attractive, and recognizable as representational to a greater or lesser degree.

Peering at the remains of this strange ancient world, early explorers found glyphs everywhere. They were carved on buildings, painted on ceramics, incised into jades, bone, and shell. They wrapped themselves around lintels and doorways. They covered the backs of stelae and the fronts of large panels and were delicately painted in the few Maya books preserved from destruction by the Spaniards. The sheer quantity of hieroglyphs used in Maya elite life and the way texts were inextricably associated with art made them impossible to ignore. And, not least of all, the glyphs were beautiful. The labor and skill it took to carve them in limestone or jade and the ways in which the symbols themselves were ornamented or varied to produce an aesthetic statement of their own, in addition to what they may have said, left no doubt that the Maya considered these signs as important and valuable.

Maya hieroglyphics were exotic and strange, but though arguments were made as to what these signs said, it was rarely doubted that they said something. Sir Eric Thompson slowed the course of the study of Maya writing because he believed that the glyphs could not be deciphered and that they only offered information on calendrics and astronomy. They could not be deciphered, he believed, because the system on which they were based was not a logical one and the glyphs were not constructed phonetically. Information on calendrics and astronomy was of interest, but, in Thompson's opinion, studying these arcane topics would not advance the understanding of larger issues of Maya politics, economy, and other matters that could only be revealed through field archaeology.

In the Andes, the keepers of the khipu, the Inka, have not appealed to the romantic sensibilities of Westerners as have the Maya. Scholars of the Inka have too long been caught up in debates as to whether the Inka Empire was the ideal socialist state or the worst example of fascism, and in the process, the Inka have come across as a people as cold, aloof, and abstract as their chilly mountain home. If the Maya have been cast as the Greeks of the New World, delighting in the pleasures of the senses, the Inka have been given the role of the dull, stoical Romans, building a huge empire but not having much fun while doing so. These images, of course, are the conceits of Western minds, but they have contributed much to how the fields of study of these ancient peoples have fared.

There are many differences between Maya hieroglyphs and Inka khipu. Although glyphs could be carved as three-dimensional statues, they generally were inscribed in two dimensions, on flat surfaces, much as in any other writing system. Khipu, however, are expressions of linearity, made up of cords, which are at the same time three-dimensional objects. Hieroglyphs are graphic and have a strong representational component, while khipu are material and kinesthetic: the medium of the khipu is bound up with the message in a way that hieroglyphs are not. Although there are variations in how glyphs are rendered, ultimately, it is the formal characteristics of the inscribed signs that convey meaning. For khipu, knots and their placements, the twists of cords, color combinations, and likely other elements all seem to play important roles in conveying information.

Perhaps as important as the points already made, if Maya glyphs can be appreciated as art as well as a medium for communication, khipu seem to be merely an extension of folk craft. Knotted strings, perhaps too reminiscent of macramé wall hangings, do not appeal to Western artistic tastes as do the brush strokes of a Maya scribe or limestone blocks carved as if they were butter. This point could be a departure for discussion of the whole issue of the Western distinction between art and craft, but it is more worthwhile to emphasize that khipu express their own form of beauty once one is familiar with them and that such beauty is in their tactility—an aesthetic realm severely underappreciated in the visually oriented West, where "Do not touch" signs are all too prevalent. The aesthetic sensibilities of Andean peoples are still waiting to be adequately discussed in Western literature. An assumption that khipu were merely utilitarian devices—like tying a string around a finger so as not to forget to feed the neighbor's cat—may be partly to blame for lack of interest in them. That khipu were more than simple reminders, that they were sophisticated and complex systems, is another important message of this book.

These various issues have led to different perspectives on approaching khipu. Many scholars presumed early on that the Maya were conveying meanings with their hieroglyphs, and khipu scholars knew that the Inka kept accounts with knotted strings. But did they do more than record numbers with khipu? As both Robert Ascher and Marcia Ascher point out in their chapters in this book, numbers may be interpreted as magnitudes or quantities, but they can also be interpreted as labels, and these labels may have narrative properties and functions. Thus, numbers signified by knots, along with knot directionality, the colors of cords, and other elements that made up khipu, may all have been used to convey narrative information.

It is interesting that despite differences in the nature of Maya hieroglyphs and Inka khipu, issues of understanding how these systems worked have revolved around similar problems of interpretation. These interpretive problems have not been due primarily to the inherent ways khipu and glyph systems were constructed but rather to the ways in which they were explained to Spanish investigators in the sixteenth century and to subsequent assumptions about and interpretations of what the Spanish said concerning native recording systems.

Diego de Landa, bishop of Yucatán, interviewed Maya scribes and had a Spanish alphabet written with what he thought were the Maya equivalents of Spanish letters. In doing this, he created a document that has been essential in deciphering Maya script but has also produced confusion and debate. Landa made the assumption that Maya writing was alphabetic, but it is not. When Landa or an assistant spoke the sound for a letter of the Spanish alphabet, the Maya scribes heard what they took to be a word or sound in their own language. They then wrote down the glyph for the word or sound, not a letter. This issue of the relation of signs to sounds was a chief problem in accepting the Landa syllabary. Once what had happened in Landa's scriptorium was clarified, the Landa document became an important tool for understanding how to read Maya hieroglyphs.

We have had no single authoritative colonial record of an extended investigation of khipu but rather a number of colonial authors who discuss khipu at greater or lesser lengths. Much of what they say is unclear or contradictory, yet there are strong indications that khipu did not simply record numbers but also kept records of poems, histories, and other narratives. The earliest breakthrough in understanding Maya writing was the identification of number notation, but scholars now can read histories and other texts. For khipu, though, there has been no significant advancement beyond the ability to read numbers. In this book, however, Carlos Sempat Assadourian provides a careful discussion of the remarks by the colonial author Antonio Calancha on how to make a khipu that tells a story. The degree to which we can rely upon Calancha's account, however, is uncertain, and we may be on the same kind of uncertain ground concerning his understanding of what Andean people were telling him, as well as the accommodations those people made to allow their system to be understandable to Calancha, as was the case with Landa and his Maya scribes.

Understanding khipu thus involves not only finding and examining colonial sources about how knotted-string records were made and used but also evaluating how we should interpret those accounts, when found. Are the apparently contradictory and unclear discussions of khipu due to lack of full understanding by colonial authors of how khipu worked, or do at least some of these different accounts express variations among khipu themselves? Added to the exploration of narrativity and the demonstration of complexity in khipu, the issue of standardization versus idiosyncrasy in khipu is the third theme of this book. Although writing systems can tolerate a certain degree of deviation from some standard, such as variant spellings of words in British as opposed to American English, there must be general intercommunicability in a writing system, as was the case with Maya glyphs. For khipu to approximate a writing system, then, they had to have been able to be read by more than just their makers.

Many of the important advances in the study of Maya hieroglyphs occurred at or through the agency of Dumbarton Oaks. In this tradition, it seemed appropriate to hold a meeting here on the issue of narrativity in Andean khipu. Gary Urton and I thus organized a meeting of leading scholars on the topic at Dumbarton Oaks in April 1997. Many of the chapters in this book are revised versions of papers given during that meeting, and others are new contributions.

The book opens with two chapters intended to provide the reader with background material. In the first chapter, Gary Urton reviews the history of khipu studies, beginning with accounts from the early Spanish chronicles and documents. Rosaleen Howard's following discussion of narrativity in contemporary Quechua stories underlines the critical role language and linguistic studies hold for decipherments. The study of Coptic was crucial for the eventual decipherment of Egyptian hieroglyphs, and Maya studies languished for many years because the earliest attempts at translation disregarded the study of contemporary Mayan languages and instead worked primarily on solving the riddle of the glyphs solely as symbolic representations. Thus, emphasizing language studies in the investigations of khipu may be critical for future advances in decipherment, although alternate approaches will undoubtedly have their own contributions to make, as well.

William Conklin discusses important technical issues on his way to developing a theory for the organization of information in knotted-string records. Additional theoretical perspectives are provided in the contributions of Marcia Ascher and Robert Ascher, with the former discussing numbers as both magnitudes and labels while the latter explores how encipherment and decipherment may have been performed on khipu without depending directly on language.

Examining common features and variability in how communication systems develop and change will be important in future work, as I discuss in my contribution. Gary Urton looks at such variability in khipu themselves in his second chapter, comparing Garcilaso de la Vega's commentaries in light of a class of known non-numerical khipu. Carlos Sempat Assadourian provides insight on another colonial commentator on khipu, Antonio Calancha, as noted above. A third colonial writer, Blas Valera, is discussed in Sabine Hyland's contribution.

Following the discussions based on colonial sources' descriptions of khipu is a series of chapters on khipu use in the same era: Tristan Platt comments on one of the few eyewitness accounts of khipu in use in a public setting, and Regina Harrison focuses on rosaries and khipu in the arena of competing yet analogous religious practices.

The chapters of the next section demonstrate that khipu are still vital parts of native Andean culture. The contributions by Frank Salomon and Carol Mackey reveal the rich sources of information available on the maintenance of special places for khipu themselves in the Andes and the continuing use of knotted-string records, respectively.

Gary Urton and I and the other authors represented in this collection hope that this book will not only establish a benchmark in khipu studies but also stimulate a wider interest in investigating these materials. Considering recent advances in mathematics, information theory, and other scholarship involved with computers, software, and communications technology, the study of khipu should appeal to many who are challenged and intrigued by puzzle solving. Perhaps this book will stimulate someone to be the Jean François Champollion of the khipu.

To call for a Champollion of the khipu implies that someone can "crack" the code of the knots in the same way that the great French Egyptologist made a breakthrough in understanding Egyptian hieroglyphics. But it remains uncertain whether we will ever be able to "read" a khipu the way we now can read an Egyptian or a Maya hieroglyphic text. Even in the case of Maya writing, its code was not so much broken—as a Gordian knot cleanly sliced through—as it was chipped away at: a little hole was gradually made bigger until, eventually, a critical mass was reached and the wall of ignorance came tumbling down.

Will we be able to unravel the mysteries of the khipu? We are at such an early stage of investigation—only beginning to tug at the knot—that it is hard to say. The archives are full of documents yet to be seen or studied, and the tradition of knotted-string records in the New World is hardly known. There is not yet even a thorough inventory and description of the known khipu in existence, a corpus of about five hundred items, which, if documented and made easily accessible for study, would serve as a fundamental reference work for all future khipu scholars. No code can be cracked, chipped at, or unraveled unless the material to be worked on is within easy reach of those who wish to solve the riddle.

Given that we barely understand even the limits of what we may know about this complex and unique system of record keeping, it is quite likely that our abilities to understand khipu at the end of the twenty-first century will be far more advanced than they are now, at its beginning. If this book inspires others to start tugging at the knots that secure the codes of Andean khipu, then it will have served scholarship very well indeed.



“A veritable encyclopedia of the khipu, this volume pulls together new and groundbreaking work by the foremost experts, attacking the problem from a wide variety of perspectives and integrating analysis from historical, archaeological, and ethnographic perspectives.”
Thomas A. Abercrombie, Associate Professor of Anthropology, New York University


Available for Kindle
Available on Google Play
Available on Kobo
Available for Nook
Available on the Apple Store

This book may also be available on the following library platforms; check with your local library:
3M Cloud Library/bibliotheca
UPCC/Project Muse