The Cord Keepers: Khipus and Cultural Life in a Peruvian Village / Edition 1

by Frank L. Salomon

Overview

None of the world’s “lost writings” have proven more perplexing than the mysterious script in which the Inka Empire kept its records. Ancient Andean peoples encoded knowledge in knotted cords of cotton or wool called khipus. In The Cord Keepers, the distinguished anthropologist Frank Salomon breaks new ground with a close ethnography of one Andean village where villagers, surprisingly, have conserved a set of these enigmatic cords to the present day. The “quipocamayos,” as the villagers call them, form a sacred patrimony. Keying his reading to the internal life of the ancient kin groups that own the khipus, Salomon suggests that the multicolored cords, with their knots and lavishly woven ornaments, did not mimic speech as most systems of writing do, but instead were anchored in nonverbal codes. The Cord Keepers makes a compelling argument for a close intrinsic link between rituals and visual-sign systems. It indicates that, while Andean graphic representation may differ radically from familiar ideas of writing, it may not lie beyond the reach of scholarly interpretation.

In 1994, Salomon witnessed the use of khipus as civic regalia on the heights of Tupicocha, in Peru’s central Huarochirí region. By observing the rich ritual surrounding them, studying the village’s written records from past centuries, and analyzing the khipus themselves, Salomon opens a fresh chapter in the quest for khipu decipherment. He draws on a decade’s field research, early colonial records, and radiocarbon and fiber analysis. Challenging the prevailing idea that the use of khipus ended under early Spanish colonial rule, Salomon reveals that these beautiful objects served, apparently as late as the early twentieth century, to document households’ contribution to their kin groups and these kin groups’ contribution to their village. The Cord Keepers is a major contribution to Andean history and, more broadly, to understandings of writing and literacy.

Product Details

ISBN-13:	9780822333906
Publisher:	Duke University Press Books
Publication date:	10/29/2004
Series:	Latin America Otherwise
Edition description:	New Edition
Pages:	368
Product dimensions:	6.12(w) x 9.25(h) x (d)

About the Author

Frank Salomon is Professor of Anthropology at the University of Wisconsin, Madison. He is the author of Native Lords of Quito in the Age of the Incas and coauthor of The Huarochirí Manuscript: A Testament of Ancient and Colonial Andean Religion. He is a coeditor of the two South American volumes of The Cambridge History of the Native Peoples of the Americas: South America (“Prehistory and Conquest” and “Colony and Republics”).

Read an Excerpt

The Cord Keepers

Khipus and Cultural Life in a Peruvian Village

---

By Frank Salomon

Duke University Press

ISBN: 0-8223-3390-2

---

Chapter One

UNIVERSES OF THE LEGIBLE AND THEORIES OF WRITING

From early contact to recent times, South American indigenous groups learning the alphabet have often denied that writing is new to them. They say they "already" had it, because they could "already read" various things such as petroglyphs or spoors (Gow 1990; Hugh-Jones 1989:65-68; Perrin 1986; Platt 1992:143). Yet Andean mythology also abounds in motifs to the effect that letters were deeply puzzling when first imported. This polarization suggests that many Andean persons perceived European writing as a member of some larger familiar class of legible signs, but a strikingly deviant member. Certainly the few examples of South American inscription systems about which we have ethnographic evidence-grafismo indigena, as Lux Vidal (1992) classes them-do differ drastically from writing "proper" in their principles of legibility (Arnold 1997; Langdon 1992).

Where do South American models of legibility and legible objects fit in relation to grammatology as usually defined? Do nonphilological theories of inscribed meaning yield promising heuristic avenues for interrogating inscriptions where mimesis of speech is far from being an assured property?

"A Certain Kind of Writing": Grammatology and the Problem of "Writing without Words"

Byway of estimating the distance between indigenous South American notions of legibility and the extant corpus of theory and method about writing, we might give ear to an obscure man made famous by testimonial literature, a cargo porter in Cuzco named Gregorio Condori Mamani. In the early 1970s he tried to explain, via a legend, that Andean and European ways of recording information were maximally different:

The Inkas didn't know anything about paper or writing, and when the good Lord wanted to give them paper, they refused it. That's because they didn't get their news by paper but by small, thick threads made of vicuna wool; they used black wool cords for bad news and for the good news, white cords. These cords were like books, but the Spaniards didn't want them around; so they gave the Inka a piece of paper. "This paper talks," they said. "Where is it talking? That's silly; you're trying to trick me." And he flung the paper to the ground. The Inka didn't know anything about writing. And how could the paper talk if he didn't know how to read? And so they had our Inka killed. (Valderrama Fernandez and Escalante Gutierrez 1977:50. Translation from Valderrama Fernandez, Escalante Gutierrez, Gelles, and Martinez Escobar 1996:57)

This legend was not just Don Gregorio's. It has been collected and discussed over and over for 430 years. It seems among the most indelible of all Quechua-language traditions. Literally foundational to the Andean-European dialogue, it comments at the same time on its failures. It occurs in the oldest known book of Andean authorship, Titu Cusi's 1570 brief for the intransigent Inkas who had been defending a jungle redoubt since 1535 (Titu Cusi Yupanqui 1973:15-16) . Over 400 years later, it still occupied a central place in Andean folk history (Ortiz Rescaniere 1973:134-41, 146-49). In each version the rejection of writing (with or without provocation by the Spanish) triggers the first act of the colonial tragedy.

The story is no less a canonical theme in academic history (MacCormack 1988; Wachtel 1977:39). In some cases, it has become a charter myth for misguided treatments of Andean culture as being in an essential way "oral" or even antigraphic. One such article bears the title "Literacy as Anticulture" (Classen 1991). Another (Seed 1991) bears the astonishing subtitle "Atahualpa's Encounter with the Word," as if absence of alphabets meant absence of language! This misperception underwrites an attitude on the part of educators that Peru's rural literacy deficit results not from socio linguistic difficulties, but from some fundamental dissonance between Andean culture and literacy.

And yet the myth has not been fully interpreted. European and Euro-American readers seem obsessed with Quechua insistence that Andeans found the alphabet radically strange. They are understandably fascinated by the world-historical importance tellers attach to this fact. But they rarely ask why the book, of all things, stands for insoluble cultural difference.

Condori was telling us precisely that Inka society was a graphic society. Since the Inka already possessed a system of visible "messages," a graphic artifact as such would not seem a fundamentally confusing artifact. By telling us first of all that Inkas used messages on cords-q'aytu, incidentally a word Tupicochans likewise apply to khipus-Don Gregorio tells us that the problem concerns kinds of graphic system, not the presence or absence of them. His story is about a comparison, not a singularity.

The reason for the Inka's confusion is mentioned, but unanalyzed: "paper talks" (papelmi riman). Rimay means speaking in words, specifically and exclusively (Lira 1982 [1941]:251). Although Condori says cords are like a book, he does not apply rimay to cords. He implies, then, that European paper used words, but in a way that did not involve sounds. In other words, the Inka was putatively puzzled not over the concept of legibility, but over the assertion that paper was legible by virtue of its being rimay.

To us, nothing could be more routine than the assertion that writing is a secondary code for a primary, oral-aural code, namely, speech. After all, thinkers from Aristotle to the modern heirs of Ferdinand de Saussure and Leonard Bloomfield have been insisting that this is its essence (Olson 1994:3). But the Aristotelian-through- Saussurean viewpoint is far from cross-culturally obvious.

Consider, then, what putatively stopped an Inka well versed in Peru's graphic code short of recognizing the alphabet as an alternate graphic code-and why this difference appeared as a fateful gulf between cultures. Writing "proper" operates not as a code whose signs correspond to objects in the world, but as a code whose signs refer to the sounds which a given spoken language uses as parts of speech in naming such objects. Whatever the fineness or coarseness of divisions at which written signs encode speech (real cases range from minute articulation features in Korean to whole words in Chinese, with alphabets occupying a middle ground), the mechanism is basically the same: visible signs stand for sound-segments of the speech flow, and the morphemes (e.g., words, suffixes) that make up the flow in turn stand for "things."

Perhaps the Inka was not unfamiliar with visual code, but unfamiliar with the notion that it stands for the sounds of "talk" as opposed to the referents of "talk." Commentators have emphasized the negative-what the Inka failed to understand- without asking the more interesting question of what his story did presume about legibility. This book sets out to consider the positive: what legibility did Andean culture posit? If not talk, what?

Let us restate the point in terms of theory. What the Inka was unfamiliar with was "writing with words," which Barry Powell (2002:64) terms lexigraphy. (Lexigraphy is a broad term including all signaries that correspond to segments of speech, whether large ones, such as whole morphemes, or small ones, such as phonemes.) His expectations were apparently based on some other principle. Just to begin somewhere, we could consider the most broadly conceived alternative: what Geoffrey Sampson, following Gelb's 1952 neologism, calls "semasiography." Semasiography is applied to the generally ill-theorized area of "mnemotechnologies," "pictography," "notations," and "tokens" (Sampson 1985:26-45).

Semasiographs purportedly do not stand for the sounds of the name of a referent, but rather for the referent itself. Taken abstractly, they are not "in" any particular spoken language. Productive competence consists of knowing or inventing signs that correspond to objects within the presumed common semantic grid. The reader's competence is to recognize the correspondence and, optionally, verbalize it in any appropriate tongue. The "No left turn" traffic sign, consisting of a leftward-bent arrow in a slashed red circle is a semasiograph of two sematograms, one iconic (the shape of the turn) and the other perhaps arbitrary (the slashed red circular cartouche). Whether a police officer verbalizes it as "no left turn," or "no doblar a la izquierda," depends only on local translation habits. This detour via verbiage is the operation which the Inka apparently failed to guess.

Because in any pure semasiography, elements of speech need not be retrieved in order to grasp a message, some authors call such systems "writing without words" (Boone 1994). Grammatologists disagree about whether this is an acceptable extension of the word writing. William Bright's and Peter T. Daniels's superb compendium The World's Writing Systems does cover dominantly semasiographic musical, dance, and numerical scripts. But in his theoretical keynote essay, Daniels enshrines only what the counter grammatologist Jacques Derrida calls "a certain kind of writing" as the real graphic McCoy (1979 [1967]:83; see Goody 2000:109- 18). Daniels's definition demands "a system of more or less permanent marks used to represent an utterance in such a way that it can be recovered more or less exactly without the intervention of the utterer.... It is thus necessary for a writing system to represent the sounds of a language" (Daniels 1996b:3). In other words, for Daniels, "writing" must imply paper (etc.) that "talks." John De Francis, rebutting Sampson, argues for even more restrictive postures (1989:211-47). These, like most historians of scripts (for example, Coulmas 1989:18; Gaur 1984:22-23,77-79; Gelb 1952:4), regard khipus as irrelevant to writing.

Influenced by such theory, and by the unilineal evolutionism of its precursors (I. Taylor 1899 [1883]:1:18), encyclopedias and syntheses tend to treat semasiographies as "primitive" precursors of "true writing." Isaac Taylor's Victorian idea of a single continuum from signs for ideas, to signs for words or morphemes, to more analytical systems like the alphabet, has long outlasted other pre-Boasian unilinealist theories. Although practicing grammatologists no longer embrace it, the notion of "progress" from more aggregated to more analytical signaries still saturates virtually all Western popular literature. The Encyclopedia Britannica even retains (as of 2004) a special dungeon for nonlexigraphic systems under the heading of "subgraphemics." But before assuming that semasiography has something to do with simplicity, one should notice that music notations, chemical formulas, mathematical scripts, machine-readable waybills, and circuit diagrams are semasiographies.

Could khipus' reference system be a general-purpose semasiography, as opposed to functionally restricted modern semasiographs? Sampson imagines a limiting extreme: "There would appear in principle to be no reason why a society could not have expanded a semasiographic system by adding further graphic conventions, until it was fully as complex and rich in expressive potential as their spoken language. At that point they would possess two fully-fledged 'languages' having no relationship with one another-one of them a spoken language without a script, and the other a 'language' tied intrinsically to the visual medium" (Sampson 1985:30). In fact no such language has been found. Sampson suggests that such a system would be impossibly cumbersome because it would be unmanageably prolific of signs. The crux of the matter seems to be that semasiographs are superior where different users have a substantial domain of culture in common, but little spoken language in common (e.g., musicians), or where for other reasons the verbal detour is undesirable. (Such reasons include cases where the syntactic structure of speech obscures dissimilar logical structures, as when one obscures an algebraic equation by phrasing it as a sentence.) W. C. Brice, an expert on the ancient Cretan scripts, calls attention to the "self-sufficient function" such signs achieve:

Most studies of the history of writing relegate non-phonetic systems, under various names, to a brief preliminary chapter, which stresses their limitations, before passing on to early varieties of so-called "real" or "true" writing in which, through the familiar device of the rebus, the symbols are used to designate sounds rather than ideas. But in fact the only truly independent writing is that which is non-phonetic, and can express a meaning without reference to sound. Phonetic script is writing deprived of its self-sufficient function and used not to express an idea but to freeze a spoken sound. Both systems have their separate advantages. But if non-phonetic writing had been so clearly inferior to phonetic ... it would not have survived in use for special purposes. (1976:40)

Highly successful and persistent nonphonetic scripts such as math notation, music notation, chemical formulas, formalizations for linguistics, choreographic labanotation, and knitting and weaving codes, satisfy such purposes.

It would be pointless to deny that movement toward writing "with words" is a repetitive course of history. The association occurs in archaeological cases as diverse as Sumer, Egypt, the Maya lands, and ancient China. The system of glyphs associated with Aztec rule appears to have been acquiring some lexigraphic features when Spanish intervention diverted its history. It seems "real writing" in the Aristotelian-Saussurean sense did over and over again develop in sign systems that served complex institutions: states, priesthoods, and commerce. It has been suggested (see chapter 5) that Spanish invasion itself stimulated khipu development in that direction.

But the ethnographic evidence presented in chapters 6 through 8 does not con- firm any systemic speech-mimicking function. It does suggest a khipu system elaborately keyed to mimicking other forms of action. What little we know ethnographically of modern and recent khipu competences depends on semasiographic composition with a "readout" method superficially resembling logography. Without prejudice against Urton's argument that khipus are theoretically capable of lexigraphic reference such as logosyllabography, I have found that actual cord lexigraphy remains conjectural. So it seems to me we should still give a high priority to asking how khipus could have served complex data registry independent of lexigraphy, with or without supplementation by it.

Could it be that Peruvian practice came closer than our own to that limiting case Sampson asks us to ponder theoretically-the extreme at which visible code and speech become "two fully-fledged 'languages'"? If one considers the phylogenesis of graphic practice in its pristine instances, thinking in a forward or historical direction rather than assuming a teleology toward "real writing," one quickly runs into considerations that bring real writing and its alternatives much closer together.

The first is that when people set about to inscribe, they do not a priori have the goal of recording speech acts. It would be silly to posit a pair of puzzled Neolithic ancestors asking, "How can we write down what we've said?" because the concept of representing evanescent, unself-conscious speech was itself unavailable. Indeed the psychologist David Olson (1994:105) persuasively argues that "writing introduced a new awareness of linguistic structures," only after it had arisen from nonlexigraphic processes. Even Florian Coulmas, a strong exclusivist, concedes that "the invention of writing ... was not a clearly defined problem.... Rather there were a number of practical problems (such as record keeping [by priests, who would be held accountable by fellow priests], counting, conveying messages indirectly, etc.).... The establishment of convention is a kind of social problem solving and that is what the invention of writing amounts to" (1989:9). Coulmas may have been persuaded by Konrad Ehlich's (1983) formalization of such problem solving in semiotic terms.

The useful perception here is that the development of a writing system is nothing other than the practical case-by-case solution of social tasks, which produce as an emergent a new data registry system. An interesting research venture in the direction of studying legibility as problem solving is that of Axel Steensberg (1989), a Scandinavian archaeologist who associates complex signaries with problems of "labeling" in certain food technologies. Steensberg, like the Aschers, reminds us that the record-keeping art takes shape around the social problems it solves.

(Continues...)

---

Excerpted from The Cord Keepers by Frank Salomon Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---

Table of Contents

List of Illustrations and Maps ix

List of Tables xiii

About the Series xv

Preface xvii

The Unread Legacy: An Introduction to Tupicocha’s Khipu Problem, and Anthropology’s 3

1. Universes of the Legible and Theories of Writing 23

2. A Flowery Script: The Social and Documentary Order of Modern Tupicocha Village 41

3. Living by the “Book of the Thousand”: Community, Ayllu, and Customary Governance 55

4. The Tupicochan Staff Code 77

5. The Khipu Art after the Inkas 109

6. The Patrimonial Quipocamayos of Tupicocha 137

7. Ayllu Cords and Ayllu Books 185

8. The Half-Life and Afterlife of an Andean Medium: How Modern Villagers Interpret Quipocamayos 209

9. Toward Synthetic Interpretation 237

Conclusions 267

Notes 283

Glossary 295

References 299

Index 317

Editorial Reviews

The Cord Keepers is a brilliant and pathbreaking book. It forces us to reconceptualize what writing is or can be, what it encodes, and whether we should even think of writing as something that records, rather than as a performative practice that engages more actively with the world.”—Joanne Rappaport, author of The Politics of Memory: Native Historical Interpretation in the Colombian Andes

“Strokes of good fortune brought Frank Salomon to villages and archives with extraordinary potential for research. Imagination, erudition, and perseverance permitted him to retrieve insights of great importance from these sources. The world seems subtly yet significantly different after reading The Cord Keepers, since Salomon convinces his readers that the human capacities to record meanings and to transmit cultural forms are deeper and broader than they had thought.”—Benjamin Orlove, author of Lines in the Water: Nature and Culture at Lake Titicaca

“This is without doubt the most important work ever produced on the khipu. It is extraordinarily bold and inventive in its suggestions for how these ‘texts’ may have been produced and interpreted. The hypothetical reading of the khipu is an absolutely astonishing accomplishment.”—Gary Urton, author of Signs of the Inka Khipu: Binary Coding in the Andean Knotted-String Records

From the Publisher