(Rosaleen Howard-Malverde) Creating Context in and

Creating Context in
Andean Cultures
Oxford Studies In Anthropological Linguistics
William Bright, General Editor
1. Classificatory Particles in Kilivila
GUNTER SENFT
2. Sounds Like Life: Sound-symbolic Grammar, Performance, and Cognition

in Pastaza Quechua
JANIS B. NUCKOLLS
3. Plastic Classes and Church Fathers: Semantic Extension from the Ethnoscience Tradition
DAVIOB. KRONENFEI.D
4. American Indian Languages: The Historical Linguistics of North America

LYLE CAMPBELL
5. Telling Our Selves: Ethnicity and Discourse in Southwestern Alaska
CHASE HENSEL
6. Creating Context in Andean Cultures
ROSALEEN HOWARD-MALVERDE (ED.)
7. Disorderly Discourse: Narrative, Conflict, and Inequality
CHARLES L. BRIGGS (ED.)
8. Understanding Cultures Through their Key Words: English, Russian, Polish,
German, and Japanese
ANNA WIERZBICKA
9. The Korowai of Irian jay a: Their Language in Its Cultural Context
GERRIT J. VAN ENKAND LOURENS DE VRIES
10. A Language of Our Own: The Genesis of Michif, the Mixed Cree-french Language of
the Canadian Metis
PETER BAKKER
11. Referring to Space: Studies in Austronesian and Papuan Languages

GUNTER SENFT
12. People, Countries, and the Rainbow Serpent: Systems of Classification Among
the Lardil of Mornington Island
DAVID MCKNIGHT
13. Vernacular Literacy Revisited
PENELOPE GARDNER-CHLOROS, ROBERT B. LE PAGE, ANDREE TABOURET-KELLERAND GABRIELLE VARRO
(EDS.)
CREATING CONTEXT IN
ANDEAN CULTURES
Edited by
Rosaieen Howard-Malverde
New York Oxford

Oxford University Press
1997
Oxford University Press
Oxford New York
Athens Auckland Bangkok Bogota Bombay Buenos Aires
Calcutta Cape Town Dar es Salaam Delhi Florence Hong Kong
Istanbul Karachi Kuala Lumpur Madras Madrid Melbourne
Mexico City Nairobi Paris Singapore Taipei Tokyo Toronto
and associated companies in
Berlin Ibadan
Copyright © 1997 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.
198 Madison Avenue, New York, New York 10016
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise,
without the prior permission of Oxford University Press
Chapter 4, "The Art of Ethnic Militancy: Theatre and Indigenous Consciousness in Colombia" by Joanne Rappaport is
reprinted from Cumbe Reborn: An Andean Ethnography of History, Joanne Rappaport, copyright © 1994 by permission
of the University of Chicago Press.
Chapter 11, "On the Margin: Letter Exchange Among Andean Non-Literates" by Sarah Lund is reprinted from
Exploring the Written: Anthropology and the Multiplicity of Writing, ed. E. Archetti, copyright © 1994 by permission
of the Scandinavian University Press, Oslo.
Creating context in Andean cultures / edited by Rosaleen Howard-Malverde.
p. cm. —- (Oxford studies in anthropological linguistics : 6)
Includes bibliographical references.
ISBN 0-19-509789-0; ISBN 0-19-510914-7 (pbk.)
1. Quechua language—Discourse analysis. 2. Quechua language
—Semantics. 3. Culture—Andes Region—Semiolic models.
4. Discourse analysis, Narrative—-Andes Region. 5. Anthropological
linguistics— Andes Region. 6. Semantics—Andes Region.
7. Criticism, Textual—Andes Region. I. Howard-Malverde, Rosaleen.
II. Series.
PM6303.C74 1997
498'.323—dc20 95-42967
1 3 5 7 9 8 6 4 2
Printed in the United States of America

on add-iree paper
In memory of
Eleudoro Gabriel Bias
This page intentionally left blank
PREFACE AND ACKNOWLEDGMENTS
Over the years, in the course of fieldwork in different parts of the Andes, many
Quechua-speaking storytellers have shared with me their understandings of the
complex social and historical setting within which their cultures have evolved.
Through the narration of oral tradition, as well as the other media of expression
discussed in this book, Andean peoples interpret and sometimes challenge the condi-
tions in which they live, creating culture anew with each generation of speakers. This
book is dedicated to the memory of one of those storytellers.
Ten of the papers in this volume were delivered to the conference on "Textual-
ity in Amerindian Cultures: Production, Reception, Strategies," convened by myself
and William Rowe and held in May 1991 under the joint auspices of the Centre
for Latin American Cultural Studies of King's College London, and the Institute
of Latin American Studies of the University of London. The paper by Willem
Adelaar was given at a workshop on "Language, Discourse and Society in the
Andes" held at the Institute of Latin American Studies of the University of London
in February 1990. The paper by Denise Arnold was presented at an anthropology
workshop on "Gender Relations, Work and Proprietorship among the Indigenous
Peoples of South America," convened by Joanna Overing, also at the Institute of
Latin American Studies, University of London, in December 1990.
The conference on "Textuality in Amerindian Cultures: Production, Reception,
Strategies" was funded by the British Academy and The Nuffield Foundation, in
addition to the host institutions; grateful thanks for their support are due to all
these bodies. Thanks also to the Institute of Latin American Studies of the
University of Liverpool, who met the translation costs of the papers by Lienhard
and Millones. I should like to acknowledge the contributions of those who took
part in the conference who are not represented in this volume, whose papers and
viii PREFACE AND ACKNOWLEDGMENTS
participation in the roundtable on Sunday morning were invaluable to the debate:

Hernan Aguilar, Margit Gutrriann, Silvia Hirsch, Billie Jean Isbell, Peter Masson,
Rodrigo Montoya, Joanna Overing, and Susanna Rostas, A special thanks to
William Rowe, who has been involved with this project from the outset; I am
grateful.to him; to William Bright, OSAL editor-in-chief; and to Rodolfo Malverde,
for their comments on a draft version of the introduction. Acknowledgments are
due to the editorial staff at OUP for their expert advice. Thanks also to my father,
Ben Howard, who, in late night conversations over the phone, helped me think
through the ideas involved in the terms textuality and context and made me see
the importance of having an active verb form in the title of the book. Tristan Platt
gave much appreciated encouragement, and his comments on the text/context
distinction were especially useful. Thanks most of all to the contributors, for their
patience and diligence in seeing their papers through to the final versions presented
here.
Liverpool R.H.-M.
March 1995
CONTENTS
Contributors xi
A Note on Orthography xiii
1 Introduction: Between Text and Context
in the Evocation of Culture ROSALEEN HOWARD-MALVERDE 3
I Textualising Histories and Identities in Cultural Performances
2 Peruvian Independence Day: Ritual, Memory,

and the Erasure of Narrative PENELOPE HARVEY 21
3 Saint Rose through the Eyes of Her Devotees:

From Flower of Lima to Mother of Carhuamayo LUIS MILLONES 45
4 The Art of Ethnic Militancy: Theatre and Indigenous

Consciousness in Colombia JOANNE RAPPAPORT 55
II Multiple Media: Ritual, Weaving, and the Making of Meanings
5 When Pebbles Move Mountains:

Iconicity and Symbolism in Quechua Ritual CATHERINE J. ALLEN 73
6 Cultural Transpositions:
Writing about Rites in the Llama Corral PENNY DRANSART 85
x CONTENTS
7 Making Men in Her Own Image:

Gender, Text, and Textile in Qaqachaka DEMISE ARNOLD 99
III Inside the Text: Processes of Textualisation in Written Quechua Discourse
8 Spatial Reference and Speaker Orientation in

Early Colonial Quechua WILLEM F. H. ADELAAR 135
9 Point of View and Evidentiality in the Huarochiri

Texts (Peru, 17th Century) SABINE DEDENBACH-SALAZAR SAENZ 149
IV Orality and Literacies as Reflexes of Colonisation and Resistance
10 Writing from Within: Indigenous Epistolary Practices

in the Colonial Period MARTIN L1ENHARD 171
11 On the Margin: Letter Exchange among

Andean Non-Literates SARAH LUND 185
12 The Sound of Light: Emergent Communication

through Quechua Shamanic Dialogue TRISTAN PLATT 196
Glossary 227
Bibliography 231
Index 247
CONTRIBUTORS
Willem F. H. Adelaar Rosaleen Howard-Malverde

Department of Comparative Linguis- Institute of Latin American Studies,
tics, University of Leiden University of Liverpool
Catherine J. Allen Martin Lienhard
Department of Anthropology, George Romanisches Seminar, University of
Washington University Zurich
Luis Millones
Denise Arnold
Seminario Interdisciplinario de Es-
Instituto de Lengua y Cultura Aymara,
tudios Andinos, Lima
La Paz, and University of London
Tristan Platt
Sabine Dedenbach- Salazar Saenz
Institute of Amerindian Studies, De-
Seminar fur Volkerkunde, University
partment of Social Anthropology, Uni-
of Bonn
versity of St Andrews
Penny Dransart Joanne Rappaport
Department of Archaeology, Univer- Department of Sociology and Anthro-
sity of Wales, Larnpeter pology, University of Maryland
Penelope Harvey Sarah Lund
Department of Social Anthropology, Department and Museum of Social
University of Manchester Anthropology, University of Oslo
A NOTE ON ORTHOGRAPHY
The different contributors to this volume have not opted for a uniform style of
representation in writing for the Quechua and Aymara languages. Variation depends
in part on whether or not a phonemic transcription is being offered. There exist official
alphabets for both languages, used in Peru and Bolivia particularly for language
planning purposes. However, where authors have made other orthographic choices,
the editor has respected these.
Creating Context in
Andean Cultures
1
ROSALEEN HOWARD-MALVERDE
Introduction
Between Text and Context in the
Evocation of Culture
A tAt its inception, the 1991 conference on "Textuality in Amerindian Cultures"

preface) proposed to examine diverse forms of cultural expression in indigenous Latin
America, with a particular focus on the textualising processes and textlike attributes
that such expression displays. The notion of text was taken in its very broadest sense:
thus the text of textiles—two words with a common root in Latin texere meaning 'to
weave'—could be as valid an object of our interests as the alphabetic script of the
seventeenth-century Huarochiri manuscript or the oral discourse of speech makers and
dramatists in modern day Peru and Colombia. The mutual influence between speech
and writing is shown to be crucial in a number of the chapters and forms a recurrent
feature of the volume. The metaphorical use of the concept of text as a paradigm for
the description of other areas of cultural practice proved to be more problematical;
this is a matter I discuss in more detail here, while some of the contributors also address
the issue in relation to particular ethnographic cases.
In this introduction I shall explore some of the diverse ways in which concepts
such as discourse, text, context, textuality, texture, and narrative have been used in
recent approaches to the study of culture, be it verbally manifested or otherwise. These
remarks are intended to clarify some of the terms of the debate from my personal point
of view; in addition, each author represented here has his or her own usages which
may further expand the discussion. In this way, the book presents a diversity of
viewpoints around a set of issues currently topical within the disciplines of social
anthropology and anthropological linguistics alike.
3
4 INTRODUCTION
With regard to the concept of "text", in recognition of the different disciplinary

traditions within which this concept is employed, I shall separate discussion of its
primary, linguistic, sense, from discussion of the more metaphorical uses of "text"
that have arisen in anthropology. This is not to suggest, however, that the contributions
contained in the volume necessarily make such a stringent separation: many of the
more anthropological papers make use of a rather linguistic notion of text (e.g.,
Harvey, Rappaport); Allen's chapter stresses that the Western notion of linguistic text
is based on premises quite distinct from those at work in many Andean processes of
textualisation; furthermore, in the Andes, the metaphor of "culture-as-text" may not
necessarily be the construct of the ethnographer, but part of the indigenous conceptual
system held by culture members themselves (e.g., Arnold, Platt).
The remainder of my essay will explore common threads between the chapters not
picked up in the discussion of the linguistically derived terms and concepts. Thematic
affinities will be shown across papers of different methodological orientations, from
the more text-linguistic contributions (e.g., Adelaar, Dedenbach), to the literary-tex-
tual ones (e.g., Millones, Lienhard), to the ethnographic studies (e.g., Harvey,
Rappaport, Allen, Dransart, Arnold, Lund, Platt).
Discourse and text as social process

The term discourse is applied in a variety of ways in linguistics, where it is taken to
be an essentially verbal phenomenon manifested in either speech or writing.1 Whether
spoken or written, discourse is a communicative tool necessarily arising within a
situation of utterance; in this sense, discourse is of the performative order of language
use and is socially grounded. A related view of discourse as constantly reconstituting
itself in the moment is to be found in Foucault's writings on the nature of discourse,
in his words:
We must be ready to receive every moment of discourse in its sudden irruption;

in that punctuality in which it appears, and in that temporal dispersion that
enables it to be repeated, known, forgotten, transformed, utterly erased, and
hidden, far from all view, in the dust of books. Discourse must not be referred
to the distant presence of the origin, but treated as and when it occurs. (1972: 25)
Foucault's observations pertain specifically to written discourses, such as the

documents of history which, according to him, are "monumentalised" by conventional
historiographical method. He advocates an approach that refuses to treat such docu-
ments as unassailable monuments, but seeks to analyse them by persistently asking
the question, as he puts it, "what was being said in what was said"? (ibid., 28). The
meanings of discourse are thus shown to be inextricably tied to the social and political
conditions within which it is produced; the applicability of Foucault's approach to the
analysis of oral—in addition to written—history, is something I have discussed
elsewhere (Howard-Malverde 1990).
The groundedness of discourse is as true of the spoken medium as it is of the written
one, although in the case of the latter the mode of reception is utterly different. The
place and time of reception of the written text are removed from the place and time
Between Text and Context 5
of its moment of production, so the naturally dialogic interaction that occurs in the
oral medium is missing, or can only be artificially constructed by the interpretive
activity of the reader (Ricoeur 1971). These observations may be borne in mind in
relation to a number of the essays in the volume, where it is shown that oral and written
modes of discourse are by no means always separate in practice. In light of the Andean
evidence, there is scope for a revision of theories regarding the differential social roles
and symbolic functions of literacy and orality in particular historical and cultural
contexts: Lund, Lienhard, Allen, Arnold, and Platt make this revision part of their task.
Also in connection with the "literacy-orality" discussion, Adelaar demonstrates the
effects on discourse structure and linguistic form of the transposition of a traditionally
oral language (Quechua) into the written medium.
It should also be mentioned, in relation to the meanings of "discourse", that
Foucault's agenda provides the critical discourse analyst with tools to unmask the ways
in which language is used to manipulate and control, especially in institutionalised
domains such as the mass media, education, and religion (cf. Fairclough 1989). This
leads us into a somewhat different area of analytical tradition, insofar as Foucault's
conception of discourse is not strictly speaking linguistic. As Duranti and Goodwin
point out, for Foucault discourse is a set of cultural practices and attitudes, a domain
of knowledge that regulates social life (Duranti and Goodwin 1992: 30). Such cultural
practices and attitudes frequently surface in language, although they are not confined
to it. Thus, from a more anthropological perspective, discourse can mean something
akin to ideology, and analysis of discourse can be conducive to an understanding of
the dynamics of power relations in society and processes of social change (cf.
Fairclough 1992; Street 1993: 15).
Following upon the above remarks about discourse, how shall we take the primary
meaning of text? For Paul Ricoeur, when manifested in writing, discourse gives rise
to the birth of text, which is held to be "any discourse fixed by writing" (Ricoeur 1991:
106ff.).2 In this view, fixity is a most characteristic feature of text, and indeed, in
commonplace usage, "text" bears connotations of an immutable script to which
members of a society are constrained to adhere in their cultural beliefs and practices,
as, for example, in the case of the constitutionalised "texts" that provide the basis for
world religions and legal systems.
By contrast, William Hanks, in a useful review article, proposes a definition that
orients us toward the idea of text as essentially social, and the notion of "fixity" thus
recedes: "a communicative phenomenon located in the social matrix within which the
discourse is produced and understood, towards which there is a social orientation,
rendering text interpretable by a community of users" (Hanks 1989: 96).
Such a definition coincides with the views of text adopted by Lienhard, Adelaar
and Dedenbach, according to which the literary and linguistic features of text are best
apprehended and interpreted in light of the specific historical and social conditions of
production within which it emerges.
Textuality, again to cite Hanks, can be taken as "the quality of coherence or
connectivity that characterises text" (ibid.). Thus conceived, the concept of textuality
stems from a prior concept of text in the terms just defined, and Hanks proceeds to a
useful summary of the achievements of text-linguistics in identifying those aspects of
linguistic form that help structure text and enable textuality. The use in text of deictics
6 INTRODUCTION
and orientational suffixes as analysed by Adelaar, and the patterning of evidential

markers in discourse as Dedenbach describes, both pragmatic issues relating to the
management of information, are excellent examples of the kind of text-linguistic
feature that Hanks is referring to here.
In considering what text is, it is instructive also to think about what it is not. Hanks
counterposes text to nontext in this respect: whereas text comprises a set of coherent
symbols, the nontextual would consist in the random, unmotivated, juxtaposition of
markings or objects, such as the senseless cacophony of a crowded street or scuff
marks on a public wall (ibid., 96). In this respect we can appreciate the way in which
the pebbles symbolically handled by the pilgrims in Allen's account are indeed
components of text: the same pebbles on some other occasion, outside the pilgrimage,
in the eyes of the goatherd casually passing by, might be meaningless, thoughtlessly
kicked underfoot, a nontext in Hanks's terms.
I would argue then for an open and flexible notion of text along the lines that
William Hanks suggests, whereby text is considered not merely as product but as
process, a process that can be seen as a mode of social action whereby power is
exercised, and human potential to bring about change in the world is fulfilled. Such a
view of text emphasises the interpenetration of textual and extratextual factors,
necessary for adequate interpretation; in Hanks's words: "The boundaries of text are
best conceived as extremely permeable, incomplete, and only momentarily estab-
lished" (ibid., 105). Such a view allows for the dissolving of the boundaries that
separate the text from the world, boundaries that in their turn serve to maintain the
illusion of the text's authority. This point connects with Ricoeur's conception of the
interpretive process, to which I return later in this chapter.
Text in context, text as context

A dynamic and flexible notion of text requires a rethinking of the related concept of
context. If the boundaries between the text and the world are conceived as permeable,
it is hard wholly to sustain the traditional view of text as being located in a context,
from which it can be disengaged for the purposes of analysis or, alternatively, to which
it must be related to enable a fuller, contextualised, interpretation. In the customary
view, context is taken to be something external to text, and the analysis of context in
its own right is rarely included in the search for meaning. Clearly, context has an
influence on the shape of text: the forms of discourse emergent in institutional settings,
for example, can be related to a context that is "there" in the minds of participants to
some extent, both prior and subsequent to interractions that take place within those
settings. Nonetheless, there is an argument for revising the somewhat unidirectional
conception of the text-context relationship as it has traditionally been held. While, on
the one hand, context can rightly be said to be constructive of text, from another
perspective, context can also be taken to be constructed by text in processes of
interactivity. Thus conceptualised, context itself becomes a phenomenon shaped in
social interaction. Text is then conceived not so much as in context, but rather as
creative and constitutive of context, as Duranti and Goodwin propose: "Neither the
physical nor the social setting for talk is something that is fixed, immutable and simply
'out there'. Instead these phenomena, and the very real constraints they provide, are
dynamically and socially constituted by activities (talk included) of the participants
which stand in a reflexive relationship to the context thus constituted" (Duranti and
Goodwin 1992: 7).
A number of the papers in this volume (e.g., Harvey, Rappaport, Dransart, Allen,
Platt) illustrate how texts and textualising practices are active agents in the shaping of
experience and, effectively, the creation of context, whether this be by verbal or other
symbolically expressive means.
Text as a model for the interpretation

of culture
Recent studies have called into question the appropriateness of text as a model or
metaphor for the description and interpretation of nonverbal forms of cultural expres-
sion (cf. Keesing 1987; Strathern 1987; Thompson 1990). The culture-as-text ap-
proach stems in particular from Clifford Geertz's inspired interpretations of social
interaction observed during fieldwork in Morocco and Indonesia. Geertz describes his
concept of culture as an essentially semiotic one, and elaborates his position in this
much-cited passage from "Thick description": "Believing, with Max Weber, that man
is an animal suspended in webs of significance he himself has spun, I take culture to
be those webs, and the analysis of it to be therefore not an experimental science in
search of law but an interpretive one in search of meaning" (Geertz 1973: 5).
The metaphor of culture as "webs of significance" calls to mind the linguistic
interconnections between 'web', 'weave', 'text', and 'textile', which allow for the
Aymara weavings discussed by Arnold (in this volume) to be as valid an object of
present concern as the Huarochirf manuscript; later in the same article, Geertz makes
use of the text, as opposed to textile, metaphor more specifically: "Doing ethnography
is like trying to read...a manuscript—foreign, faded, full of ellipses, incoherencies,
suspicious emendations, and tendentious commentaries, but written not in con-
ventionalised graphs of sound but in transient examples of shaped behaviour" (ibid.,
10). He further describes culture as an "acted document" (ibid.).
However, despite Geertz's quite explicit use of the textual metaphor in theory, a
perusal of his method of ethnographic description and interpretation in practice—as
with the complex episode of the Berber sheep-stealing raid in the same essay—reveals
it to be less reifying and reductionist than recent critics of the model claim (e.g.,
Keesing 1987: 165; Thompson 1990: 134). For Geertz, the "text" of culture amounts
to what he otherwise calls "social discourse", which may be conducted, as he puts it,
"in multiple tongues and as much in action as in words" (1973:18). The cross-cultural
misunderstandings and conflicting political interests of the actors involved in the
sheep-stealing episode are brilliantly unveiled by Geertz as part and parcel of his will
to demonstrate how culture operates as text. The subtlety of his insights, and his skill
in textualising events in the construction of his ethnographic account, lead to no
reduction in meaning—quite the opposite. As I read Geertz's reading of the power
game between the Jewish trader Cohen, the French commandant, and the Marmusha
tribal sheikh, I am caught up in the excitement of the bluffs and tactical skirmishings,
8 INTRODUCTION
and intrigued by both the psychology and the cultural significance of it all. However,
it seems to me that the text metaphor is somewhat superfluous to Geertz's ends. In
light of the subsequent criticism to which his theory has fallen prey, Geertz can be
said, in this case at least, to have got somewhat caught up in an interpretive web of
his own making.
J. B. Thompson, for example, takes particular issue with Geertz's approach,
arguing that the culture-as-text metaphor leads to the analyst eschewing problems of
power and social conflict in the interpretation (Thompson 1990:134). Thompson also
sees the metaphor of text as likely to impose a monochrome interpretation upon
situations of struggle in which multiple intentions and conflicting interpretations arise.
My own reading of "Thick Description," as briefly expressed here, does not allow me
fully to agree with Thompson on this point. Geertz's handling of the Moroccan data
in my view skillfully lays bare the interstices of power relations in a colonial setting,
as these are lived and experienced by human individuals of flesh and blood. Geertz's
mistake, if any, is to overlabour a semiotic theory of culture that he neither particularly
needs nor applies to the case study in point.
In reviewing the model of culture-as-text, Thompson also takes issue with Ricoeur,
to whom we owe an important elaboration of the paradigm (Ricoeur 1991). He sees
Ricoeur's definition of text—by dint of its "distantiating" quality as written dis-
course—as inadequate to engage with its own social conditions of production:
To proceed in this way is to ignore the ways in which the text, or the analogue
of the text, is embedded in social contexts within which, and by virtue of which,
it is produced and received; it is to disregard the sense that it has for the very
individuals involved in creating and consuming this object, the very individuals
for whom this object is, in differing and perhaps divergent ways, a meaningful
symbolic form. (Thompson 1990: 135)
This statement poses certain problems if juxtaposed with the interactive concep-
tion of the text/context relationship proposed earlier; the characterisation of text as ob-
ject somewhat contravenes the spirit of such a conception. But my main point here is
that Thompson's argument provides a rather one-sided view of Ricoeur's wider theory
of the hermeneutic process, a theory that in fact allows for both the internal structure of
the text to be exposed, and for the text to be related to its conditions of production; this
two-way process—the hermeneutic arc, as Ricoeur describes it—crucially involves
both explanation (text-internal) and understanding (text-external) in the interpreta-
tional equation (Ricoeur 1991:119). Understanding is a crucial dimension of the activ-
ity of reception, in so far as it involves appropriation, which Ricoeur defines as a
"struggle against cultural distance", a process, to cite him again, of "making one's own
what was initially alien"; emphasis in the original). Elsewhere I have examined how
just such a process of interpretive appropriation allows Quechua-speaking storytellers
to use communally held oral traditions verbally to shape the conflict-ridden contexts of
their individual lives (Howard-Malverde 1994). In this connection Ricoeur's thinking
ties in with Hanks's characterisation of text as social process.3
In attempting to determine the appropriateness of the culture-as-text analogy, it is
important to distinguish the perspective of the analyst from that of the cultural agent
(the producer of culture). Clearly, the analyst's ethnographic enterprise leads to the
textualisation of the observed. Here there is little to dispute, leaving aside for the
moment the fact that there are ways and ways of writing ethnographies, some less
disempowering than others with regard to the groups and individuals whose cultures
are described.4 What is at issue, rather, is whether the cultural practices being observed
can suitably be characterised as textlike in their primary unmediated forms, before
ever the ethnographer sets to work. Does not the metaphor of culture as text risk
formalising the non-formalisable? Is it not too convenient to presume that social
interaction is textlike in its patternings, when it comes to having to give a verbal
account of the non-verbalised?
From culture-as-text to the evocation

of culture
The idea of text cannot capture the experiential dimensions of socially and psycho-
logically grounded human behaviour: be it ritually framed or otherwise, such interac-
tion involves participants in strategies of positioning with regard to each other, a
positioning process bound up with relations of power, and whose meaning emerges
in its performative dimensions; the treatment of such processes as semiotic phenom-
ena—which the text metaphor necessarily entails—cannot but obscure these socially
situated meanings, positing an overly structuralist account, removed from the lived
significance of cultural actions from the point of view of the actors involved.
If we are not to get caught up in the semiotic web as interpreters of culture, what
approach must we then adopt? At this point, the analyst may take her cue from cultural
agents themselves, by attempting to get at how they engage in the chain of reception,
interpretation, and production of meaning. As Keesing (1987: 164) points out, mem-
bers do not "read" the meaning-embodying symbols of their culture, but rather
"evoke" them; that is to say, they engage with them and meaningfully reproduce them
in an experiential way. Marilyn Strathern, in her comment on Keesing (Strathern 1987:
173), sees in this concept of evocation both a path to a more performative notion of
culture, and a means to enabling a more dialogic encounter between the observer and
the observed. The culture-as-text metaphor is radically overturned in Strathern's view,
and along with it such text-derived notions as signification and reference: "Under the
umbrella of post-modernism, representation itself has become questionable, discourse
displaces text, alternities dissolve frames, and it becomes thoroughly appropriate that
in symbolic anthropology the idea of evocation or elicitation should push aside
signification and reference" (ibid.).
The implications of such a stance are far reaching, for they challenge the ethnog-
rapher to set herself free from the shackles of literacy, and look for new ways of
producing cultural accounts—by elicitation, evocation, and dialogue—that stand to
be less prepossessing than those produced by conventional textualising means. It
might only be by such a route that the "Western" voice inscribed in so many
ethnographies, and the voice of the indigenous subject, can ever conceivably meet.5
Increasingly, anthropological narrative is ceding ground to the voices of the people;
a commitment to such dialogue is surely imperative if the ethnographic enterprise is
10 I INTRODUCTION
to be meaningful in the postcolonial world, and will bring about gradual and radical
changes in the aims, objectives, and methods of anthropology in the years to come.6
All this being said, there is no doubt that Andean culture members themselves are
prolific producers of text (be it spoken, written, woven, inscribed). Furthermore,
indigenous Andean peoples have long incorporated the symbolism of Western writing
into their theories of history. As Platt demonstrates, for example, the Macha shaman's
Quechua discourse incorporates metaphors of alphabetic literacy and legalisrn as a
means of cohering ideas about alternative sources of power: local autochtonous—the
mountain gods, and national—the government and the state. To reject the culture-as-
text analogy as an artifice of the ethnographer, might lead us to overlook certain
features of symbolic practice actually produced by the people themselves. Other
contributors to the volume will also have reason to investigate the usefulness of the
model in application to particular cases (cf. Rappaport, Dransart, Allen).
From representation to performance

Strathern also refers to the concept of representation, the appropriateness of which
has been recently questioned in social anthropological debate as well as within cultural
studies. The problem with the idea of "representation" is that it suggests some
objectified reality that pre-exists and stands outside the moment, then to be enacted
before an audience who supposedly perceives it as an account of an external state of
affairs. Thus, for the sake of argument, the Aymara herders in the llama corral
(Dransart) or the goose-stepping paraders in the streets of Ocongate on Independence
Day (Harvey) would be deemed to be "representing" perceived relationships between
humans, animals, and landscape, or between local community and nation-state, rather
than living these perceptions in the process of conducting rituals and ceremonies, as
these authors make plain.
A performative conception of cultural process, by contrast, displaces the concept
of representation—an analytical construct in the first place, and reality is conceived
as constructed and emergent, an experiential process grounded in time and place, and
in relation to the individuals and groups who, through their participation, position
themselves in relation to one another and the wider social order. The following extract
from Stuart Hall's discussion of the way in which cultural identities emerge in
interaction illustrates this perspective:
Perhaps instead of thinking of identity as an already accomplished fact, which

the new cultural practices then represent, we should think, instead, of identity as
a 'production' which is never complete, always in process, and always consti-
tuted within, not outside, representation. This view problematises the very
authority and authenticity to which the term 'cultural identity' lays claim. (Hall
1993: 392; emphasis added)
There is obviously common ground between this emergent conception of cultural

practice and the social interactive approaches to the text/context relationship emerging
in linguistics, as outlined earlier.7
Enactment and the use of texture in

visual performances
A performative conception of culture can be carried further if we introduce the related
notion of enactment, for a number of reasons. Enactment is multifold in its levels of
reference: it at once suggests involved participation by culture members (the "en-"
prefix is the active component here), a playing out of experience (as in the theatrical
sense of "act"), and an actualisation (whereby latent meanings are realised in relation
to a here-and-now). Enactment enables us, yet again, to conceptualise text not so much
as being in context, but rather to think of text as context. Rappaport' s chapter provides
an illustration of this approach: the politicisation of Cumbal youth is a process that
emerges in the activity of doing theatre in Narino; theatre here may be seen not merely
as a medium for the representation of social and political circumstances, but as a means
of empowering actors to interiorise, dramatically to interpret, and eventually to exert
change upon those circumstances.8
The activity of interpretation referred to here is not the textualising procedure
of the anthropologist, but a performative activity on the part of the agents of culture
themselves. This interpretive activity may involve retellings of narratives, re-en-
actments of ceremonies, and so on. What is of interest in this process is the way
in which the distinction between production and reception is blurred. In each
performance, the particular shape of production echoes an aforegoing process of
reception, of some previous performance in the past; at the same time, it contains
the seeds of future performances to come, which in their turn will channel receptive
responses and understandings into new instantiations, new productions. The process
is a circular one of continuous and ongoing interpretation—-Ricoeur's hermeneutic
arc in living practice.
Theatre in a multilingual and pluricultural society—where linguistic codes are not
necessarily shared between performers and audiences, or across different sectors of
audiences—relies heavily on the visual for successful communication, on texture
rather than text, as Rappaport puts it. We take texture to include expressive media such
as dress, heraldry, visual icons, use of colour, sound, and body language, some or all
of which enter into the literally theatrical enactments of the cases studied by
Rappaport, but which also figure in the ceremonial practices relating to Peruvian
Independence Day described by Harvey, and which perform a key symbolic function
in the Aymara rituals, appropriately defined as "acts of remembrance", analysed by
Dransart. In addition, the texturelike qualities of sound are well evoked in Platt's
discussion of the shamanic seance, which takes place almost entirely in darkness.
"Enactment" also enables us to grasp what is happening as the pilgrims to Qoyllur
Rit'i play out their aspirations and desires through the ritual handling of stone
miniatures (Allen). The Quechua language itself offers support in favour of these links
and cross-references, insofar as the verb pukllay can be taken to mean 'play' in the
sense of 'act out'. "Pukllay" crops up across a number of contexts, suggestive of the
full extent of its semantic range: as the word for the fight between the condor and the
bull (Harvey), as the word the pilgrims use to describe their activities at Qoyllur Rit'i
(Allen), and as part of the lexicon of theatre (Montoya 1993).9
12 I INTRODUCTION
Rethinking the theory of the linguistic sign

The conceptualisation of cultural expression as performance or enactment, as well as
its application to the Andean examples considered in this volume, leads us to think
about the construction of meaning as entailing a somewhat different relationship
between signifier and signified from that commonly pertaining in twentieth-century
Western linguistic theory. The Saussurean triangle of signification posits what could
potentially be a non-hierarchical relationship between ideas, words, and the objects
to which the words refer. Nonetheless, in certain logocentric traditions of the West,
the linguistic sign tends to occupy the prior position, and words are taken to be signs
for things, rather than vice versa.
A theory of meaning as process, if applied in the domain of symbolic anthropology,
allows us to collapse the triangle and envision instead a synthesis of signifier and
signified, in a signifying relationship whereby neither can be abstracted out from the
other or be considered superordinate to the other. Kenneth Burke (1966) has demon-
strated for the case of religious language that perceptual processes occurring as part
of spiritual experience may imbue profane words with new, sacred, meanings. In this
sense, he proposes that the signified (the object perceived and endowed with sacred
power by the perceiver) becomes a sign for the signifier (profane words take on sacred
meanings and can be put to new effect in discourse addressed to God).
Burke bases his discussion on language in Judeo-Christian tradition; however,
similar processes seem to occur in Amerindian settings. For example, among the
Aguaruna of Northern Peru, shamanic chants are learned by the ritual specialist
through contemplation of objects in nature. Plants are the "owners" of healing songs
and release their words into the consciousness of the shaman in response to ritual
observances (Brown 1986). Bearing Burke's discussion in mind, were we now to
reintroduce, for the sake of analysis, the notion of "standing for" into the synthetic
signifier/signified relationship suggested in the preceding discussion, then, by main-
taining our conception of meaning as process, we would be led to question, along with
Burke, whether the signifier is necessarily always the sign for the signified or whether
this relationship cannot be turned on its head in some cases, and the signified be seen
as the sign for the signifier.
This strategy of inverting the accepted Western canon can be useful for
understanding Andean ways of thinking, particularly in relation to cosmological
beliefs: the papers of Allen, Dransart, and Lund provide evidence for this. Objects
(stones, sites on the landscape, letters) are found to be meaningful tokens in their
own right, endowed with message-embodying powers, which humans utilise for
particular ends. This utilisation is part of a reciprocal interpretive process whereby,
on the one hand, humans exert their agency on the external elements of the animate
universe in order to control and bring about desired change in that universe and,
on the other hand, elements of the animate universe exert their agency on the world
of humans.
The two-way tension inherent in the signifying process that the ethnographic data
suggests, seems all the more viable as a model for understanding ways of meaning in
the Andes, when we consider that the Quechua language itself possesses a number of
terms that are semantically complex in a related way. For example, the verb unanchay
means both 'to make a sign' and 'to interpret a sign' (Goncalez Holguin 1952: 355).
The verbyachay 'to know' provides a derived form yachakuy 'to become accustomed,
to settle down', from whichyachakuchiy can be further derived, meaning 'create, bring
into being', which is commonly used to refer to acts of mythic creation in the
Huarochiri texts (e.g., Taylor 1987a: 52). Knowledge at once brings the world into
being and can be derived from observation of the world, as the features of the landscape
embody the past and keep the oral tradition alive in the present.
Narrative and nation building

Another conceptual tool used by a number of the contributors is that of narrative.
Narrative is a mode of storytelling that lends itself to the presentation of events in
linear fashion and as such can act in a hegemonic capacity, imposing and sustaining
one preferred version of events over and above all others. Thus, in the telling of history,
narrative can create a sense of inevitability, a naturalising sense of the unquestionable
in the given order of things. Narrative amounts to the discourse of history, whereas
non-narrativity may better characterise the discourse of myth. In the latter case,
spatio-temporal relations are non-linear and open to contestation; in the former, these
relations tend to support a monolithic ideology of nationhood, couched in a linear
concept of time, and are harder to deconstruct and rearrange.
Harvey's contribution to this volume demonstrates the tension between narrative
and counternarrative (the state's version of history is contrasted with an indigenous
expression of structural relations that invokes the pre-Hispanic past), and it touches
on the effects of terrorist insurgency in Peru, which challenges all modes of narrative
alike. Rappaport also makes use of the concept of narrative as she shows how indianist
Cumbal playwrights construct a counterstate version of history in their attempts to
rewrite the past.
The narration of history in any language raises issues with regard to the nature of
sources of knowledge and the degree of veracity and authenticity that narrators
attribute to these. In Quechua, such concerns are embedded in the grammar, in a series
of sentence suffixes of epistemic modality ("evidential" or "testimonial" suffixes).10
That epistemological concerns should be intrinsic in grammar suggests that "source
of knowledge" is a deep-rooted conceptual category, the analysis of which is likely to
prove fertile for demonstrating ties between language and culture. At the level of
cultural analysis, Harvey's chapter suggests that this category is indeed operating in
the conceptualisations of history that are ceremonially enacted in Ocongate on
Independence Day. From the linguistic angle, Dedenbach's chapter shows how the
use of evidentials in text is central to the construction of multiple points of view on
the part of the narrator(s) of the Huarochirf myths.
The chapter by Millones also deals with the role of cultural expression in "nation-
building", in particular the role of the Catholic Church in its mission to shape the New
World in its own image, and Andean challenges to this. The uses to which the image
of Saint Rose is put in her different cultural settings (as expressed in poetry and song)
sheds light on the alternative modes of consciousness—Latin American, Peruvian
national, local indigenous—that evolve, as religion is used to shape society.
14 INTRODUCTION
A voice for indigenous peoples

I referred earlier to the challenge to develop a more dialogic approach in ethnographic
research and writing. We need to keep separate the issue of whether and how our
scholarly pursuits might contribute to empowerment and audibility for indigenous
peoples from the ways in which Andean indigenous peoples themselves have striven
to make their voices heard in the colonial setting, where cultural and linguistic barriers
have persistently hampered them in the exercise of power. I here turn my attention to
the latter.
The voice of indigenous peoples—as long as it is projected into the wider social
arena where those who listen do not share the same position in the social structure, or
the same experience of colonisation, as those who speak—will necessarily have to be
shaped in order that the non-indigenous will understand the words. This process does
not only require the acquisition of European linguistic codes but may also incur the
adoption of values and concepts actually derived from the colonising group, a strategy
of conceptual accommodation that arises from the need to acquire both communicative
and merely linguistic competence across the ethnic divide. As usual, it is the subordi-
nated group that makes the accommodating gesture in order to set up a framework for
communication in the first place. This point is made by Lienhard in his discussion of
the epistolary discourses that arise from indigenous attempts to coopt alphabetic
literacy for the purposes of communication with successive colonial administrations.
Hybridisation as a strategy of accommodation and

resistance
Spanish colonisation of the indigenous people of the Andes produced profound
cleavages in existing structures at every level of human experience: in systems of
communication, knowledge, religious belief, and social and political organisation, to
name but a few. The strength of those who resist oppression and try to maintain
integrity against all odds is a factor in explaining how five centuries of colonialist
domination have not completely eradicated linguisitic and cultural diversity and
creativity among Latin American peoples of indigenous origin. The very cleavages
mentioned here have been moulded into spaces in which innovative forms of cultural
expression are produced, acting as tools in the fight against total assimilation.
Hybridisation is a phenomenon naturally generated at such interfaces, and many of
the contributions to this volume deal with the problematic of what happens in this
"space between". I shall explore how this is so.
To take linguistic practice as a starting point, in verbal discourse hybridity pertains
at a number of levels: in the medium, the form, and the content of the message.
Regarding the medium of the message, hybridity was at work early on in colonial
indigenous texts that combined alphabetic script with visual images, such as the
illustrated manuscript of Guaman Poma de Ayala. In this volume, several chapters
deal with transpositions and articulations between one medium and another: Arnold
provides evidence that weaving and writing perform complementary and interdepen-
dent functions among the Bolivian Aymara, each medium expressing differential
relations of production of agricultural crops and knowledge, along gender lines.11

Lund's study of the uses of literacy by non-literates in highland Peru illustrates how
orality and writing may merge into a common set of mutually reinforcing practices in
specific social-historical circumstances. With regard to hybridity of form: the excess
of honorifics in the written Mesoamerican discourses studied by Lienhard exemplify
the way oral habits of speech may intrude into fledgling styles of writing; the effect
of incipient literacy on language form and use is also the theme of Adelaar's chapter.
As far as content is concerned, the Huarochiri manuscript is an example of textual
production emerging from a "place between" on this level, too: the Quechua narrator's
conflict of loyalties toward two systems of religious belief is a feature discussed by
both Adelaar and Dedenbach.
The emergence of hybrid forms of religious belief and practice is also dealt
with in Luis Millones's paper, where the cult to Saint Rose is shown to have
evolved into a specifically Andean form of devotion, in which it is compatibilised
with other focuses of belief and constructs of history, involving both the cult to
the earth and commemorative ceremonies that re-enact the capture and death of
the Inka Atahuallpa.
The practices described by Rappaport and Harvey, emergent in theatrical and
ceremonial space, respectively, are seen as modes of resistance to state-interested
versions of history and social structure. In the case of the Cumbal dramatists
(Rappaport), theatre provides the medium for a "re-invented" ethnicity that serves to
reinforce identities at the interface between indigenous and national society. In the
case of Harvey's study, a series of interlocking symbolic relationships are observed:
between past and present, landscape and urban space, Quechua and Spanish, province
and metropolis, non-Christian and Christian, and, most ambivalently, condor and bull.
None of these is an oppositional relationship, but together they are a series of
ambivalent intermeshings that produce hybrid discourses in cultural practice.
If landscape embodies counterhistory, as suggested by public ceremony in Ocong-
ate, private rituals focused on pebbles, the product of that landscape, may also be
interpreted in light of the concept of interface. The stones used in "play" by pilgrims
to Qoyllur Rit'i, as described by Allen, are seen to be a means of communication with
the upper and lower worlds by the human inhabitants of the world in between, the kay
pacha of Andean cosmology.12 The importance of stone(s) in Andean theories of space
and time cannot be overemphasised: the lithomorphosis of the culture hero or heroine
in myth has the effect of leaving a lasting trace of his orherpast activities in the present,
visible to the living eye for generations to come (cf. Duviols 1979); stone stands at the
crossroads between past, present, and future, and between inner and outer worlds, and
perfectly epitomises the merging of space and time in Andean thought.
In the wake of colonisation, cultural hybridisation may be the only means whereby
an effective indigenous discourse of resistance can evolve: the "authentically indige-
nous" is an inescapable anachronism in such a setting, while the alternative to
hybridisation is the engulfment of the indigenous by the hegemonic cultural forms and
values of the colonising society. From this point of view, hybridisation generates a
space for the formulation of new meanings, by combining re-use and transformation
of the indigenous, with appropriation and adaptation of symbols originating with the
dominant culture.13
16 IINTRODUCTION
Concluding remarks
The essays in this volume represent Andeanist scholarship within a number of
fields—anthropology, linguistics, history, literature—and reflect the diverse scholarly
traditions of a range of countries. Despite the diversity of approach, several recurring
themes unite the chapters and make of the volume a coherent whole. This coherence
arises in part from the geographical focus of the book: the Andean area, although
physically extensive, and displaying linguistic and cultural heterogeneity still today,
also enjoys a degree of common cultural history reaching back to pre-Hispanic times.
Furthermore, European colonisation, while not a uniform experience for all the human
groups involved, nonetheless drew indigenous Andean peoples into types of structural
relations with the conquerors that can usefully be compared from group to group:
relations of social, cultural, and linguistic domination, as well as ones of resistance
and accommodation.14
The studies presented here focus on indigenous responses to the experience of
colonisation, as revealed in diverse forms of cultural and linguistic expression ongoing
today. Common themes include social identity; state-indian relations; the concept of
nationhood from the indigenous perspective; political violence and its traumatic
consequences; gender relations; the construction of knowledge and multiple theories
of history; cosmological beliefs about different sources of power; and alternative
media of communication, and the techniques and ideologies associated with these. In
all, it is seen that language and culture are powerful instruments for creating contexts,
and shaping identities within those contexts, in ways that challenge the official view.
Culture, be it enacted verbally, in ritual, or by visual or other means, emerges in the
Andean practices described in this book as an effective political tool.
I shall end by reflecting on one feature of Andean cosmology that has proven to
be particularly dynamic and adaptable to circumstances through the ongoing process
of colonisation: the apparently changing symbolic role of the condor. The condor can
be said to be the embodiment of indigenous Andean identity and focus of cosmological
beliefs par excellence. In many parts of the Southern Andes, where it most abounds,
the condor is believed to be an ancestor god and one of the forms adopted by the
mountain spirits when the latter are invoked by humans in ritual contexts (see Platt,
this volume). In my own fieldwork, I have noticed the symbolic role of the condor in
the indigenous fiestas of Northern Potosf, Bolivia: the fiesta sponsor's brother-in-law
(tullqa)—a spouse-taker who forms the affinal link between two family groups—dons
a feathered condor skeleton as a headdress and cavorts around the periphery of his
wife's brother's fiesta-going group. This example lends itself to the interpretation of
condor symbolism in terms of in-group kinship ideologies and local cosmovisions,
which I cannot develop here.15 However, the second and final chapters of this volume
(Harvey and Platt) also accord a special place to the condor, and here, by contrast with
the case I have just cited, what strikes the reader is that the symbolism attached to the
condor is very much a product of the colonial process. In these accounts, the power
of the condor is seen to extend not only over the autochtonous sphere (ancestor god;
brother-in-law of a local patriline), but to incorporate elements of power deriving from
the outside, mestizo, society. The condor of Ocongate (Harvey) is on a par with other
visiting dignitaries representative of the state; the condor of Macha (Platt) is a
lawyer-judge whose Quechua discourse is peppered with Spanish-derived legalisms

and who masters alphabetic writing. I can advance a closing speculation here, in line
with my previous remarks about hybridisation. While the condor is the paramount
symbol of the Andean past and local autochtonous power, perhaps the key to its
survival as a symbol for the future lies with its flexibility to adapt, and an ability to
relate to the outside as well as to the world within, as new hybrid contexts are
symbolically, as well as actually, created.
Notes
1. See Schiffrin 1994 for a survey of the many approaches to the study of discourse in
pure and applied linguistics. I do not here discuss the autonomous linguistic notion of discourse
as a unit of analysis above the level of the sentence, as my focus is strictly on discourse as a
social, albeit verbalised, phenomenon. For extensive treatment of the concept of discourse as
social practice, see Fairclough 1992.
2. Ricoeur's use of the term discourse can be contrasted with that of Emile Benveniste,
also writing in French, for whom discours is necessarily spoken while, by contrast, histoire
supplies the written order of language with its particular formal characteristics (Benveniste
1966).
3. For further discussion of Ricoeur's concept of text, see Harvey in this volume.
4. For a discussion of problems of power in social science research and the writing of
culture, see, for example, Clifford and Marcus 1986 and Cameron et al. 1992.
5. I recognise that the definition of who or what is "Western" is highly problematic; I am
here referring to the classical type of anthropological writing that emerged as part and parcel of
the colonisation process in many parts of the world and which is becoming increasingly
challenged today.
6. The spirit of this imperative is well captured in the innovative collection of essays in
postcolonial criticism aptly entitled De-scribing Empire. Post-Colonialism and Textuality
(Tiffin and Lawson 1994); cf. Arnold et al. 1992 for a further example of dialogic ethnography.
7. The intellectual debt of these perspectives to the theory of practice of Pierre Bourdieu,
whether it be direct or indirect, also needs to be acknowledged (Bourdieu 1977).
8. Rodrigo Montoya's contribution to the conference, published elsewhere (Montoya
1993), describes the comparable use to which theatre is put among young militant theatre groups
in Peru.
9. Montoya (ibid.) also points out the similarity in semantic range between Quechua
pukllay and French jouer, in this connection.
10. Evidentially is a feature of the grammatical systems of many languages, cf. Chafe and
Nichols 1986; Howard-Malverde 1988; Dedenbach, this volume.
11. That Aymara exegeses of weaving make recourse to the language of ploughing is
interesting in light of Walter Mignolo's recent observations regarding "connections between
writing, weaving, agriculture, and distribution of labour according to gender", not only in
Mesoarnerica and the Andes but also in the Latin Middle Ages (Mignolo 1994: 296-297).
12. The notion of a conceptual "space between" that acts as a place of regeneration is an
enticing one for Andean scholars, insofar as it can be related to a properly Andean construct,
that of the chawpi in Quechua (Aymara taypi) or 'middle place', which has been examined in
a number of studies of Andean cosmology and symbolism (e.g., Bouysse-Cassagne and Harris
1987).
13. The consequences of such a process are not without its perils, however; hybridisation
can entail something of a double bind. This point has been explored in relation to the aboriginal
18 INTRODUCTION
populations of Australia by Gareth Griffiths, who argues that as a result of his/her attempt to
construct a discourse that will be listened to across the cultural divide, the aborigine is seen as
"less authentic" by the white colonial society, and this very lack of authenticity provides yet
another argument on the part of the whites for disallowing the legitimacy of the aborigine voice
(Griffiths 1994). A comparable Latin American case of double bind is discussed in relation to
the polificisation of the Tukano of lowland Colombia by Jean Jackson (1991). For a theoretical
critique of the wider issues involved here, see Spivak 1993.
14. The Andean focus is with the exception of the chapter by Lienhard, which compares
Andean, Mesoamerican, and Guarani sources.
15. See Harris 1994 for an interpretation of condor symbolism in Northern Potosi in terms
of gender ideologies.
I
TEXTUALISING HISTORIES AND

IDENTITIES IN CULTURAL
PERFORMANCES
2
PENELOPE HARVEY
Peruvian Independence Day

Ritual, Memory, and the Erasure
of Narrative
Our society has become a narrated society in a threefold

sense: it is defined by narratives, by quotations often, and
by their interminable recitation.
Michel de Certeau (1984: 52)
P
Peruvian Independence is commemorated annually on 28th July. Throughout the
nation, this date is marked as significant by official ceremonies, speeches, parades,
and other public displays of "patriotic civic consciousness". In many small Andean
towns, these official ceremonies are followed by sporting events, and in some places
by a bullfight, which is billed as the highlight and grand finale of the celebrations.
As might be expected, the official acts of commemoration pay particular attention
to that moment in the nineteenth century when Peru was declared an independent state.
The historical processes that link contemporary Peruvians to that moment and render
it significant are made explicit in public discursive accounts. By contrast, the bullfight,
an event that offers no explicit narrative or chronology of events, allows quite different
connections to be drawn between past, present, and future. Most importantly, it offers
a shift in focus away from the moment of independence to that of the encounter
between the Hispanic and the indigenous worlds.1 Thus, despite the fact that the
bullfight is an integral part of the commemorative activities, it invites a reinterpretation
of the accounts produced in the official ceremonies. Above all it stresses the presence
of indigenous peoples in the contemporary nation state. It is an occasion that produces
21
22 TEXTUALISING HISTORIES AND IDENTITIES
intense interaction and an image of fruitful cooperation between agents of the state,
supernatural powers of the landscape, and local people.
However, while the public ceremonies and the bullfight provide contrasting focal
points for the central commemorative event, I will argue that in many ways the effects
of these rituals are complementary rather than contradictory. The indigenous presence
is not an alternative, but an integral aspect of the Peruvian nation. Highly stylised and
conventional accounts of the relationship between past and present are enacted in both
parts of the extended Independence Day rituals which systematically convey a positive
evaluation of the role of the state in contemporary Peru. In this respect it is not
surprising that Shining Path, the guerrillas at war with the state in Peru, have
consistently intervened in the commemorative ceremonies, bombing public installa-
tions and threatening, even murdering, state officials.2 Their activities constitute a
direct challenge to the ways in which the official ceremonies depict Peru as a nation
of free citizens. Shining Path do not look to the past to motivate their actions. When
they talk publicly, they focus on a very particular configuration of contemporary social
relations, and their narrative is directed toward the future. For them, the significant
historical event is the moment of the state's destruction.
This chapter is not directly concerned with these very dramatic and salient acts of
contestation, but rather with the ways in which the annual commemorative activites
are realised in the interpretative acts and bodily practices of those participating in
them. It will be argued that latencies of memory, connecting past experiences with
present concerns, motivate particular understandings of the activities in which people
are engaged. As collective practice, commemorative rituals re-present the past,
depicting and interpreting it, shaping it to particular social ends, yet, as embodied
practice these rituals are also generated by the past, by the former experiences and
assumptions of participants.3 The relationship between history, memory, and ritual
will thus be considered by looking at the ways in which events are brought into
connection with each other and at how these connections, the articulation of past and
present, constitute people's experience of history. The focus is on lived experience
and owes much to Munn's discussion of how regionality is constituted in experience,
as part of the lived world of the island of Gawa in Melanesia. She stresses the
importance of looking at the "experiential syntheses that actors create in practices and
the events that transpire in their terms" and that these syntheses occur "through
microhistorical processes in which Gawans carry forward certain past events (spatio-
temporal 'moments') and configure certain futures in their present experience" (Munn
1990: 2).
I take historical narrative as a parallel example of such a cultural synthesis. The
official ceremonies of commemoration produce explicit narratives of Peruvian his-
tory, which depict a chronology of significant events generated through the agency of
a number of notable and named figures whose actions guided the emergence of the
modern state. The telling of this story in the ritual of commemoration produces a
particular ordering of the past for conscious consideration. This kind of official history
constitutes a very different mode of retrospective interpretation from what we usually
understand by "memory". Indeed, given that the event lies outside contemporary
experience, its memory has to be actively constructed in the ritual.
Peruvian Independence Day 23
Two problems occur to me as I consider the possibility of working with this idea
of the contrasting connective force of historical narrative and remembered experience
in Peru. The first is that this particular distinction between history and memory is
grounded in a very specific European tradition, with all the obvious limitations for a
cross-cultural application.4 The second is that the consideration of history and memory
in these terms raises questions about the relationship between discursive and non-dis-
cursive practice, narratives that link past and present through the production of explicit
connections, and fragmented, disconnected embodied memorisation which may or
may not be put into words. A consideration of these divergent ways of remembering
and recounting will be used to generate a critique of the ubiquitous use of textual
metaphors in contemporary Western cultural theory.
History and memory

As thinkers in the nineteenth century began to explore the nature of the agency of the
human subconscious, memory was represented in philosophical discussion as a
privileged site of subjectivity that fascinated because of the creative power of its
hidden side. In the emerging schools of psychoanalysis, conscious memory was
apprehended as the subjective construction of past events, a coherent narrative
sequence that coexisted with other possible but hidden meanings. This realm of the
psychological other—the alternative, subconscious, forgotten versions of the past—
could be triggered into consciousness by moments as apparently insignificant as
Proust's tasting of the famous petite madeleine (Proust 1987: 48).
Memory occupied a totally different space of preoccupation from objectifiable
history, the emerging discipline that sought to achieve an analytical distance from
subjective memory and arrive at a truth undistorted by the complex unconscious
concerns of individual psychology (Nora 1989: 5). History and memory as modes of
retrospective interpretation in the European tradition thus carry the overtones of a
distinction between objectivity and subjectivity, between general truth and individual
experience.
It is now generally acknowledged that both history and memory are social
constructions whose difference lies in their conditions of production and reproduction
rather than in their truth value.5 Within this framework it is usually assumed that
memory furnishes the raw material from which history is abstracted. Such memories
might be either verbal or written archives of past experience, but they exist as concrete
articulations and are thus available for redefinition and interpretation. The agency of
non-discursive memory, the forgotten experiences, the silences and spaces that ani-
mate the unconscious, is harder to assess but obviously lies at the heart of the
psychoanalytic tradition and the theoretical application of this perspective in recent
discussions of social history.6
Despite the cultural specificity of these abstractions, there are justifications for
applying some distinction between history and memory, as ways of telling, to the
Peruvian context. First, the commemorative event that I am discussing here is itself
rooted in nineteenth-century European culture. The declaration of independence in
1821 referred to a political struggle between two groups of Hispanics and did not in
any sense imply the end of European cultural influence, particularly at the level of
state activities. If anything, these influences increased as the independent state sought
to integrate the population in the service of the new political economy. History, as
coherent authoritative narrative, is thus an important aspect of the state's symbolic
production. Second, local concepts of narrative genre are also concerned with truth,
accuracy, and narrative authority. The Quechua language distinguishes the recounting
of subjective experience from events reported by others through the use of grammat-
ical markers that enable a speaker to establish the greater or lesser distance entailed
by direct, as opposed to indirect, knowledge. While these linguistic markers are
frequently used to denote respect or deference, the reluctance to take responsibility
for the facts of an account can also open up potential ambiguities as to the account's
accuracy or impartiality. In this respect, the claim to truth or accuracy of narratives
on the basis of personal experience is established in terms similar to those with which
we are familiar.
There are, however, narratives that discuss events that occurred in other worlds,
prior to the one in which people live today. These are the narratives of ancestor spirits,
of Inkas, of fantastic animals. These stories also inform and order people's understand-
ing of the past, but they are less concerned with the chronological sequencing of events
that explicitly connects the past to the present. The notions of time implicit in these
accounts are not primarily evolutionary, insofar as the contemporary world is said to
have emerged suddenly from the radical and catastrophic transformation of former
ways of being and is not presented as a result of the accumulation of past events. The
ancestor spirits who lived in the age of the moon were displaced by the coming of the
age of the sun, just as the Inkas were displaced by the Spaniards and the advent of
Christianity.7 Nevertheless, these categories of this world and other world narratives
are not definitively distinguished. Ancestors and Inkas were displaced, but not
destroyed. They are still significant and potentially active forces in contemporary
social life. Inka history is taught in school, ancestor spirits appear to the living.8
Ultimately, the validity of any of these accounts depends on the extent to which
the teller can get others to accept it. The articulation and sequencing of past events
carries more or less authority, according to the circumstances of the telling. Both the
official commemoration ceremonies and the bullfight provide coherent, optimistic,
and stylised accounts of the relationship between past, present, and future. Further-
more, these accounts contextualise mundane memory. They are immediately recognis-
able in terms of various practices in which people are habitually involved. However,
the authenticity of such accounts is not absolute, and, as this chapter will show, these
privileged and overt connections are vulnerable to contestation in the interpretative
process. This co-existence of privileged and alternative interpretations brings me back
to the question of the relationship between discourse and experience and the appro-
priateness of the textual metaphor for the analysis of events of this kind.
Commemorative ritual as text

Much has been written about the ways in which cultural theorists have extended the
metaphor of text to cultural practice in general, including the anthropological practice
of doing ethnographic research. In these debates the work of Paul Ricoeur on mean-
ingful action and the concept of text has emerged as a convenient point of reference.9
For Ricoeur, the essential features of text are associated with the processes of
distantiation that objectify particular cultural activities and render them available for
interpretation. A recent article by Moore (1989) traces very clearly the development
of Ricoeur's treatment of the human subject in the tradition of hermeneutic phenom-
enology, from a concern with spoken discourse, to written texts, and finally to
meaningful action. In all these cases, Ricoeur seeks to articulate a new relationship
between sense and reference, system and performance, semantics and semiotics. If we
take the case of discourse, these paired dimensions of meaning, which exist in a
dialectical relationship to each other, embrace both the language system and the world
outside to which language ultimately refers and in which discourse is realised. Ricoeur
stresses that all discursive interaction is interpretative, as language is inherently
polysemic: speakers thus have to reduce linguistic ambiguity through interpretation
in order to communicate, but this polysemic quality of language also expands the
possibilities of reference through metaphorical extension.
Despite these common features, Ricoeur is concerned to distinguish speech and
writing on the basis of the autonomy of the written text from the conditions of its
production. It is in this regard that he elaborates four forms of distantiation:
The first form of distantiation concerns the fact that what writing preserves is
the meaning of discourse in such a way as to transcend the passing moment of
the instance of discourse itself. Thus the event of saying is surpassed by the
meaning of what is said.. .. The second form of distantiation concerns the
relationship between the intention of the author and the meaning of the text. .. .
In written discourse the author's intention and the meaning of the text cease to
coincide. .. . The third form of distantiation concerns the gap between the text
and the conditions of its production. Unlike spoken discourse which is addressed
to a hearer, the written text is addressed to an unknown audience, and, in theory,
to anyone who can read.. .. The fourth form of distantiation designates the
emancipation of the text from the limits of 'ostensive reference'. ... A text does
not refer to the situation of its production, but to the world. (Moore 1989: 94)10
Such a notion of text does not in itself provide the framework for the dynamic
performative understandings that Ricoeur is seeking to establish. Structuralism oper-
ates with a similar model of text, which might indeed appear to encourage a disen-
gagement from the world of social relations. It is the dialectic between polysemy and
interpretation that continues to interest Ricoeur. Thus while readers will engage with
the sense of the text (in the Levi-Straussian sense) they will also look at what the text
refers to, at how it relates to the world.
Finally, Ricoeur's theory of interpretation is extended to what he calls "meaningful
action". Human actions are analagous to written texts in their plurivocality and insofar
as they undergo the necessary process of objectification as set out in the four forms of
distantiation. Meaningful action can be conceptualised as a text that discloses a world
open to a variety of competing interpretations.
Ricoeur's notion of text has obvious attractions for post-structuralist anthropology
with its stress on polyvocality and contested readings, the anti-essentialist approach
26 TEXTUAI.ISING HISTORIES AND IDENTITIFS
to meaning and the possibility of a new synthesis of structuralist and performative

approaches. However, there are certain potential problems that I will briefly outline
before proceeding to a more detailed analysis of the particular commemorative event
with which I am primarily concerned.
Moore's objections to Ricoeur centre on his failure fully to appreciate the im-
plications of the fact that texts are produced in social relations and that the symbolic
ambiguity on which both interpretation and creativity rely is the primary site for the
assertion and contestation of social inequality. Social texts are not produced in a vac-
uum. They are produced for an audience with expectations and wishes that have deter-
mining effects on the kinds of texts that are produced and recognised as such. In this
respect, the following analysis of the Independence Day rituals clearly demonstrates
that, while interpretations of meaningful action are multiple and contested, they are mo-
tivated by material experience and mediated by the dynamics of social power which do not
enable all participants to author or to read texts in the same way.
Moore's objections have been echoed in a slightly different form by those who
point to the flattening effects of the textual metaphor, which fails to distinguish
between knowledge and experience because of the assumption that meaningful action
merely represents experience. Such an approach fails to look at the way in which action
constitutes experience.11 Even stylised practices such as rituals are not simply occa-
sions for the expression of relationships but are rather the events in which these
relationships are constituted. Ricoeur does recognise that, due to the intrinsic poly-
semy of communicative forms, interpretation simultaneously involves the con-
textualising processes that reduce meanings and the creative metaphoric extensions
through which innovation occurs. Interpretation could thus simultaneously comprise
a, reductive act of contextualisation and the innovative extension of previous inter-
pretative possibility.
The second set of objections to the textual model for social practice is concerned
with the fact that such models are so firmly language based. The textual model relies
on the metaphor of a message to be read and has no consideration for the non-expres-
sive, non-representational aspects of social life. Criticisms of this kind acknowledge
that anthropologists are deeply involved in textual practices, since our analytic work
implies the objedification of social relations which thereby become susceptible to
interpretation.12
My interest in the motivating force of memory is that it can embrace both discur-
sive and non-discursive practice. A focus on memory enables us to consider the links
between knowledge and experience and the simultaneity of interpretation and creativ-
ity. Representational acts are embodied practices which then exist as latent memories
for the generation of future events. Thus mindful of the possible analytic separation of
textual and non-textual practice, I will now look more closely at the role of memory in
articulating knowledge and experience in the Independence Day rituals of Ocongate.
The official commemoration

The commemoration of nationhood is a phenomenon that dates back to the nineteenth
century. As the new nations of the world emerged, a consciousness of their existence
was instilled in their populations through symbolic practices that allowed people to
imagine themselves as part of an extended community of citizens, an identity entailing
new lines of demarcation between self and other. Symbols of nationhood, such as flags
and anthems, were the concrete manifestations of the imagined communities which
required that the myriad inequalities and differences that exist between people be
forgotten in the project of abstracting that sense of sameness on which the nation's
integrity depends.13
Independence Day in Peru is a national holiday. All public institutions are closed,
and people's energies are redirected to the enactment of particular rituals that com-
memorate the emergence of the modern state. In Ocongate, an Andean village of some
1,500 people in southern Peru, people were required to paint their houses in the colours
of the Peruvian flag—red and white. Any household which had not repainted those
walls that constitute the public space of the village was liable to a fine, levied on the
morning of the 28th. Town hall officials went round checking to see that the houses
were painted, that the streets were clean, and that each house was hanging a Peruvian
flag from its windows. The official programme of events for the 28th, publicised over
loudspeakers in the market and in the central square, was the same year after year. On
the night of the 27th, one of the hillsides that frame the village was lit up. One hundred
old cans filled with kerosene-soaked rags were arranged to inscribe the words "viva
el Peru" (long live Peru) across the dark slope. At 5.00 A.M. the following morning,
the guardia civil, the civilian police, fired a 21 -gun, or rather -pistol, salute. The houses
and streets were inspected during the morning, and the main activity of the day began
at about noon with the desfile, a parade or march past of certain sectors of the
community, particularly the school children and state employees. This parade entered
the village square to the applause of the gathered townspeople, accompanied by the
school's brass band and records of military marches. As each group appeared, a town
hall official, acting as animator of the event, announced their identity over the
loudspeaker.
The goose-stepping marchers saluted the assembled authorities and then lined up
facing the town hall. When all were gathered, the local police raised the Peruvian flag.
With hats off and hand on heart, people accompanied the crackly record of the national
anthem or stood looking suitably solemn until the moment came to shout "viva el Peru!
viva!"
The schoolchildren then performed various "numbers"—folkloric dances or poetic
recitations—and both the mayor and the district governor made speeches. Finally,
there was a general invitation to drink a toast in the town hall and to bear witness to
the minutes of the foregoing activities. These minutes were prepared in advance by
the secretary to the town hall, who meticulously copied them out from the minutes of
the previous year, carefully reproducing this official account of the day's significance
to the village.
The mayor made another short speech, called on the secretary to read the minutes,
and then proposed the toast. Again, all these proceedings were broadcast over the
loudspeaker attached to the balcony of the town hall. All present stood and raised their
glasses, and then one by one filed up to the mayor's desk to sign the minutes. After a
few glasses of sweet "champagne", the local authorities produced a case of beer and
talk turned to local concerns and particularly to the forthcoming bullfight that was to
be celebrated on the following day.
It is not difficult to see how participation in the ritual practices described here
might invoke a particular historical consciousness that both validates the present
existence of the state and provides tacit support for its future projects. Painting the
houses and hanging out the flags involves considerable time and some expense. The
walls are first whitewashed, then edged with red and some people paint flags or words
as further embellishment. The flags that each household must display have to be large
and sewn in cloth. Those who try to get away with sewing red and white plastic are
fined for disrespect. A statement of national identity is thus literally inscribed on the
most intimate social space in the village, the individual family dwellings. Furthermore,
it is inscribed indelibly. The flags come down after the festival, but the red and white
houses and their national slogans remain as permanent testimony to the existence of
the nation.
The communal singing of the anthem, the public raising of the flag, the signing of
the minutes, and the parades themselves also create an image of unity and discipline.
They are ritual actions that in the moment of their enactment constitute those present
as a collectivity and also link them to all the other villages and cities in the nation who
are performing the same acts at the same time. In these aspects of the ritual, the state
is produced as a concrete entity. This national unity is invoked in the words of the
anthem and of the speeches. Thus people sing—and here I paraphrase the national
anthem without the strange grammatical distortions that the music requires:
We are free, may we always be so. The sun's rays will die before we renounce
our solemn vow which the fatherland has raised to eternity. Through the ages
the oppressed Peruvian dragged the chain of a cruel servitude, through the ages
he wept in silence, in humiliated servility basely did he raise a sacred cry. We
are free, may we always be so ... .14
Similarly, an extract from an official speech made in 1987, but typical of many I
have heard on such occasions, reads as follows:
The central commission of festivities for the 166th anniversary of national

independence has given me the responsibility of speaking these words, to offer
you in my name and in that of the commission, this commemorative declaration,
on the very day that we celebrate once more the historic date of our independence
from the Spanish yoke. In my capacity as first political authority of the district
of Ocongate, I offer the following words. It simply remains for me to offer a
moment of reflection so that in this way we can determine to continue working
for the integral development of our district. As Peruvians, who love Peru so
much, we will contribute to the development of the country, beginning with our
own district.15
In this speech, the district governor exhorts people to unity appealing to their
shared love of the nation. Other speeches traced the chronology of events more
explicitly and in the very particular style of public oratory associated with literacy and
prestigious hispanic knowledge. The authority of the following account is constructed
through the use of this style and its copious citing of names, dates, and figures, as well
as the use of direct quotation:
In that year of 1821, on this very date, national independence was proclaimed in
the city of the kings, today Lima, capital of Peru. San Martin entered Lima on
12th July 1821, accompanied by a small escort. He was received by the Marquis
Montemira, colonial governor of the city, and on 14th July he sent a message to
the court in which he invited them to proclaim independence, in a full assembly
of the court, under the presidency of the Count of San Isidro, permanent officials,
titled by Castille, members of the ecclesiastical council, military orders, and the
notable residents of the city. More than 3,400 of them signed with full flourish
the Act of Independence in a solemn ceremony on 28th July, amidst popular
jubilation on the platform in the main square of San Martin. Independence was
proclaimed with these celebrated words: "From this moment Peru is free and
independent by the general will of all the people and through God-given
justice".16
The speeches concentrate on the common past of all Peruvians just as the national
anthem articulates their common freedom. The parades represent the state as an
effective source of power. The order in which people file through the village square
implies a particular categorisation of the population into groups which all refer in some
way to the agency of this power. The main body of the parades throughout the nation
is composed of schoolchildren, all dressed in the same grey and white uniforms and
carrying notebooks. In Ocongate, they were followed by the adult literacy classes
whose presence further affirmed the state's connections with certain forms of knowl-
edge. After them came the local authorities, the forces of order and administration in
the district, the people through whom government money and food aid is channelled,
the people who impose a standard of morality through the sanction of the courts, and
the police force. After them the workers from the health post, then the breadsellers,
parading as a group in white overalls, indicating the imposition of health controls.
Finally the PAIT (Programa de Apoyo al Ingreso Temporal, or Seasonal Income
Support Programme) workers, employed by the state on government projects as part
of their policy to bring work and wages to rural areas. The identification of these
groups was directed by the words of the town hall animator:
In these moments the women of the association of bread-bakers of Ocongate are

marching by, the truly hardworking women who with their efforts feed the
village. .. . The directors who work in our primary education centre are also
coming with their hearts bursting and overflowing with joy and filled with
patriotism they march in a martial fashion past the official stand. ... Dear public,
we are seeing how the children from the private school of the education centre
of Cunicunka are participating in this patriotic festival and are performing their
triumphal civic patriotic march past, etc.17
These various groups cohere as an army, an image which they themselves are well
aware of. The schoolchildren are drilled for several weeks by the police force, even
the kindergarten kids make wobbly attempts at the goose-step, and some of the
schoolchildren, boys and girls, carry wooden machine guns as part of their festive
attire.
The dominant or intended vision of Peruvian history as invoked in this official
ceremony is an extremely optimistic one. The state is represented as a liberating
institution that provides work, food, health, wages, order, discipline, and coherence.
Democracy is emphasised as the people make a series of statements about their sense
of themselves as part of this wider community. The state thus consists of its citizens
and constitutes their freedom. Furthermore, it is important to emphasise that it is
through the ritual that everyone in the village is experientially involved in constructing
this image of the state. Even if they do not march, they at least paint their homes and
put up the flag, and thus provide the physical setting that lends meaning to the parades.
This act of commemoration thus operates on two levels. In one sense, it concerns
a distant past which nobody present experienced. The rituals present this event to
subjective experience in terms that depict the state as a force of effective and desirable
historical agency. The state requires people to participate in a series of actions that
might retrospectively constitute a collective reminiscence, through a symbolic state-
ment of national unity, of the original act of the declaration of independence. People
are thus forced to collaborate in the telling of a particular version of their past, the
glorious emergence of the Peruvian Republic. However, a further step is needed to
connect this creative act of liberation to the contemporary presence of the state.
An event cannot be meaningfully commemorated unless it in some way resonates
with the subjective experience of those involved in its commemoration. It is thus
important to note that, although the original event lies outside the experience of any
of those present, the terms of its remembrance, the symbolic elements of the ritual,
resonate with the daily lived experience of all those taking part. The displays of
patriotic and civic sentiment on the occasion of Independence Day are simply a large
scale orchestration of smaller utterances of this kind made throughout the year. I have
mentioned how the painted houses provide a permanent if fading link between one
year's celebration and the next. Every day, at noon and at midnight, the Peruvian
national anthem is broadcast over all radio stations. Every Sunday in the larger cities,
schools take turns in parading in the square and saluting the national flag. It is most
particularly in the schools that these nationalist sentiments are fostered. Here the ritual
is enacted on a small scale every Monday morning throughout the nation, and it is in
this social space that children are taught about the importance of the independence of
Peru, an event that is not remembered in the oral tradition of the region. The continual
repetition of these symbolic practices in the everyday lives of the people of villages
such as Ocongate enables the state to use an event as distant and removed from
contemporary experience as the signing of a document by a Colombian army general
in 1821 to motivate national consciousness and to have the kind of emotive impact on
people's lives that I have described. In the act of watching and interpreting the
groupings of the parade, people are directed to connect experiences and to remember
the ways in which the state is active in their region. Schoolchildren are connected to
the state's investment in the future, the PAIT workers with employment and wages,
the local authorities with order and administration.
The ritual creates a context of consensus and authority. The commemorated event
is presented as self-evidently significant and the particular narrative of the pastthrough
which it is presented naturalises the continuing presence and social agency of the state.
It could be argued that this event merely reproduces the effect of a centralised
education system. The difference, however, between the Independence Day presen-
tation and the history lessons in schools is that the former uses the experiential and
consensual nature of the ritual to conceal the contingent and partial nature of the
particular connections that its narrative implies.
Before I go on to look at how commemoration rituals can also lead to a questioning
of the narratives that are presented in them, I wish finally to emphasise the symbolic
power of these rituals by examining how the state has used them to inaugurate
important policy shifts and changes in the practice of government. New governments
are always sworn in on 28th July. During their term of office, this is also a day that
has frequently been chosen to announce quite dramatic measures that will affect the
lives of all Peruvian citizens, and which usually belie the notion of a democratic state
in which policies supposedly emerge as the negotiated outcome of discussions with
all sectors of the population. Thus, President Alan Garcia announced on 28th July
1985 that Peru was not going to repay the national debt according to the terms of the
International Monetary Fund; two years later, again on 28th July 1987, he nationalised
the banks. The year 1990 saw the swearing-in of President Alberto Fujimori, who was
elected to protect the peasant and shanty town populations from the stringent mone-
tarist policies of his rival, Mario Vargas Llosa, and who promptly announced an
economic package of undreamt-of severity that terrified the poor and middle classes
alike. I mention these recent political events to stress that the Independence Day rituals
invoke a sense of continuity that can be used to legitimate future innovation. The sense
of continuity is linked to the power of the state itself, its importance as the central
institution of government whose offices can be occupied by changing individuals, each
claiming that they will realise the potential of state administration and provide
effective government for their nation.
However, contemporary Peru is wracked by civil war as the army attempts to
protect the very existence of the state itself against the challenge of various armed
guerrilla groups. The pervasive awareness of a split between state and civil society
indicates that consensus is by no means the inevitable outcome of these rituals,
although their symbolic power is such that guerrilla groups find it necessary to
intervene in them directly.
At this point I want to look at the more detailed ethnography of how the narratives
that emerge in the rituals of commemoration are contested in the interpretative
processes. We are dealing here with a gap between representation and experience.
Memories are subjective abstractions of human experience. To remember a particular
past is to forget another. Thus, while the state can attempt to control the representation
of its past and thus validate its actions in the present, the past experience of those
involved in the rituals will also be brought to bear on their interpretations. I have shown
how many aspects of people's daily lives might be activated in this sense to provide
the interpretation that effectively constitutes state power; nevertheless, they also
provide a dimension of interpretation that the state cannot control. Thus, the enactment
of the Independence Day parades across the nation does not imply a homogeneous
understanding of, or response to, these ritual practices. The attempt on the part of the
state to construct an objective history through these rituals is both made possible and
challenged by the ways in which subjective reminiscence intervenes in the process of

interpretation.
The parades give a good example of this process. While the parades provide a very
positive image of state agency in the locality, this particular representation of the
community of Ocongate is also very problematic. Many of those who live in the district
are not represented at all in this apparent cross-section of citizens. Peasants, shopkeep-
ers, traders, families, political parties, and migrants are nowhere grouped and named
as such. Yet these are all extremely important social categories in the daily lives of
the people in this area. Furthermore, the work groups that are represented in the
parades are as likely to invoke feelings of exclusion as feelings of support for the
positive agency of the state. The workers on the PAIT scheme, for example, were
widely thought to have been recruited along nepotistic lines of favouritism, whereby
local officials looked after their own to the exclusion of the majority of local people.
The breadsellers dressed in white also provoked memories of the struggle that occurred
between these workers and the local authorities. The latter tried—unsuccessfully, as
it happens—to enforce standards of hygiene that were seen as totally arbitrary by the
women concerned. They saw the obligation to wear white overalls as a step toward
trying to control their work practices, and their resistance to the move was effective.
As a result, they now only wear the overalls for the Independence Day parade. Similar
alternative interpretations exist for all the possible positive messages that the repre-
sented groups carry: there are those who have never been to school, who left illiterate,
and who do not speak Spanish; there are those who were made ill at the medical post
or whose relatives died from lack of skilled attention; there are those who have been
harassed by state employees, maybe even fined that morning for not having the right
flag; there are those who are extremely poor and who do not have food for their
families. For many, the state narratives are so irrelevant to their daily experience that
they stay away from the celebrations. Others make fun of them, even when forced to
attend. Goose-stepping can easily become a kick in the bottom for the person in front
of you(Gose 1986).18
The official ceremony affirms a particular history of the Peruvian state, backed up
by the consensus of experiential participation in the rituals. Villagers enact an image
of the state of which they are themselves a part, a state which provides essential
services, a state which is disciplined and united. The real problem for the Peruvian
state is that this image, despite the power of its symbolic affirmation, simply does not
ring true. Furthermore, this contradiction with other areas of lived experience finds an
accessible mode of articulation in alternative understandings of the agency of state
power, kept alive in the oral accounts and ritual practices that present the state and the
locality as distinctive entities. In these accounts, local people depict state power as
outside power, unpredictable and difficult to access. It is associated with the conquer-
ing forces of Spanish-speaking foreigners, destroyers of local autonomy through the
incursion of abusive institutions, particularly the army and the police, and with the
implementation of arbitrary and destructive economic policies characterised by ne-
glect of local needs, particularly in terms of schools, medical provision, roads, water,
and electricity; finally it is seen as corrupt, where state officials look after their own
and take from others for their own ends.
The clearest or most articulated counter-statement made by the people of Ocongate

to the official history of the Peruvian state occurred in the bullfight, the most popular
part of the commemoration ceremonies. While many villagers stay away from the
official rituals, nearly everyone attends the bullfight; furthermore, people come from
the most distant settlements of the district. As part of the overall commemorative
event, the bullfight makes constant reference to, and therefore comments on, the
official history of the state. It remembers and re-presents that part of the ritual that
preceded it, and that constitutes the immediate conditions of its production. In doing
so it does not erase this history, but rather renarrates and re-emplots events. The
connections that are drawn in the bullfight produce a different, but not necessarily
incompatible, history from that narrated on the previous day.
The bullfight
The Ocongate bullfight differs in many respects from the Spanish bullfight. There are
no professional bullfighters, and the bulls are not harmed in any way. Ideally, a condor
should be brought down from the highlands above the village to "play" with the bull
in the village square.19 The bullfight is organised by the officials of the town hall who
need the collaboration of the owners of the bulls and of the specialists in catching
condors. These people live outside the village in small isolated hamlets in the puna,
high above the valleys.
Some several weeks prior to the bullfight, representatives of the Town Hall travel
out of the village to request the assistance of these puna dwellers, and they make every
effort to diminish the potential social difference that their status as village authorities
might evoke. They have to demonstrate that they are able and willing to welcome the
bull owners and their families as honoured guests to the village on the day of the
bullfight, and they thus offer gifts of food, drink, tobacco, and coca and show
friendship and affection by drinking and chewing coca with them in their houses.
The bulls that are brought down for the bullfight usually belong to several
individuals who collaborate with each other along lines of kinship and compadrazgo,
under the guidance of the senior male of the family. Two or three days prior to the
bullfight, they begin to round up the animals and, together with a group of contracted
horsemen, the laceadores, they begin the journey down to the village. As they
approach the village, they are joined by young men who are sent out on the day before
the bullfight to assist in the final stages of the roundup.
Catching the condor is a more lonely task. The catcher sets off up to two weeks
before the bullfight, and laying a trap up by the snowline of the high mountains in the
district waits patiently, requesting the mountain spirits to allow the condor to come
down to the village and "play" with the bulls.
In the village itself preparations are also organised along lines of kinship and
compadrazgo. Sponsoring authorities begin to prepare the food and drink and to hire
musicians. On the eve of the bullfight they go round to every house in the village
requesting small gifts, such as articles of clothing or plastic goods, which can be given
to the bullfighters and bull owners. Finally the village square is closed off with wooden
poles to form the "ring" in which the bullfight will take place.
The owners of the bulls and their families arrive in town on the day of the bullfight
itself, the women and senior men usually arriving in a lorry which is sent to collect
them. The condor is prepared for the event. It drinks substantial amounts of wine and
the tips of its wings are tied. In the assembling crowd, puna dwellers in their homespun
bayeta clothing jostle with villagers in Western dress for a good view, cramming
themselves behind the wooden barriers, onto balconies and lorries. The owners of the
bulls are invited up onto the balcony of the Town Hall from where they watch,
squeezed in with the musicians and with friends and relatives of the village authorities.
The bulls are released into the square one at a time, although two or more
sometimes get out at once. Those men who wish to show their skill and daring
demonstrate their passes, using ponchos or even jackets and hats as capes. When the
bull tires or refuses to "play", the laceadores rope it back into the corral or pozo, and
another one is pulled into the square. There is usually less action during the first half
of the bullfight as the bullfighters need to get slightly drunk to build up the necessary
courage to go in and fight. Many men and occasionally a young woman will go in to
face a small harmless looking animal, but few will risk their lives with heavy fierce
bulls or with an agile cow who knows to go for the body and not for the cape. A meal
is served halfway through the event and throughout the day the authorities have to
keep the owners, the laceadores, and the bullfighters supplied with drink. The condor
is released drank into the square and faces the bull alongside the bullfighters,
supposedly using its wings as capes.
At the end of the day the bulls are released to find their way back to the high
pastures. The condor is set free on the following morning during a final farewell
ceremony for both the condor and the owners of the bulls. With ribbons attached to
its wings it walks slowly up the mountain side. When it catches sight of Ausangate,
the dominant snow peak in the region, it is supposed to take flight. If it circles the
village, people know that it has enjoyed itself and may return to play another year.
The connections between past and present that the bullfight invokes are quite
different from those emphasised in the ceremonies of the previous day. The bullfight
appeals quite openly to a source of power that exists outside the realm of the state, and
which resides in the landscape of the locality. This landscape is inhabited by a
pantheon of supernatural forces, among whom are the apu and awki spirits that inhabit
the hills and mountains, and the Pachamama or tierrakuna, the earth powers present
in pastures, roads, paths, fields, and houses. The interaction of these spirits with human
beings is problematic. It is necessary for the regeneration of the local community, yet
it is also dangerous, as the powers are unpredictable and essentially outside human
control. The daily experience of the ambiguous nature of these powers has generated
a similarly ambiguous sense of the nature of the connection between their present
existence and the past events which displaced them from the centre of the human
universe. Were they defeated by the Spanish powers at the time of conquest? Were
they irrevocably weakened by the neglect that religious repression entailed? Has the
Christian god in fact rendered them redundant? Are they still available for effective
intervention in contemporary society? There are no clear answers to these questions.
However, there is the sense that the local population must do all it can to maintain
effective communication with these forces, because the state does not offer a viable
alternative source of regenerative power In and of itself. Much local ritual practice
thus works to combine the possibilities offered by state and landscape, and the
Independence Day rituals are no exception.
As was the case in the official ceremonies, the ritual practices of the bullfight
ostensibly present a very optimistic version of the history and agency of outside power.
The forces of the landscape are depicted as powerful and alive, as available and well
disposed toward the community, and as respected and honoured by the state through
their officials. These positive images are created in the whole process of preparation
and enactment of the bullfight.
Both the condor and the bulls dwell in and belong to a powerful non-human
landscape. The bull has long been associated in mythology with wild, non-human
power, and the condor, so wild to the human, is the domestic chicken of the apu hill
spirits. These animals are strong; they cannot be forced down to the village. Their
presence must be requested, their journey facilitated by their supernatural owners.
Those responsible for rounding up the bulls and herding them down to the village
engage in complex ritual activities involving divination with coca leaves, inspection
of cow dung, major offerings to the hill spirits before the journey begins, and smaller
offerings of coca and alcohol during the journey and in the corral in the village where
the bulls are kept. The catching of the condor involves a similar attention to the hill
spirits, offerings, divination, and, once the condor is caught, careful attention and
respect. Both the owners of the bulls and the herders continue these magical practices
in the village as they attempt to encourage and support their bulls. The owners, for
example, throw hot peppers, alcohol, and oranges from the Town Hall balcony to make
their animals more tierce through the contagious magic of "hot" substances, and the
herders offer alcohol to the bulls to establish a relationship of support. Bullfighters
say that the owners interpret injury and death as a good omen and that they believe
the spilling of blood in the bullfight constitutes an offering to the earth spirits, and can
thus ensure a good harvest and good fortune. The bullfighters also have their "secrets".
They, too, make offerings to the hill spirits for protection and have magic amulets
which they keep on them during the fight to protect themselves.
There are many more ways, apart from this direct appeal to the powers of the
landscape, in which the practices associated with the bullfight constitute a statement
of an indigenous identity that asserts a continuity with a non-Hispanic past. Work
practices are based on traditional reciprocal labour patterns that stress kinship ties; the
bull owners dress in their most elaborate homespun clothes, a statement that is
reinforced by the traditional music and dances performed as part of the event. These
practices must also be understood in terms of the very different symbolic statements
made in the same physical space on the previous day—through the marching parades
of schoolchildren, the brass bands, and the records of military music. The Town Hall,
for example, is converted from the space in which elaborate speeches were made in
florid Spanish to the scene of a wild party of unrestrained drunkenness and where all
talk is in Quechua.
Nevertheless, this symbolic reversal does not negate the power of the state. For a
start, it is the state authorities who organise and sponsor the bullfight. The bulls
themselves not only represent untamed wildness but are also a substantial source of
material wealth, a wealth that confers status on their owners in the value system
supported and upheld by the capitalist market. The bulls are owned by people as well
as by the apus, and the owners of the bulls are well aware of this status when they are
welcomed into the village by the elected state officials. A similar symbolic ambiguity
surrounds the condor. Unlike the bulls, the condor has no single human owner; as a
protected species, condors are in fact the property of the state, and people talk openly
about the condor belonging to the state. This status is reflected in the way they behave
toward it. The respect accorded the condor is analogous to that extended to a powerful
visitor from the city, a respected outsider with the sensibilities of an urban Hispanic
world. The condor has refined tastes and does not eat and drink to excess, behaviours
that would be seen as cultural imperatives for adequate attendance on indigenous
dignitaries. One of the village authorities informed me that the condor should really
only mix with high-status officials in the village: "I've been noticing that anybody can
hold onto the condor nowadays, even get their photo taken with it, but it didn't use to
be like that. Before it was very sacred, the only people who were authorised to hold
onto the condor were the mayor and the governor, nobody else".20
Nevertheless, however much the town authorities would like to claim the condor
as their own, they cannot actually gain access to the bird without the help of the
specialists and ultimately the hill spirits. The important point to emphasise here is that
neither the bulls nor the condor can be associated unambiguously with either an
indigenous or a Hispanic tradition. The bullfight distinguishes these "traditions" by
contrasting the powers of the landscape with those of state institutions. However, the
ritual itself also brings these two possibilities together to constitute a more totalising
environment. In the bullfight, the power of the landscape is paramount, but it is linked
to that of the state—not in opposition, but in a complementary relationship offering a
positive statement about the collaborative possibilities of indigenous and state
power.21
Thus the bull and the condor in the village square can embody the power of the
landscape, contained within the village but not reduced or diminished by such
containment. The danger and strength of outside power is experienced in the bullfight
as real and tangible. The power of the landscape cannot be defeated in the confronta-
tion between bull, condor, and human. Human beings enter into a social interaction
with the outside powers. The interaction involves a confrontation between the bull
owners and the bullfighters, but more important than the confrontation is the intensity
of the interaction. A failed bullfight does not result from injury or death, but occurs
when the bulls show no interest in the "game", and interaction is thus refused. One
man expressed it to me thus:
In the bullfight we sometimes see what we want to see: Will the bull play or not,
or will it perhaps catch somebody? And sometimes in some bullfights we see
the bull catch someone, sometimes even kill them. These things happen in the
bullfight, and this is what makes people enthusiastic. That is why Ocongate is alive
with its bullfight. Before, Ocongate was an insignificant place; now it's a better place
with its bullfight.22
Thus, although the ritual practices of the bullfight are not explicitly discursive,
and unlike the official ceremonies of the previous day do not provide a continuous
commentary of intended meaning, they do nevertheless allow the construction of
a particular set of connections between present and past events. They invoke a
non-Hispanic world, a source of creative power that establishes a continuity that
stretches back beyond the event of independence, which acknowledges the exis-
tence of the state, but stresses its incorporation into an enduring indigenous
tradition. It is in this sense that the bullfight could be said to renarrate the events
of the previous day and to establish different connections with the past, to the
extent that the event of independence itself is rendered insignificant. The significant
axis of history is redefined as the moment of encounter between indigenous power
and the state.
However, as with the official commemoration of independence, the ritual
practices of the bullfight also generate contradictions between the representation
of the past from the perspective of the present and the memories of past experience
that subvert this performative control. If we consider much of the ritual practice
of the bullfight, there is considerable emphasis placed on preventative, placatory,
and exhortatory measures. Interaction with the powers of the landscape is extremely
problematic and by no means entirely under the control of the local population.
The offerings and the magic performed by all those who plan any direct
interaction with the bulls and the condor are not necessarily effective, and people
are aware that their lives are at risk. This risk should be understood in the wider
context of the general preoccupation about the reluctance of the powers of the
landscape to communicate with human beings. I was told that these powers never
talk to people directly. Might not this silence also imply that they do not listen,
just as the state has consistently not listened to the petitions and requests that have
been sent from the village? Those who participate in these events most directly
know from past experience that the ideal of producing an intense interaction with
these powers in the ritual context might be achieved at the cost of their own lives.
They remember the deaths of others; their scars remind them of their own
vulnerability.
These memories of pain and fear also motivate the interpretations of the families
of those who face the bulls. While the general atmosphere of the bullfight is one
of animated festivity, there are also many spectators whose attention is focused on
a spouse, parent, or sibling, ready to prevent them from entering the square if
necessary. It is common to see women and even children, shouting, crying, or
running into the square to drag away a man fired with an alcohol-induced bravado.
The bullfight is also pervaded with an air of uncertainty. People never know
until the last minute whether the bulls will be brought down from the pastures,
whether the condor will be caught. Sometimes they elude capture, and sometimes
when the bulls arrive in the village they will not "play". In my experience, the
condor never "plays". I have never seen it use its wings as capes, nor does it fly
to greet the spirit of the snow mountain Ausangate with its ribbons flowing behind
it. It usually got somewhat knocked around in the square, having been made too
drunk to defend itself, and was fairly quickly removed to a place of safety. On
the following day in the farewell ceremony, it walked slowly up the hillside,
stopping only to pull off the coloured ribbons until it was a mere speck on the
horizon, presumably waiting for the hangover to wear off before it flew back to
the place it had come from.
38 TEXTUAL1SING HISTORIES AND IDENTITIES
In these circumstances, it is not surprising that there is often a sense of failure

surrounding the bullfight. It is never as good as it should be. There is nostalgia
for past bullfights, the ones where everything did turn out just as it should. It is
interesting here to consider Rosaldo's discussion of nostalgia, in which he com-
ments that nostalgia is an emotion usually evoked in an attempt to establish one's
innocence while talking about what one has destroyed (Rosaldo 1989). Rosaldo
was referring to ex-patriot colonialists, but his insight is relevant here, especially
in the context of the Independence Day commemoration which celebrates local
involvement in a modern state that has only a cosmetic interest in the notion of
an indigenous past. The fascination of taking pictures of themselves holding the
tip of the condor's wing might be part of the same phenomenon: preserving the
memory of an unusual and possibly unrepeatable event in the concrete form of the
visual image.
Finally, the bullfight also reveals the fragility of the local community as an
integrated social unit. In the official commemoration ceremonies, this fragility is
demonstrated by people's absence, their failure to participate in the events of the
day. In the bullfight, the fragility is dramatic and confrontational and directly
related to the importance of drinking in all rituals that address the supernatural
powers of the landscape. In the bullfight, drinking creates a local community—the
owners of the bulls drink with the authorities of the Town Hall, the bulls and the
condor are offered drink, and frequent offerings are poured on the earth or flicked
into the air for the hill spirits. The interaction between the humans and the bulls
is intensified as people dare to enter the square. However, the drunkenness also
highlights the rifts in the human community and belies the unity that the ritual
practices construct. Fights frequently break out, particularly between the bullfight-
ers and the owners of the bulls. I have argued elsewhere that, in Ocongate, drinking
ideally serves an integrative function; it should generate social cohesion and ensure
social reproduction through the integration of human beings both among themselves
and with supernatural powers (Harvey 1987, 1991, 1994). Such groups are actively
created in the initial stages of drinking and then celebrated and expressed in the
later stages of drunkenness. However, the abstractions required to achieve this
sense of group or sameness are problematic. The process is never completely under
the control of either drinkers or non-drinking participants, as people operate with
conflicting models of the group that is being formed. Thus, drinking often appears
to achieve or enforce a particular interpretation of community at the cost of social
relations. Because drunkenness is about creating community from the available
relationships in people's social experience, it is also a time when the fragmentary
nature of "society" is revealed. This fragmentation is revealed precisely as people
are frustrated in their constructive aims and as they voice their understandings of
the contradictions on which the fictions of social cohesion and community are
built. These frustrations are highly visible in the hostilities and local tensions that
are often brought into the open by drunks on the occasion of events such as the
bullfight.
It can thus be seen that while the bullfight provides an alternative history and
one that is potentially more empowering to the local community than that of
the previous day, this version is itself constantly undermined in practice by the
subjective reminiscences of the participants in the ritual.
Conclusion
The commemoration of Independence Day in Peru engages people in both textual and
non-textual practices. As public events, both the official celebrations and the bullfight
produce connections between the past and the present. However, the articulated
historical knowledge or consciousness which emerges from textual practice is mediated
by the creative recuperations of the past that individual memories afford. These
memories, based in concrete material experience, both enable the recognition of
the intended symbolic content of the text and simultaneously work against its
implications.
The hegemonic history of the official ceremonies attempts to establish the decla-
ration of independence in 1821 as the significant moment in the historical process that
defines contemporary Peruvian identity. This was the moment which granted the
freedom that all Peruvians supposedly share today, and which founded the overarching
institution of the nation to which all are tied with strong emotive bonds. The narratives
that depict the importance of this moment of liberation thus contrast the positive
agency of the modern state with the negative oppression of the Spanish colonial
regime. The authority of this narrative lies in the way in which it is produced in a ritual
of such public and widespread consensus. The massive participation in the telling of
this account, along with the implicit collaboration in its continual retelling in the
everyday reaffirmation of its symbolic elements, creates a powerful experiential
synthesis that, as Munn suggests, not only carries past events into the present but
configures certain futures. It is in awareness of this that state officials attempt to
legitimate radical political change by introducing innovation in the context of this
powerful statement of continuity.
I would argue that it is also in this awareness that the Maoist guerrilla group,
Shining Path, expends such considerable effort on the disruption of the Indepen-
dence Day celebrations. Shining Path has consistently disrupted these events,
placing bombs and murdering local state officials. Indeed, the memories of these
attacks are now sufficient for Independence Day celebrations themselves to invoke
the experience of terror and disruption that Shining Path seeks to instill in the
population. People notice it if nothing happens. Such emotions strongly undermine
the narrative that tells how the modern state perpetuates the freedoms established
for all Peruvians by the liberator San Martin. You do not have to sympathise with
Shining Path to experience the contradiction between the words of the national
anthem and the banning of all public ceremonies and parades. Shining Path subverts
the memory of state symbolism in local consciousness and thus begins to appro-
priate the public images of state control. They practice a politics of erasure and
of retextualisation. On the one hand, they produce new innovative, confrontational
narratives as they wipe out the patriotic writings on the houses and replace them
with revolutionary slogans or rearrange the kerosene lanterns to inscribe their own
symbolic repertoire on the hillsides. On the other hand, they may simply produce
spaces and absences that disrupt the continuity of the official narratives. Instead
of drawing innovative connections between past and present, they work to sever
the connections that the state has produced.23
At particular historical junctures, such as that of contemporary Peru, the challenge
to official memory can thus be extremely violent and explicit. The inability of the
contemporary Peruvian state to deliver in the terms of its own historical projections
has led to a situation in which counter-memory no longer simply comments on official
memory but begins actively to eradicate it.
This challenge to the state is dramatic and highly confrontational. It is also
very public and pervasive. The alternative accounts more commonly generated in
Ocongate are less subversive. As I have discussed, the redefinition of the significant
axis of history which the ritual practices of the bullfight produce, can and does
co-exist with an account of state history that accepts its legitimate continuity,
although it doubts its efficacy. In this account, I have produced the bullfight as a
public text, drawing connections and making explicit the links that enable this
interpretation. In classic anthropological style, I have moved between what is
explained and what is taken for granted, the explicit and the implicit, attempting
to make visible that which an outsider might not expect to see. However, I do not
want to over-emphasise the salience of this interpretation. Strathern has recently
reminded us that our concept of culture compels us to make culture explicit to
ourselves and leads to our privileging of the representational (textual) aspects of
social practice (Strathern 1992: xvii). In this vein, anthropology operates to
disclose, order, and contextualise the practices of others, to reduce the ambiguity
of polysemy, and to produce a coherent and plausible text. My argument here is
not that such a text might not be recognised by the people of Ocongate, nor that
its effects are negligible. They, too, have a concept of culture as performative,
representational practice. However, we should also be mindful of Taussig's warn-
ings as to the nervousness of our systems: he points out how order, stability, and
structure also operate through normative disorder, through secrecy, silence, and
paranoia (Taussig 1992). The force of disorder is not a state monopoly. Disartic-
ulated subjective reminiscences can challenge hegemonic histories. Many of these
alternative interpretations of the significance of past events are never made public,
never move from the space of (inarticulate) memory to the articulate abstraction
of an alternative historical narrative.24 Textual models cannot account for this
dimension of social life, the memories that disarticulate text from experience.
The state undoubtedly produces a text on 28th July in the commemoration
ceremonies. The bullfight and the activities of Shining Path both erase and partially
renarrate the terms of this account. However, while it may appear that beyond
these forceful public statements there are as many meanings as there are inter-
pretative acts, we should not be satisfied with pluralist models that fail to
distinguish the agency of the state from that of a drunken peasant. If we consider
the role of memory in interpretation, it is possible to appreciate the simultaneity
of normative contextualisation and disruptive creativity as people's material experi-
ence enables them to recognise, but also to disarticulate, themselves from the texts
which they themselves produce.
ACKNOWLEDGMENTS Earlier drafts of this chapter were presented to the Department of

Anthropology at the School of Oriental and African Studies, London, in 1990; in 1991, to the
Department of Anthropology at the University of Manchester; and to the symposium on
Textuality in Amerindian Cultures held by the Centre for Latin American Cultural Studies,
London. I am grateful to the participants of these seminars for their comments and suggestions.
I am grateful also to Peter Getzels and his assistant Benigno Diaz for their collaboration in the
recording of interviews on the nature and meaning of the bullfight which were carried out as
part of a film project on the Ocongate bullfight (see Getzels et al. 1990).
Notes
1. In the light of the polemic over the use of the term "encounter" to describe the nature
of the contact between native Americans and Europeans in 1492, it is worth pointing out that
my use of the term here is deliberate and refers to the explicit avoidance of the use of the term
"conquest". As will be explained, local narratives are ambiguous about whether the Spaniards
did or did not conquer/defeat indigenous powers.
2. The fieldwork on which this account is based was carried out between 1983 and
1988, a time of intense guerrilla and army activity in the central and southern Peruvian
Andes, although during that period the village of Ocongate was not the site of any such
confrontations.
3. See Terdiman 1989 and Ingold 1994.
4. Thomas Laqueur has written a fascinating article on memory and naming in the
Great War which looks at the adoption of new techniques of memory in the mid-twentieth
century. His general point is to stress the historical specificity of commemorative practice
(Laqueur 1994).
5. See for example Tonkin et al. 1989, Hastrup 1992, and Gillis 1994.
6. See for example Theweleit 1989 and Taussig 1990, 1992.
7. There is a widespread oral tradition in the Andean region which tells of the three ages
of mankind: the age of God the Father, the age of God the Son, and the age of the Holy Spirit
(Gow and Condori 1982, Nunez del Prado 1970, Ossio 1973, Urbano 1980, Wachtel 1977).
People presently live in the age of God the Son. This Christian millenarian chronology meshed
with Andean millenarian beliefs summed up in the notion of pachakuti (the world turned upside
down) which refers to the Inka belief that contemporary worlds are inevitably overturned to
make way for new world orders, the transformation occurring through the forces of nature (Gow
1976, Ortiz 1973, Ossio 1973, Wachtel 1977). The sense of ambiguity in local narratives
concerning the historical source of civilisation reflects these separate traditions. At times the
narratives are quite explicit that the coming of the Age of the Sun coincided with the emergence
of the Inka state; at other times the dawn of civilisation is associated with the advent of
Christianity. In this latter case, the association between Christianity and Spanish rule is left
implicit, as is the implication that the Inkas were therefore living in pre-civilised times.
8. See for example Allen 1988 and Harvey 1987.
9. See Keesing 1987, Fernandez 1985, and Moore 1989. A key article by Ricoeur in this
respect was Ricoeur 1971.
10. Moore is summarising Ricoeur's concept of distantiation as set out in Ricoeur (1981:
134-203).
11. For elaboration of this point, see Friedman 1987.
12. According to Wagner (1981: 8-10), this is all we can do as anthropologists, and this
is what differentiates our practice from that of those we describe. In a somewhat different way,
Bloch (1991) has also argued that much of what anthropology is trying to understand is precisely
this non-articulated, non-lingustic knowledge, what is commonly designated as the taken for
granted or common sense.
13. For further discussion of these issues, see Anderson 1983, Segal 1988,Spivak 1988,
Fox 1990, Segal and Handler 1992, and Lofgren 1993.
14. The Peruvian National Anthem:
Somos libres, seamoslo siempre, seamoslo siempre

Antes niegue sus luces, sus luces el sol
Que faltemos al veto solemne,
Que la patria al eterno elevo.
Largo tiempo el peruano oprimido la ominosa cadena arrastro,

condenado a una cruel servidumbre.
Largo tiempo, largo tiempo,
Largo tiempo en silencio gimio,
Mas apenas el grito sagrado.
La humillada serviz levanto,

Serviz levanto.
Somos libres, seamoslo siempre . . .
15. The original words of this speech were as follows:
Es un cargo de parte de la Cornision Central de festejos del 166° Aniversario de

la Independencia Nacional; hace que mi persona se dirija la palabra para
ofrecerles a nombre mio y el de la Comision la presente manifestation con-
memorativa, justamente por cumplirse un ano mas la historica fecha de nuestra
Independencia del yugo espanola. En mi condition de la primera autoridad
politica del distrito de Ocongate a mas de mi palabra ofrecimiento. No me queda
sino a un momento de reflexion para que mediante ello nos propongamos seguir
trabajando por el desarrollo integral de nuestro distrito. Como peruanos que
amamos tanto el Peru contribuiremos al desarrollo del pais, empezando desde
nuestro distrito. (Transcription by Janett Vengoa de Oros)
16. This speech was made by a resident of Ocongate in 1987:
Aquel ano de 1821 como en esta fecha fue proclamada la Independencia

Nacional en la ciudad de los reyes, hoy es Lima capital del Peru. Hasta que San
Martin entro en Lima el 12 de Julio de 1821 acompanado por una pequena
escolta, lo que fue recibido por el marques Montemira gobernador colonial de
la ciudad para que el 14 de Julio envio al Cabildo una comunicacion por la cual
le invitaba a proclamar la independencia reunida en el Cabildo baj o la presidencia
del conde de San Isidro, regidores perpetues, titulos de Castilla los miembros
del Cabildo eclesiastico, los ordenes militares y los vecinos notables de la ciudad
y luego rubricaron en la acta de la Independencia con mas de 3,400 en un acto
solemne el dia 28 de Julio en medio de jubilo popular sobre el tabladillo en la
plaza mayor San Martin proclama la Independencia con las celebres palabras:
"Desde este momento el Peru es libre e independiente por la voluntad general
de los pueblos y por justicia que Dios defiende. (Transcription by Janett Vengoa
de Oros)
17. The following words were broadcast over the town hall's loudspeaker into the square:
En estos momentos esta haciendo su paso las senoras de la asociacion de

panaderas de Ocongate, verdaderas senoras trabajadoras que con su esfuerzo
alimenta al pueblo. .. . Los directores que trabajan en nuestro centra educativo
inicial vienen tambien con el corazon enchido y lleno de alegria y llena de
patriotismo hacen paso marcialmente marchando al frente de la tribuna oficial.
. .. Amable concurrencia estamos viendo como los ninos de la escuela particular
del centro educativo de Cunicunka, participan en esta fiesta patriotica. Asi de
esta manera esta haciendo su paso civico patriotico triunfal, etc. (Transcription
by Janett Vengoa de Oros)
18. Gose's account of the Independence Day parades in Apurimac emphasises the way in
which the "didactic and expository approach to performance", which the parades and speeches
require, contrast with other expressive performative genres, which are in fact usually taken more
seriously (1986: 172-73).
19. The English term "bullfight" is somewhat misleading as it overemphasises the
confrontational aspects of the encounter between bulls, men, and condors. In Quechua the term
is turu pukllay, in Spanish corrida. The Quechua term describes the activity as play (turn 'bull',
pukllay 'to play"), and this concept is reinforced in the ways that people talk about the bullfight.
A condor catcher remarked in 1988: "The old condor is also going to go as a player. Old condor,
you are going to play well in the square of Ocongate, you are going to cape well with the bulls"
("El viejo condor tambien va a llegar de jugador. Condor viejo, pues vas a jugar en la plaza de
Ocongate bien vas a capear bien a los toros"). This concept of play is taken up in the idea that
bulls, condors, and men are supposed to enjoy their encounter and return to play again in
subsequent years.
20. This conversation was recorded in 1988: "Ahora mas bien he visto ya cualquier
persona agarra y hasta fotos se hacen tomar, pero antes no era asi. Antes era muy sagrado, los
unices que eran autorizados para agarrar al condor eran el alcalde y el gobernador nada mas"
(Transcription by Janett Vengoa de Oros).
21. It should be noted that this interpretation of the Ocongate bullfight diverges in
important ways from other interpretations of Andean bullfights which depict the encounter as
a confrontation between the indigenous and the Hispanic, the bull standing as an image of
Spanish power and domination. Such accounts, popularised in films such as Luis Figueroa's
classic "Yawar Fiesta", or Paul Yule's "Our God the Condor", refer primarily to the Yawar
Fiesta (Blood Festival), in which the condor is tied to the back of the bull and the bull is
eventually killed. The Yawar Fiesta is also the theme of the novel of the same name by Jose
Maria Arguedas (Arguedas 1941). My criticisms of these analyses are elaborated in Harvey
1992 and illustrated in "The Condor and the Bull" (Getzels et al., 1990). Even taking into account
the ethnographic differences between the Yawar Fiesta and the Ocongate bullfight, I am critical
of the tendency in these films and novels to fix symbolic values into simple oppositions such
as bull/condor, Hispanic/indigenous, and of the tendency in Andean ethnography to conflate
the indigenous or non-Hispanic: with the Andean. At this point, I would also stress the strong
differences between the contemporary Ocongate bullfight and the Spanish-style bullfights
which also take place in Peruvian cities. Analyses of these latter confrontations, whether they
focus on gender and honour or social class, tend to rely on the fact that the bull is killed. See
Douglas (1984) for an overview of such analyses.
22. These comments were made by the man who had caught the condor in 1988:
En la corrida a veces vemos lo que queremos ver, si juega o no juega el toro, o

si tal vez lo pesca a la gente, y como a veces vemos en algunas corridas que lo
pesca a la gente tambien el toro, y hasta a veces lo mata, esas cosas pues pasan
en la corrida, con eso se calienta el pueblo, por eso pues Ocongate vive siempre
con la corrida, antes Ocongate era insignificante y ahora es un mejor pueblo con
su corrida.
23. See Taussig 1992 for further elaboration on the workings of terror through processes
of fragmentation and the generation of absence and disconnection.
24. It is interesting in this respect that the words of the drunks, who are most likely to
produce public discursive countermemories, are themselves forgotten as the social norms that
allow for their articulation also require that they are not received as significant (see Harvey
1991).
3
LUIS MlLLONES
Saint Rose through the

Eyes of Her Devotees
From Flower of Lima to
Mother of Carhuamayo
Bc Dorn in Lima, Isabel Flores de Oliva was the most important of all the saints to
emerge from the Viceroyalty of Peru. Her period (from the end of the sixteenth century
to the second half of the seventeenth) is notable for its bumper crop of holy men and
women (Saint Toribio of Mogrovejo, 1538-1606; Saint Martin of Porras, 1579-1639;
Saint John Macias, 1585-1645; Saint Francis Solano, 1549-1602; Pedro Urraca,
1583-1657; Francisco Camacho, 1629-1698; Nicolas Ayllon ?-1677; Francisco de
San Antonio, 1593-1677; Ursula de Cristo, 1629-1666; and others). Not all of the
aforementioned attained canonisation: some remained as blesseds; others failed even
to get that far. Even so, the number of persons who felt singled out by God, and
achieved official recognition for it, is considerable. But besides this official list, there
was another route to acceptance which did not necessarily coincide with the church's
verdict in any given case—namely, the popular fervour that viewed the holy persons
in question as a means of access to the divinity.
In those years there began to develop a group of spatial centres of power and
communication, around which there grew up a mixed population composed of
indigenous people uprooted from their communities (through fleeing from taxation or
as a result of the reducciones), freed or escaped blacks, mestizos, and poor whites.
With the establishment of the mining circuit which fed the colonial economy, the
relations of the viceregal capital with Potosi and Huancavelica converted Lima into
45
an administrative focal point, serving as a magnet for people from the whole Andean
area. This heterogeneous population was to be the setting for the actions of many of
the above-mentioned saints, and their attributes and biographies took shape in this
setting, side by side with the official version prepared by the prelates of the church.
In some instances, despite attaining canonisation, certain saints (e.g., Torobio de
Mogrovejo or Juan Macfas) steadily lost their popular following and ended up as mere
names in the rolls of the saints; in other cases, holy persons have been constantly
reinterpreted, acquiring a status and an enduring appeal far beyond whatever rank the
decisions of the church conferred on them.
The situation described changes dramatically when the cult of a saint takes root
in areas with a high proportion of indigenous peoples. In such places, Catholicism was
only one element of a whole spectrum of belief, ranging between intense acceptance
on the one hand and militant rejection on the other. In these places, the veneration of
the saints was filtered through systems of regional and local belief, which adapted the
teachings of the missionaries to their own logic and context. As a result, the "urban"
hierarchy of the saints became ever more distorted as the community established its
own order of precedence, in disregard of the decisions of the church. Even those saints
who had become "patrons" of the community or were established as part of its
immediate pantheon had to modify their attributes, symbols, and hagiography so as
to be incorporated into the common history—an oral history, obviously—of the social
group that took them to its bosom.
If we stick to the official version, the life of .Saint Rose (1586-1617) was divided
between her great charitable works for the poorest sectors of society and the severe
physical self-chastisement that she practised as part of her quest for heavenly glory
and the capacity to intercede for the souls of sinners. Her beatification and canonisa-
tion were achieved in a record time for the period (1668 and 1671). In 1669 she was
already patroness of the Viceroyalty of Peru, and by the following year her sainthood
extended through all America and the Philippines (Millones 1989: 891-905).
The popularity of Rose seems to have very ancient roots, directly related to her chari-
table works, already alluded to, and which she devoted especially to the patients in the char-
ity hospitals. On her death, the "plebe colonial" came together en masse to say farewell to
their benefactress. The crush as people tried to get close to her, touch her, and take her
garments as relics was so great that her body had to be rescued by the viceregal guard. The
Dominican order assumed this popular interest as its own and hastened the process that
ended with her consecration as a saint. Rose, however, was never more than a tertiary (i.e.,
lay) member of the order, and never entered a convent, despite having worn the habits of
both the Franciscans and the Dominicans at different stages of her life.
Her cult had immediate repercussions in Lima, where two churches were built in
her honour (Santa Rosa de los Padres and Santa Rosa de las Madres). Her name
reached beyond the confines of the church: even before her canonisation, all manner of
places—streets, squares, towns, and cities—were dedicated to her. Today there is no
Latin American country where Rose is not known: candles with prayers to her are sold
on the U.S.-Mexico border; her image is borne in processions in Manila; and there are
centres of pilgrimage in places such as Pelequen (Rancagua, Chile) orQuives (Lima, Peru).
This multiple perception of the saint is not an unknown phenomenon in the
Catholic Church. Even in countries with a more homogeneous social base than the
Saint Rose through the Eyes of Devotees 47
Andean states, there are marked differences between the popular forms of cult and
those which appeal to the middle or upper classes. When we add, to the variable of
class, the factors of ethnicity and the passage of time, the spectrum of types of
veneration to which sacred figures are subject becomes much wider.
Still more complex phenomena lurk behind these diverse perceptions of the saints.
It could be postulated, for example, that contemporary urban veneration of Saint Rose
has more to do with the evangelical policies of the Dominican order, and that the saint,
as she is envisioned in rural areas with a largely Indian population, draws together a
whole series of processes of religious superimposition and interpretation.
By way of elaboration, consider the stress that the Dominican order placed on the
need for a holy figure to replace Our Lady of the Rosary as a maternal force. When
writing about Rose, the Dominican biographers reiterated time and again their joy on
discovering that there was now an American saint, born, moreover, in the capital of
the viceroyalty. It is no accident, therefore, that it should be Our Lady of the Rosary
who, in Rose's accounts of her frequent visions, gives her son in marriage to Rose.
This hand-over of responsibilities does not form part of the "sacred history" of the
Peruvian sierra. As can be seen in the verses that her faithful dedicate to her, the saint
seems to be linked more strongly to the local environment. Mountains, streams, fauna,
and flora rejoice at the mention of her name and seek to ensure the blessings of heaven
that they achieve through Rose's intercession. Having been incorporated into the local
pantheon, she ceased to be the exclusive property of Lima, and became—with or
without ecclesiastical approval—the patroness of the people, responsible from then
on for the well-being of her charges.
Let us consider, in the light of all this, the work of two poets. Luis Antonio Oviedo
y Herrera, count of La Granja, was born in Madrid in 1636 and educated at Salamanca.
A member of the social circle of the Marquess of Castell-Dos-Rius, viceroy of Peru in
the Academia de Palacio, he wrote in 1707 (or 1708) a very long poem, "The Life of
Saint Rose", in royal octaves (Oviedo 1867: III; Sanchez 1950-51: 139). The poem
recreates in verse the official story of Isabel Flores de Oliva and sets forth all the ste-
reotypes in terms of which society in the viceroyalty thought of the non-Catholic world,
eulogising in its turn the colonial Arcadia which was seventeenth-century Lima.
Pio Campos Chavez was a contemporary poet, writer, and musician from the
central highlands. Born in Carhuamayo, his career was at its peak in the 1970s, when
he directed and composed the majority of the verses, dramatisations, and songs that
were performed in the festivals of his town. He began, but never completed, his
religious studies, but he made up for his lack of formal education with curiosity and
intelligence. He succeeded Herminio Ricaldi, who had been the innovator of the
popular festivals of the central highlands. His early death put an end to an already
significant and prodigious output.
The American identity of Saint Rose in the

poet's muse
Let us begin with Oviedo's Rose. The colonial author begins by explaining his reasons
for writing the book and, to this end, places the sacrifices of Rose in the context in
which woman can attain glory—namely, in the struggle against her sinful nature.
Oviedo, entirely in accordance with the thought of the period, goes on to explain why,
on this occasion, the purpose of his writing is not the celebration of military glories:
I do not sing the deeds, the victories

Of an immortal male, warrior champion,
Nor famous memories of high renown,
Sculpted by steel in bronze and marble;
I sing the glories of a sacred heroine
Who was born a Rose that she might bring the dawn
And with deepest heart of true humility
Triumphed over Lucifer, herself, the world.
(Oviedo 1867: 1)
Further on he refers to Lima:
To the Tormes and the Henares, which in Spain

Are mirrors of one and the other science,
Rimac, which bathes its learned Areopagos,
Cedes not in antiquity nor in eminence:
There is to its children no art or science,
But they take it with ease unto their bosom;
As lively comprehend the minds that study,
So, with ingenuity, they better what they learn.
More than fifty churches on her bank
The Rimac sees, where art and grandeur both
Give form to the religion of Castille;
Increasing the richness of their fervor
With equal splendour at the one time shine
The altar and ecclesiastical purity:
Their learned desire is archive of the faith,
And each holy man a living temple.
Two thousand or more from every order drawn
Busy themselves in sacred ministeries;
Almost their equal in number and perfection
There flourish monasteries of virgins:
Other saintly folk, not formally encloistered,
Abide yet by their own strict discipline,
And join in chorus with her who is triumphant
In this militant new Jerusalem.
(8-9)
Comfortably established in the City of the Kings, where he would remain until his
death, the count compares his colonial paradise with the Old World that he left behind
him, and with pre-Columbian indigenous society:
The abysses of errors and deceptions

Threatened to o'erwhelm the ship of the church;
Africa and Asia, equally afflicted,
Were inundated with floods of sects;
Great flocks from the supreme shepherd
The wolves of Europe they robbed,
And thus was necessary a new world
In which to plant the renewal of our faith.
(Oviedo 1867: 16)
The author sketches a portrait of Catholic Europe, ruling out consideration of the
other continents, which he supposes lost to the cause of the faith. But he is anxious
also for Europe because of the Protestant Reformation that, "like a pack of wolves",
is stealing away the flock that had previously belonged to his church. One must turn,
then, to America as the setting in which to found a world freed from sin and from the
mistakes of history. He begins his account of the continent by describing its inhabitants
and religions:
Made blind by ancient ritual, they rendered

Adoration to the prince of Avernus,
To worship at false idols and altars alike,
The base alternate error inflamed them:
The three siren furies of the Cocytus,
Peopled in the Indies another Hell,
So that evils unpardonable
Might crown blame with blame.
(ibid.)
The first task of the Catholic faith, then, is to purify this great stage, peopled with
idolaters despite its magnificence, so that it may itself take root there and compete on
equal terms with the mother country. The birth of Isabel Flores de Oliva, the future
saint, answers that need for a miracle:
Pouring out aromas, putting forth sweet odours,

Revealing nectar in wallflowers,
Like daybreak between snowy splendours
Or dawn between crimson banks of cloud,
Rose was born in April, month of flowers,
And in Lima, whose orange blossom turned to rubies;
It being impossible to be born in Heaven,
She was born in Lima—the next greatest honour.
(Oviedo 1867: 19)
SO TEXTUALISING HISTORIES AND IDENTITIES
This is not the place for an analysis of Oviedo's verses. What I wish to point out
here is the way in which he expresses a profoundly Latin American consciousness,
despite his having been born in Spain. From his perspective, Rosa is the culmination
of a process of civilisation which makes it possible to achieve in America a perfection
that was never attainable in Europe. His text also reveals an utter fascination with the
Peruvian capital, in which he enjoyed great personal success. As he puts it: it being
impossible for Rose to be bom in Heaven, she was born in Lima.
Rose and Pachamama in Serrano

evocations of the sacred
While Oviedo is preoccupied by the date of Rose's birth, Pio Campos is concerned
with the current civic/religious festivity, on 30th August. This date was settled
upon by a minor alteration of the real date of her death (24th August) for reasons
of organisation of the ecclesiastical calendar. One could speculate, also, that Oviedo
felt the need to present Rose as the dawn of a history which was just beginning
to be written. His reference to April and the European spring (which does not
correspond to the Andean cycle of seasons) seeks to reinforce this sense of
beginning which tinges all his verses. As we shall see, this is not what concerns
Campos:
LET US REMEMBER
Let us be cheery,
Let us sing, then,
Let us remember
Little Saint Rose.
Thirtieth of August, feast of the people,
when the hearts of all are cheered.
Thirtieth of August, feast of the people,
when one day my devotion was born.
On which you share her feelings
like the wind in the reeds
in her beautiful lakes.
Let us be cheery ... etc.
Thirtieth of August.. .etc.
It is the tradition of Carhuamayo
to drink of its fair ancestry
which its fathers left to it. . . .
Let us be cheery . . . etc.
Thirtieth of A u g u s t . . . etc.
In the central highlands, the saint is the presiding spirit of one of the most important
festivities of the region. In Carhuamayo (department of Junin), the celebration lasts
nearly two weeks and involves a continuous round of processions, church services,
dances, and dramatic performances. During the commemoration of Rose of Lima that
I attended in 1984, the drama portraying the death of the Inka Atahuallpa was
performed, although there is both oral and written evidence that other works—such
as the "Ollantay" and "Tupac Amaru II"—have also been staged on this occasion.
Nevertheless, in August, Rose and the last Inka coincide perfectly because both of
their cults are strongly linked to the earth.
On the one hand, Rose with babe in arms appears to personify Mother Earth, and
it is in this aspect that she is prayed to and celebrated. August is the ideal month for
this, as it includes the time of first ploughing, and in turning over the soil one moves
her body, with the attendant possibility of displeasing her. The death of Atahuallpa
seems to be transformed into a ritual of appeasement, whereby the sacrifice of the
chosen person ensures the favour of the gods so that life may be renewed. It is worth
noting, moreover, that during the fiesta the body of the executed monarch disappears
immediately. This alludes to the historical fact that Atahuallpa's corpse was removed
from its Christian tomb and probably taken to Quito or to one of the snow-capped
mountains near Cajamarca. Interpreted more symbolically, when no evidence of his
death is to be found, the Inka enters the supernatural universe, ready to be reborn when
his people have need of him.
This way of perceiving death removes the element of mourning from the
ceremony and reminds us that, for the indigenous population, the concept of final
destiny in heaven or hell of Catholic doctrine, are inapplicable; in fact, dying
simply means entering another level of existence (Millones 1985: 51-58; 1992).
MY HOPE
Carhuamayo, you grow huge

in the harshness of thunder and wind,
where the hail and the pampas
roar by day and night.
Chinchaycocha made you a pearl,
and your children a beautiful town;
we will follow with firmness
the way of progress.
We work with vigour,
and tradition strengthens us,
disdaining strange fancies;
we remember the Inkas
exalting immense faith;
we venerate Saint Rose.
In contrast to Lima, Carhuamayo boasts of its ability to tolerate the harshness of

its situation, 4,200 meters above sea level in the puna. Chinchaycocha is the lake on
whose banks stands the town and its six thousand or more inhabitants, with their strong
herding tradition. Their capacity for hard work, as they tell us themselves, is a legacy
from the Inkas, and is an important motif in the civic celebrations. Next, the bard
emphasises the rural beauty of his region and offers dominion over the environment
in tribute at the feet of the saint:
O LITTLE SAINT ROSE (LONG LIVE MY TOWN)
Long live my land,

beautiful Carhuamayo,
Covered with ichus, flowers, and quihuash.
With its dear lakes,
just like your eyes,
their waters are the colour of my sky.
0 little Saint Rosie,
patroness of my town,
you are the guardian of our hopes and dreams.
Thirtieth of August,
1 sing to you
with the soul, from the heart,
that you may give to me
your sacred devotion.
The identification of Rose with Carhuamayo goes beyond the Christian blessing
that is received from heaven. As can be seen from the preceding verses, the saint is
incorporated into the lavishly depicted landscape. This is not due to any naturalistic
zeal; in making her "patroness", the poet identifies this friendly terrain with the saint
herself, whose eyes are "the image" of the "dear lakes" so characteristic of Carhuamayo.
This relationship is made clearly explicit in a wayno by Campos, which tells us once and
for all, in its closing lines, what is implicit all throughout the celebrations: Saint Rose
personifies Mother Earth, or Pachamama, in whose lap the people find shelter.
Carhuamayo, pretty and beloved land,

in your breast our hopes take on new life;
I respond with my mind and my arms
to make you a happy and a great town.
As long as I have the breath of new life,
Carhuamayo, I will increase your greatness;
I'll plough a furrow for those who come after,
because your name is so noble and so sacred.
Now your children are workers and drivers,

artisans and intellectuals,
who bring you fame and honour.
Saint Rose, tender and beloved mother,
light up for me the fate which I have chosen;
let neither distance nor time separate us,
for I live beneath the shelter of your cloak.
Conclusion
How are we to explain the transformation of a colonial mystic into one of the favourite
images of popular Andean religion? Here, there are two possible leads to explore. On
the one hand, August is the month when the earth is "turned over"—that is, prepared
for sowing. It is a dry time, before the arrival of the rains in November. This action of
turning over the soil for farming has serious ideological implications in the Andes. In
delving into the soil, dangerous channels of contact with the inner world (uku pacha)
are opened up, and it is possible thus to offend the ancestors ('old ones', machu or
gentiles) who may cause harm to the community as a whole or individually to anyone
they think has failed to show them respect. The ritual, then, is intensified by the need
to maintain a balance between the relationship with the supernatural and the fact that
it is vital to prepare the earth in the hope of a good period of rains for the next farming
season. This makes August a difficult and dangerous month, and therefore ideal for
the Catholic saints whose feast days fall around that time to assume the necessary
native attributes to enable the reworking of their cult.
The second trail is a historical one. Saint Rose lived in Quives during her
adolescence. She arrived at the age of eleven and was confirmed in the chapel of the
town by Saint Toribio of Mogrovejo. But Quives already had a long memory of
relationship with the sacred. The curacas (native chiefs) had warded off with spells
the ambitions of their highland neighbours (the Cantas) and their coastal ones (the
Colliques). Their lands were well known for a variety of coca (Erythroxylon
novagranatense var. truxillense), highly valued in all corners of the Andes. In the end,
the Inka, from his base in Cusco, crushed the Quives and their rivals and took
possession of the coca-growing areas. The people of Quives were not daunted by this,
and they even tried to "poison" the "Son of the Sun", which brought down fresh
reprisals upon them (Rostworowski 1988: 178-179). Years later, under Spanish
administration, the natives of Quives still maintained a reputation for sorcery, which
gave rise to the colonial refrain "Esquive vivir en Quive" 'avoid living in Quive[s]'
(Palma 1983: 46-47). It was there that Gaspar Flores secured an appointment as
superintendent of mines for the area, and brought his family with him. He was in charge
of the local Indian workmen who wrested the silver from the ore, prior to its despatch
to Lima. In this improvised town, which was basically a mining encampment, lived
the large Flores Oliva family; among the youngest was Isabel, later to become Saint
Rose of Lima.
It should come as no surprise, then, to find that the contemporary fiesta contains
within itself manifestations of both the Christian tradition and the Indian worldview,
interpenetrating yet mutually incompatible, to the accompaniment of processions,

bands, firework displays, and sexual licence. In the colourful diversity of these two
universes, conjoined yet contrasting, the whole culture of the Andes is contained. The
way that these worldviews are expressed in fiesta does not conceal the conflict, but
makes of it the foundation of a commitment to living together that we are only just
beginning to discover.
ACKNOWLEDGMENTS Translated from the Spanish by Christy MacHale. The ethnographic

information on Carhuamayo comes from fieldwork undertaken between August and September
1984. The songs by Pio Campos were photocopied in manuscript form from the Campos family
archive in 1984 as well, by permission of the songwriter's son, Dominique Campos.
4
JOANNE RAPPAPORT
The Art of Ethnic Militancy

Theatre and Indigenous Consciousness
in Colombia
I
An June of every year, the state-sponsored commemoration of the "Day of the
Peasant" is celebrated throughout the Colombian countryside, accompanied by the
flowery discourse of local officeholders, the cacophony of small-town bands, the
aroma of grilled beef furnished by politicians in search of votes, and the distribution
of seedlings to forest-eroded mountain slopes. In highland Narino, the celebration has
been renamed the "Day of the Indian". While the municipal mayor delivers his speech
in the town plaza, the band erupts into raucous song, the senator flies in from Bogota,
and trees are exchanged for votes, the communities for whom the celebration is
intended stage their own festivities in their community house or in the public space of
repossessed lands. Shielded from the gaze of local mestizos, community leaders
deliver their own speeches, and the public is treated to dramatic presentations and the
music of rural bands. The art of these ethnic militants recalls the exploits of their
colonial chiefs, the horrors of conquest, and the achievements of the modern land-
claims movement. Visual image and verbal cliche prompt the audience to reflect on
the rhetoric of the contemporary native rights movement on the one hand and the
narratives of community historians on the other.
Cumbales and neighbouring Muellamueses use song and drama to assert an
indigenous nationalism that at once reinforces their commitment to the construction
of a distinct ethnic identity, while it acknowledges their membership in the broader
Colombian nation. On the one hand, militant art emphasises cultural specificity
through recourse to a history conveyed through texture, and obtained through the
55
56 TEXTUALISINC HISTORIES AND IDENTITIES
dense combination of sound, movement, image, and story, as opposed to a single-

channel narrative. Texture is achieved through the choice of themes and cast of
characters, the design of costumes, and the use of language. On the other hand, implicit
in their artistic production is a recognition that indigenous historical experience
intersects with that of other Colombians. The shared nature of the past compels native
people to clarify what it means to be at once Pasto and Colombian, a task that the
artists accomplish through appeals to a national indigenous political discourse.
This chapter will examine the conflicting poles of attraction exerted by national
and ethnic identities as they are manifested in politically inspired artistic creation
in the communities of Gran Cumbal and Muellamues, Narino. These high-altitude
Andean villages, which lie along the Colombia-Ecuador border, are resguardos,
communally owned indigenous territories administered by semi-autonomous polit-
ical authorities.1 Under Colombian indigenous legislation (Colombia 1983),
resguardo lands are legitimised by colonial-era titles granted by the king of Spain.2
Notwithstanding indigenous appeals to colonial authority, many of the lands
encompassed by such documents have fallen into the hands of local non-native
landlords. As a result, indigenous people generally inhabit the mountain slopes,
surrounded by mestizo-owned cattle ranches located in the valley bottoms; many
resguardo members are forced to seek their livelihoods as day workers on haciendas
or as domestic employees in nearby cities. While the land base controlled by the
community has diminished considerably since the colonial period, the native
population has grown, forcing land-poor communities to pursue a variety of legal
and extralegal methods for reclaiming their lands. Repeated failures to regain
territory through the courts led indigenous leaders of the early 1970s to embark
on a militant programme of land claims that has included the peaceful occupation
of those ranches that lie within resguardo boundaries, as well as the revitalisation
of native culture and historical consciousness (Rappaport 1994). This new-found
militancy has caused communities to re-read their colonial resguardo titles of which
they acquired certified copies on repeated occasions since the turn of the century.
Their creation of ethnic art can be understood as an attempt to convey this
knowledge to the population at large.
Artistic representation and history

Selective reconstructions of the past
Dramatic presentation and musical composition are not new to the Narino highlands.
Dramatic spectacle has been an integral part of the celebration of Christian feasts since
the colonial period: Holy Week and Advent processions include re-enactments of
biblical scenes, while biblical tableaux are prepared for Epiphany, and Spanish-lan-
guage nativity plays punctuate the Christmas season.3
Contemporary theatre and musical presentations are, in contrast, frequently
secular in nature, performed for neighbours in rural schoolhouses or for broader
cross-sections of the population at large community meetings; some of the more
politically active sectors of Panan have introduced secular, historical theatre into
Epiphany carnival. Recognised by the community as official community or-
The Art of Ethnic Militancy 57
ganisations, these artistic groups are frequently short-lived. Nevertheless, they play
a central role in integrating young people and women into political life, providing
a chance for those not directly involved in the all-male and highly politicised
resguardo council to participate in the political process. Members of theatre groups
usually attend most council meetings, joining grass-roots public works or-
ganisations. In other words, they are an avenue for the insertion of young people
into community affairs. It is thus no accident that a few years down the line, some
of the young male actors have become council members: their theatre experience
groomed them for political office.
In communities in which memoristas, or community historians, are generally older
men who were once active in community affairs, music and drama also serve as
vehicles for more youthful forms of historical interpretation.4 Thus, it is no accident
that Los Cumbes, a Cumbal drama group, identified itself as memoristas on the title
page of its scripts:
HISTORIA DEL GRUPO ARTISTICO LOS CUMBES, representantes de esta gran historia
que surge de una gente muy joven como es EL GRUPO ARTISTICO LOS CUMBES,
quienes memorizaron este gran evento....
[HISTORY OF THE ARTISTIC GROUP LOS CUMBES, representatives of this great
history that arises out of very young people, as is THE ARTISTIC GROUP LOS
CUMBES, who memorised this great event. .. . ]
Artistic performance thus creates a space in which historical knowledge can be

interpreted by alternative sectors and shared with a broad public. The past that artists
present is a selective construction, based in part on the scanty historical documentation
that native Cumbales have had access to, and in part on the actors' and singers'
personal experience. More than narrative portrayals of particular historical milestones,
artistic interpretations of the past serve as mnemonics, references to other more
complete stories (Cohen 1985, Rappaport 1990).
The selective reconstruction of the past is most useful in the present, providing
people with a model of reality to strive for. Among the Pastos, the contents of this
model are to a large extent grounded in the distant past, offering a glimpse of what
indigenous autonomy might look like once the community wins its struggle. This
model is centred around the historical event that has most traumatically affected native
Colombians: the Spanish invasion. Let us look at the plays presented at the 1988
celebration of the Day of the Indian in Muellamues. Four of the six plays presented
treat conquest themes, portraying Spanish authorities—such as, for example, the local
landowner, Erazo—who swindle native lords, sometimes local chiefs, but also figures
from Colombian national folklore, such as Nutibara and La Gaitana. Of the two other
presentations, the re-enactment of a traditional marriage ceremony is a theme common
to Narinense theatre groups. The Muellamues rendition is particularly interesting
because of the title of the play:
Vereda Riveras presenta los novios con sus tradiciones y costumbres indigenas,
con nuestra autoridad y su ley especial que nos protege, Ley 89 de 1890. iViva
el dia del indigena!
I Vereda Riveras presents a marriage with its indigenous traditions and customs,
with our authorities and their speciai law that protects us, Law 89 of 1890. Long
live the Day of the Indian!]
Here the theatre group linked its depiction of traditional marriage customs with support
of the resguardo council, perhaps prodding the audience to remember that in the earlier
part of the century resguardo councils were charged with the moral education of the
community and frequently obligated courting couples to get married. But even this
portrayal of recent history is intimately connected to the memory of the Spanish
invasion. The title incorporates a slogan in favour of defence of resguardo legislation,
which is frequently confused by Cumbales with their colonial-era titles; even though
this particular play treats a relatively recent theme, its creators have situated it within
the broader framework of the conquest and colonial periods.
The selectivity of these presentations is due in part to the paucity of available
historical documentation. Incursions into regional archives are inspired by the need
in the course of land claims for copies of colonial-era briefs. Given that these
documents provide useful, albeit abbreviated, information on colonial political actors,
conquest plays that draw upon them concentrate almost exclusively on naming
conquest-era hereditary chiefs and colonial-period Spaniards.
It is possible to distinguish some of the other sources from which the historical
interpretations of these dramas are drawn. The funeral of Nutibara, for example, opens
with the pre-conquest burial of a hereditary chief, his body covered with gold jewelry.
The image was taken from a poster advertising a museum exhibit of pre-Columbian
goldwork; the poster hangs in many council offices.5
In a fascinating analysis of a sixteenth-century relation describing the conquest
of Queretaro, Mexico, Serge Gruzinski (1985:40) demonstrates that its Otomi authors
used costumes and dances from colonial festivals as templates for describing pre-Co-
lumbian armies. The Muellamues theatrical version of the traditional wedding simi-
larly employs ritual as a source of historical evidence. Other dramatic portrayals of
weddings highlight the process of negotiation between the families of the bride and
groom and the accompanying exchange of cooked food. The Muellamues play depicts
the wedding ceremony itself, complete with priest and mass. This is reminiscent of
farcical re-enactments of weddings that take place at the Sacada de la Vieja (Removal
of the Old Woman), a secular ritual performed at the completion of house construction,
in which the woodland spirit (the "Old Woman") hidden in the beams is removed from
the house.
History and ethnic typologies

Insistence upon the conquest-era roots of contemporary indigenous experience is a
product of more than the paucity of historical data. It also reflects the authors'
acceptance of the ethnic typologies of the dominant society, in which native peoples
are relegated to the historical past (Fabian 1983, Dorris 1987). Similarly, militant
theatre situates native communities in distant locations, especially in the tropical
lowlands. For example, Cumbal theatre presents shamans wearing headdresses from
the Putumayo, but since they omit any acknowledgment of the origins of the costume,
the actors imply that the curer is from Cumbal. In "The Story of an Indigenous Slave
from Gran Cumbal, Having Arrived at the House of Two Bad-humoured Foreigners",
a Los Cumbes actor states that the power of native peoples originates in the jungle:
Tu no sabias que nosotros los indios, tenemos nuestro estudio por medio de
la telepatia o de las brizas que cubren los cuatro vientos de la Amazonia
colombiana.
[You didn't know that we native people got our learning through telepathy or
from the breezes that cover the four winds of the Colombian Amazon.]
The practice of using geographic displacement to define "otherness" also charac-

terises traditional Narinense ritual and dance. The Jambo dancers of Panan, for
instance, represent the people of Mayasquer, on the warmer slopes of the Cordillera
overlooking the Pacific lowlands; the word jambo is a common variant of yumbo, a
term used in highland Ecuador to signify the lowland outsider (Salomon 1981, Poole
1990). Many of the actors in contemporary nativity plays carry baskets made in
Mayasquer, as if to equate terrestrial distance with the temporal remoteness of Christ's
birth.
But in the popular Colombian imagination, native people are increasingly associ-
ated with the production of ethnic art, a stereotype used over and over by Narinense
actors. Equally influential in this image is the fact that the only available evidence of
the pre-Columbian past is found in ceramic artifacts. The Muellamues version of the
conquest shows people peacefully going about their everyday lives, painting small
statues of birds. Similarly, one of Los Cumbes's plays, in which miserable sharecrop-
pers overthrow a greedy and exploitative landlord, is entitled "The History of Our
Native Handicrafts from Our Gran Cumbal".
Rewriting History
In How Societies Remember, Paul Connerton (1989) suggests that ritual constitutes
the most powerful source of historical knowledge in most societies. Commemorative
ceremonies in particular serve as vehicles for historical interpretation, insofar as they
not only remind participants of events, but re-present them, thereby lending an
instrumentality to history by shifting its locus from the past to the metaphysical
present. Once history is inscribed, not on paper, but in bodily activity, that claim upon
the past is strengthened. According to Connerton, commemorative ceremonies "do
not simply imply continuity with the past but explicitly claim such continuity ... by
ritually re-enacting a narrative of events" (ibid. 45). Penelope Harvey (in this volume)
notes that such ceremony does more than "re-enact a narrative of events", but
re-emplots them, thus in effect subverting the purpose of remembering the original
narrative. Commemorative ceremony, far from honouring the past, becomes a tool for
resistance in the present. As we shall see, this is precisely the purpose of ceremony in
highland Narino.
Resguardo council ceremony effectively re-lives the past through the voicing of
the same words that were used in the colonial period to grant land rights to native
authorities. It is thus no accident that these ceremonies are frequently re-enacted in
militant theatre. But re-presenting history implies more than the repetition of events
of the past: in Narinense drama it also signifies the concomitant correcting of history.
Muellamues portrayals of the conquest depict early battles in which native people are
always the victors. Los Cumbes sets its characters in the timeless past, living in
grinding servitude to large landowners, but the natives always win in these plays,
which close with slogans of the contemporary land-claims movement. The moment
of triumph of the native forces against Spaniards or landlords is always greeted by
cheers from the audience. The correcting of history is a cathartic experience.
But even if history is corrected by dramatic forms of establishing connections to
a heroic indigenous past, the triumphs of conquest-era chiefs are situated in a European
historical framework. For instance, Vereda Comunidad in Muellamues named its play
the "Pasion y Muerte de Nuestra Raza", a very clear reference to the life of Christ.
The roots of this Christian framework originate in the theatre of the past. Traditional
drama was always religious in nature, teaching universal Christian history to the native
South American faithful. Holy Week processions, for example, were essentially a
series of biblical tableaux in which people impersonated biblical characters, moving
across the countryside in a chronologically organised pageant of Judeo-Christian
tradition.6 I observed a truncated version of the elaborate processions of the past in
Panan in 1987, including a host of tableaux, only one of which was locally inspired:
the council with its banner, trailing behind the cross of Christianity and the souls of
humanity. In this procession, native history is absorbed within the framework of the
universal. 7
Although militant theatre employs Christian metaphors, the presentation of secular
plays is also situated within a specifically Andean vision of the past in which historical
chronology is reconstituted to conform to the organisation of topographic space.
Indigenous communities of highland Narino are organised in a series of parallel bands
of territory, which I will call sections, ordered in a hierarchy. Events of the past are
conceptually arrayed within this territorial structure (Rappaport 1988, 1994). The
Muellamues commemoration of the Day of the Indian, for example, is celebrated with
a series of plays presented by sections, although at the performance I witnessed they
were not presented in hierarchical order. The history of Cumbal theatre groups is
frequently extremely localised, situating historical chiefs in specific sections. Such
local histories are loosely organised according to the section hierarchy, where histor-
ical figures and sections are represented by their most common surnames.
The power of images

Cacique Cumbe
The history embedded in Pasto drama and song is not revealed through narrative, but
through images that move people to remember the past and also to act upon it. The
power of images finds its origins in the present, in concrete activity; Pierre Nora (1984:
xix) contrasts this active and malleable past with history by calling it memory.8 I am
not speaking here of history in the chronological sense, but of the forging of a
connection with the past by re-creating it in the present, a process that, as has already
been noted, Connerton (1989) maintains is most powerfully expressed through ritual.
The experience of the past that derives from ceremony is inchoate, vague, and
open-ended, unlike the more circumscribed reactions evoked by narrative; I will
distinguish through the use of the term texture that non-discursive experience which
is less likely to be interpreted coherently, but which is infinitely more powerful than
narrative.
Texture and its component images lie at the heart of national identity. Vague,
abbreviated, and sometimes stereotyped, they supply the foundations for a sense of
nationality (Melo 1989). In Cumbal and Muellamues, many such symbols are culled
from readings of resguardo titles. Little is known about the meaning of or the historical
referents associated with these images because they are only scantily described in the
documents. Nevertheless, or perhaps precisely for this reason, they move people to
action. The power of such symbols was expressed to me in the following story about
the repossession of the Llano de Piedras, the first of a series of militant land claims
actions in Cumbal. Confronted by repression, people invoked the name of Chief
Cumbe to justify their actions:
La policia les decia, "A Ustedes, ^quien les dice vamos a recuperar?" o "^Cual
es el que empieza? Uno ha de ser". Entonces que decian, "No, a nosotros nadie
nos avisa. No se. Ni se conoce a la gente que baja. Vamos a la tierra, pero no se
sabe quien. Es que a nosotros nos relumbra el cacique. Vamos a la tierra y todo
el mundo se levanta. Relumbra en el sueno el cacique". (Bernardita Chiran,
Vereda Cuaical)
[The police would say to them, "Who tells you to repossess the land?" or "Who
is it that goes first? It has to be someone". Then they would say, "No, no one
tells us. I don't know. You don't even know the people who are going. We go
to the land, but don't know who (sends us). It's that the chief lights our way.
We're going to the land and everyone gets up. The chief lights the way in
dreams".] (Bernardita Chiran, Vereda Cuaical)
Cumbe's name is borrowed for countless music and theatre groups, which fre-
quently call themselves by some variant of "Los Cumbes" and the chief is invoked,
time and again in plays and in music. He does not appear in any colonial-era
documents. The power of his image derives from the link between past and present
that is established through the use of a name that so closely approximates the name of
the resguardo.
The invoking of the referent-less image of Cumbe conforms to traditional Andean
forms of representation. In the pre-Columbian Andes the significance of visual images
depended on oral exegesis (Cummins 1988). Similarly, Cumbe's name serves as a
stimulus for the recall of events to be remembered and interpreted by the listeners.
Photographic images, according to Susan Sontag (1977: 18-19), are only made
meaningful to the viewing public when they are articulated within a previously
constituted ideology. Similarly, Cumbe is meaningful to the Cumbales only within the
context of the ideology of their movement, whose programme developed only after
native historians began to interpret indigenous history for community members. The
young memorista playwrights and songwriters recast history in images framed by
contemporary ethnic ideology. Their frame work, aptly conveyed by the use of political
rhetoric in dramatic dialogue, is most thoroughly experienced by the actors them-

selves, for whom theatre constitutes a form of political education. The young
memoristas develop a familiarity with political ideology and with resguardo history
as they prepare and learn their lines. What they ultimately create, though, are images
that the audience must situate within its own framework.
Visual images are essential to historical representation as it is expressed through
militant theatre. On one level, they are all that the audience receives in the course of
these dramatic productions, given that the untrained voices of the actors rarely project
enough to be heard by the public. But on a more fundamental level, in Andean culture,
knowledge is acquired through the act of seeing. The active quality of seeing is
expressed in colonial-era Quechua prayers and in contemporary women's songs, in
which divine knowledge is acquired through sight (Harrison 1989); among the
Colombian Paez, seeing is integral to political ritual and mythology related to the
legitimisation of land tenure (Rappaport 1985).
The texture of the past

What the audience sees in these dramatic presentations is a series of static images
whose meaning is conveyed through texture. In a study of the symbolism of commu-
nity among African-American migrants to Washington, D.C., Brett Williams de-
scribes the re-creation of rural Carolina culture in the metropolis as it is manifested in
foodways and in gardening. She emphasises that Carolina customs convey a sense of
texture, of "dense, vivid, woven, detailed narratives, relationships, and experiences",
which take on meaning as they are relived through expressive forms that "rely
powerfully on repetition, improvisation, and the exploration of sometimes narrow
situations through many emotional, sensory, interpersonal, and reflective voices"
(1988:47).
I would like to narrow Williams's notion of texture, removing narrative from its
component attributes, in order to convey the essentially visual and aural experience
of theatre in Narifio. The texture of Narinense theatre is depicted most clearly in
costume and in the symbolic use of language, which is conveyed through the names
of characters and the languages they speak. Texture constitutes the principal vehicle
for recalling the history of everyday life, as opposed to political history, which is
frequently expressed through plots and through dialogue. Let us look first at costumes.
In Muellamues, pre-conquest people are frequently portrayed with tropical forest
attributes. Warriors are semi-naked, wearing grass skirts, feathers, and face paint;
chiefs wear tunics. Panan's cacica Maria Panana, appearing at the 1987 Carnival,
donned a modest but makeshift tunic, similar to those worn by theatrical chiefs. Almost
all pre-Columbian characters are decked out with "gold" jewelry, made from metallic
paper. Golden headbands hold in place the long hair that these characters exhibit.
Costumes conveying a sense of the more recent past, and sometimes even the
distant era of the chiefs, are generally constructed from the clothing that old people
still wear: double-weave ponchos and flannel trousers for men; homespun skirts and
shawls for women; everyone barefoot or shod with hemp sandals. Such garb is by no
means aboriginal in origin; it is worn by non-native peasants throughout the Andes.
Nevertheless, it has assumed a symbolic importance in the Narinense indigenous
movement, representing indigenous culture as it used to be; the display of this clothing
is interpreted as a form of cultural revitalisation.
Non-native characters sport stereotypical costumes that are almost a parody of
mestizo dress. Priests are identified by their red capes, pith helmets, leather boots, and
ties. Landlords and soldiers wear leather jackets; they are frequently identified by their
sunglasses.
Clearly, theatrical costume is hardly historically authentic. Instead, it is meant to
convey historicity through contemporary stereotype and cliche, as well as through the
distancing of observer and historical subject. It is thus not so much the accuracy of the
costume as the associations it elicits which are at stake here. Narinense theatre differs
in function from the historically accurate nostalgia of contemporary North American
re-enactments of the past. It is not meant to arouse a sentimental appreciation of earlier
lifeways; instead, it invokes history in order to activate militant sentiments in the
present (Lowenthal 1985).
Costume provided the major vehicle for historical expression in processional
theatre. Biblical characters did not do anything: they were simply represented by virtue
of their clothing. This is obvious in the personal reminiscences of participants in the
elaborate processions of the past, whose descriptions are essentially lists of costumes,
arrayed in processional order. Raul Fueltala of the resguardo of Panan always played
King Saul in the Good Friday celebration in Chiles, taking his place in a procession
of sixty to one hundred tableaux. He remembers the texture of the procession,
especially the clothes he wore and the props he carried:
Asi, yo me toco de ser el Rey. Entonces, claro, yo era a componer el caballo. El

caballo era bien compuesto con capa, bien adornado desde los cascos forrados,
con papel de ese amarillo, riendas, freno, la frente, todo, el galapago tapado con
capa, brillando eso de estrellas, adornos. En eso iba pues, en Rey propio, pues
eso si, y yo iba vestido bianco con capa y con la espada y con un copon tambien
de rey. Iba manejando mi caballo y con la espada y cogido el copon y con buena
corona, todo. A mi me toco atras de los Centuriones. Los Centuriones iban
adelante, cuando ya me gritaban por lista, "El Rey Saul, siga!" Entonces, ahi
tenia que seguir, y atras de mi vuelta, los que les iba a tocar.
[So I got to be the king. And, sure, I had to ready the horse. The horse was
decorated with a cape, adorned even to his gilded hooves, with that yellow paper,
the reins, the bit, his head, everything, the saddle covered with a cape that shone
with stars and ornaments. The king himself, then, rode out, that's right, and I was
dressed in white with a cape and with a sword and also with a king's goblet.I
would steer my horse, with the sword and carrying the goblet and with a nice,
crown, everything. And I came after the centurions. The centurions would go
ahead, and then they shouted out to me from the list, "King Saul, your turn!"
Then, I would have to follow, and behind me, those who were next.]
This sense of the past is not organised according to a chronological series of events,
obeying, instead, a sequence of images that are arrayed over topographic space,
duplicating in commemorative ceremony the close relationship between time and
space that pervades the native Narinense historical memory.
Texture is also conveyed through the use of language in the plays. The Pasto
language disappeared during the nineteenth century; the indigenous peoples of
highland Narino are all monolingual Spanish-speakers. Various methods of re-cre-
ating the aboriginal tongue are employed in theatre to convey what it meant to be
a native person in the past, on the one hand, and the intention on the part of many
young people to embark on the impossible task of linguistic revival, on the other.
Indigenous names, some Pasto and some from far-flung regions of Colombia,
convey a feeling of linguistic autonomy as characters appear on stage: Cruz Angela
de los Cumbes, Cristian Sotavento, Sebastian Panan. In one of the Muellamues
conquest plays, actors, probably return-migrants from Quito, speak Quechua while
a narrator declares over the loudspeaker that the community has finally recovered
the lost Pasto language.
Toward a national indigenous discourse

An examination of Colombian native manifestos led Michel de Certeau (1986) to
declare that the discourse of indigenous organisations is marked by cultural and
geographic specificity, and not by the construction of a common ideology and
language. I would agree that indigenous militancy is not centralised, as the historical
and geographic specificity of Narinense plays eloquently demonstrates. Nevertheless,
militant drama plays an important role in the creation of a national indigenous
discourse, conveyed through texture and image, much like the costume and language
already described.
Cumbe and Quintin Lame

The birth of a pan-native discourse is especially clear in the scripts written by Miguel
Angel Alpala, director of Los Cumbes. These scripts are but imperfectly memorised
by the actors, but the formulaic utterances they contain are remembered by all. Let us
look at some of these formulas:
Nosotros, los indigenas, mas aborrecidos de este gran "Cumbal" de los blancos
espanoles que llegaron este dia 12 de octubre de 1492 hasta las tierras llamadas
Guanani, que hoy las llamamos Colombia.
[We, the natives, detested by that Gran "Cumbal" of the white Spaniards, who
arrived that October 12th, 1492, at the lands called Guanani, that today we call
Colombia.]
El hijo de un indigena se sentara en el trono de sabiduria, para defender nuestra

propia sangre, sangre que por largo tiempo se ha estado ocultando por este
vengativo feroz de nosotros.
[A native son will sit on the throne ofknowledge, to defend our own blood, blood
which for a long time has been hidden by our fierce vengeance]
La prehistoria repercute lo de nuestros antepasados, sus asientos alla en esa

colina donde estd sepultada la casa de la divinidad, segun la prehistoria de
Bochica, quien par medio de signos o parabolas donde se constataba en los

sublimes rayos del sol: cuando sepresentaba par el orientey asise consagraba
las ceremonias de los dioses que adoraban nuestros antepasados. Pero los
aventureros que llegaron el 12 de octubre de 1492 en nombre de la civilization
hicieron blandir 5a cuchilla de la mano y la intention para quitarnos la vida de
nosotros y esas grandes riquezas que nosotros teniamos heredado de nuestros
caciques antepasados. Y de hoy en dia nosotros, los Colombianos, estamos
acompanados del valor, y unidos como un concierto de dguilas encolerizadas
lograremos la defensa de nosotros para que se nos haga justicia, y seamos
amparados por las autoridades competentes de toda la naci6n, de toda Colombia,
y para todos los colombianos.
[Prehistory echoes our ancestors, their homes there on that hill where the house
of the divinity is buried, according to the prehistory of Bochica, who through
signs or parables found truth in the sublime rays of the sun: when it shone from
the east and in this way they consecrated the ceremonies of the gods that our
ancestors adored. But the adventurers who arrived on October 12th, 1492, in
the name of civilisation, brandished knives in their hands and the intention of
robbing us of our lives and of those great riches which we had inherited from
our ancestors, the chiefs. And today we, the Colombians, are accompanied by
valour, and united like a concert of angry eagles, we will achieve our defence,
so that we are done justice, and so that we are protected by the competent
authorities of the whole nation, all of Colombia, and for all Colombians.]
Each of the phrases I have highlighted is a quotation from Manuel Quintin Lame's
Los pensamientos del indio que se educo dentro de las selvas colombianas (Lame
1971). A Paez sharecropper who moved the native peoples of Cauca and Tolima to
demand their rights to reclaim land, to have an autonomous political authority (the
resguardo council), and to occupy an autonomous territory (the resguardo), Lame
wrote his book in 1939, but it was not published until 1971, several years after his
death. The treatise swiftly became a tool for political education in indigenous com-
munities, including in Cumbal, where several copies circulate.9 Let us look at how
Lame's book operates as a model for dramatic dialogue in Cumbal:
1. 12th October 1492: The playwright makes repeated reference to
Columbus's arrival in America, although the Spaniards would not invade
the Pasto region for some four decades. The insistence on 1492 echoes
Lame, who repeatedly asserts that 1492 was a turning point in indigenous
experience; Lame frequently opposes it to 1939, the year in which his
book was completed, since he perceived the 447 years between 1492 and
1939 as marking the period of indigenous oppression, which would end
with the appearance of his treatise. The playwright calls Colombia by an
alternative name, "Guananf", the Bahamian island upon which Columbus
made his first landfall. Similarly, aboriginal territorial autonomy is repeat-
edly referred to by Lame through the use of the same concept. Both Lame
and playwright Miguel Angel Alpala use 1492 and the name Guanani as
mnemonic devices: Lame, to recall the cataclysm of the Spanish invasion
and the existence of a separate and autonomous indigenous experience;
Alpala, to recall Lame.
2. Reclaiming the throne: Lame's treatise is profoundly messianic, proclaim-

ing his role as saviour of Colombia's indigenous peoples, who would lib-
erate native Colombians from the darkness of ignorance through the
power of his knowledge; then justice would be done:
Ese crimen esta oculto, senores; pero esa justicia llegara, en que el indio
colombiano recuperara su trono, etc., etc. (ibid. 21)
[That crime is hidden, sirs; but that justice will come, when the Colom-
bian native reclaims his throne, etc., etc.]
Similarly, the playwright makes reference to the corning millennium, when the
aboriginal throne will be reclaimed, and native blood, long hidden, will once
again be visible. Lame writes that his knowledge was acquired through nature,
not through a formal education, likening his intelligence to the flight of birds:
Este condor de mi pensamiento y esa aguila de mi psicologia indigena la

que se engendro cuando paso ese condor o condores como un concierto
de golondrinas parleras que visitan las estaciones del tiempo. .. . (ibid.
65)
[ That condor of my mind and that eagle of my psychology, an indige-
nous psychology that was conceived when the condor or condors passed
by, like a concert of swallows that visit the seasons. . . . ]
The same flight of birds becomes a metaphor for knowledge and political
awareness in the Los Cumbes's script where, "united like a concert of angry ea-
gles, we will achieve our defence".
3. Prehistory: the lengthy passage in the play that situates the source of
knowledge in prehistory—in particular in archaeological remains—and
identifies the sun as a mouthpiece of this knowledge is also reminiscent of
Lame's writings. Los pensamientos states that humanity learned its
crafts—goldwork and stone-carving—from the sun (ibid. 24), making ref-
erence also to the Chibcha deity, Bochica. For Lame, history is encoded
in the remains of this glorious past (ibid. 76).
All three themes—the cataclysm of 1492, the messianic future, and the roots of
knowledge in the aboriginal past—are examples of a pragmatic sense of history, as it
is articulated in metaphoric language, in potent images, that at once entreat indigenous
theatregoers, in Connerton's words (1989: 45), to "re-present" their past within their
present conditions and to appeal to their memory of the words of Quintin Lame to
recall the demands of their movement.
It is indeed significant that native Narinense theatre draws upon a printed source
for its nationalist verbal images, for the written word has served as a vehicle for
communication with the dominant society since the colonial period. In effect, written
language can be understood as one of the points of intersection between native
Colombians and the national society, one of the interpretive spaces in which we can
begin to comprehend how these communities are, in fact, Colombian. Colombian law,
which is itself encoded in written form, requires that resguardos supply documentary
evidence of their existence in order to maintain their character as autonomous political
and territorial units. Writing thus forces native Colombians to consult the past in order
to justify the present.
Nevertheless, in the past two decades, written language has also become a vehicle
for communication among native peoples themselves. Unidad Indigena, a national
indigenous newspaper, is read and contributed to by the leadership of Narinense
resguardos. Communities routinely produce pamphlets and leaflets in the course of
their struggle. As Anderson (1983) has suggested in his analysis of nationalism, the
printed word generates a sensation of an "imagined community". Among native
Colombians it inspires a feeling of the existence of an extensive population that
participates simultaneously in membership in a broad indigenous community. Cumbal
playwrights invoke the existence of this imagined community by drawing on Lame's
published works for the metaphors they include in their plays.
Conclusion
If the purpose of militant drama is to expand the space of historical interpretation so
that it includes sectors heretofore excluded from the analysis of the past, it is only
partially successful. On the one hand, theatrical expression serves as a vehicle for the
recoristitution on the artistic plane of a sense of community. This is achieved through
a communal activity that makes personal reminiscences and memorista histories
publicly meaningful. In theatre, historical mnemonics permit the collectivity to re-live
past experience and, thus, to reappropriate it for use in the present (Shopes 1986). In
this respect, militant theatre represents a democratisation of historical expression.
Moreover, this new approach to historical interpretation is successful, inasmuch as it
recasts in secular form the religious spectacle of the past that once provided people
with a sense of universal history. But when we consider militant theatre against the
backdrop of the history recounted by Cumbal's local historians, it is considerably less
successful in its objectives. In comparison to memorista narratives, which themselves
incorporate only a few historical referents, these plays display a conspicuous poverty
of historical detail. The young actors seem to have neglected their homework,
displaying little knowledge of the contents of pertinent documentation or the experi-
ences recounted by memoristas.
The space of dramatic presentation might be better understood, though, as an
opportunity for young people to learn how to remember history, as opposed to what
should be recalled. Like local historians, militant dramatists insert historical referents
into a framework describing experiences of the present or the recent past. While
memoristas weave this background with threads originating in narratives of recent
experience, actors evoke the quotidian and the recent through their portrayal of
costume. The names of the native authorities of the titles are thus projected against a
contemporary and experiential backdrop. Likewise, quotations from Lame are se-
lected as familiar images that are deployed in a visual field built upon the clothing of
recent memory.
Notwithstanding the amateurish historical analysis of the playwrights in com-

parison to that of the memoristas, it is precisely by emphasising recent experience to
the detriment of historical explanation that militant actors constitute themselves as a
new brand of historians. If their dramatic training leaves them with little in the way
of historical data, it introduces them to the craft of the memorista, whom they learn to
emulate. In time, some of the best of this youthful group will take up the mantle of the
resguardo council, acquire a greater fluidity in interpreting the past, and eventually
become memoristas in their own right. How they articulate their historical knowledge
in the future, whether through oral narrative, newspaper articles, political rhetoric,
song, or some other medium, will become evident in future years.
ACKNOWLEDGMENTS The research upon which this paper is based was conducted in Cumbal,
Narino, and in various Colombian and Ecuadorian archives in 1986-87 under the sponsorship
of the Council for the International Exchange of Scholars, the National Science Foundation,
and the Social Science Research Council; in 1988, sponsored by the Fulbright Program of the
U.S. Department of Education; in 1989, sponsored by the Wenner-Gren Foundation for
Anthropological Research; and in 1990, sponsored by the University of Maryland, Baltimore
County. Research was conducted under the supervision of the Cabildo Indigena de Cumbal.
Luz Angelica Mamian Guzman and Jesus Ivan Villota Bravo of Pasto and Gilberto Hell
Valcnzucla Mites of Cumbal all assisted in the collection of oral histories. Cristobal Landazuri
and his associates at Marka, Instituto de Historia y Antropologia Andina, collaborated in the
collection of archival materials. The Grupo Artistico Los Cumbes of the Vereda Cuaical was
kind enough to allow me to reproduce copies of their scripts.
Notes
1. Resguardo is a legal term for which I hesitate to provide an English gloss (like
"reservation") to avoid misidentification of this specifically Colombian institution with very
different kinds of entities in North America.
2. The eighteenth-century titles of Cumbal and Muellamues were registered in provincial
notarial offices during the late nineteenth century: Notaria Primera de Pasto (NP/P), "Ex-
pediente sobre los linderos del Resguardo del Gran Cumbal", Escritura 228 de 1908; Notaria
Primera de Ipiales (NP/I), "Titulo de resguardo de Muellamues", ff. 336-342, 14 de octubre
de 1885. The boundaries enumerated in the title to Gran Cumbal encompass four modern
resguardos: Cumbal, Chiles, Panan, and Mayasquer.
3. Theatrical presentations of the Three Magi, using Spanish scripts, were widespread in
the Andes and can still be found in some regions (see, for example, Beyersdorff 1988). Until
recently, such plays were performed in Cumbal and in Muellamues. In the early colonial period,
sons of hereditary chiefs trained at special schools in Quito were inculcated with the Christian
doctrine through participation in dramatic presentations. See Moreno Proano (1979) cited in
Hartmann and Oberem (1981).
4. Memoristas are people recognised by their communities as interpreters ofhistory. Most
of these people have been council members, many of them the children or grandchildren of
important leaders. Their knowledge derives from oral histories passed down in their families,
but is also supplemented by the reading of documents in the community archives, as well as
historical publications. For more on memoristas, see Rappaport (1994).
5. Pre-Columbian burials have been discovered throughout Pasto territory. Nevertheless,

it is only when memory is jogged by something as legitimate as a museum poster that obvious
historical referents are incorporated into public displays. In the same way, it was only after the
appearance of "Glory", a film about an all-black Civil War regiment, that the Massachusetts
54th Volunteer Infantry Regiment began to be incorporated into re-enactments of the Civil War
(Washington Post, 20 May, 1990).
6. The traditions of Andean Christs described for the Peruvian and Bolivian Andes (see
Allen 1988, Gow 1980, Nash 1979, Sallnow 1987) or of the Virgin Mary and various saints
among the Colombian Paez (see Bernal 1953, Rappaport 1980-81) are not to be found in
southern Narino, whose indigenous population is considerably more culturally integrated into
the dominant mestizo society. I heard re-workings of biblical stories situating Christ in Andean
landscapes: being forced to drink bitter quinua water instead of vinegar, hiding from the Romans
in a field of the typically Andean chocho or tarhui (lupins), which made noise when he moved
and betrayed his presence. But an Andeanisation of Christ comparable to that of other areas,
where he essentially takes on Andean traits, is absent from Narino.
7. Similarly, voices of the indigenous movement have rewritten native history so that it
corresponds to, legitimises, and is legitimised by the Christian past. Manuel Quintin Lame
(1971), a turn-of-the-century Paez militant whose published works have influenced the native
people of Narino, included indigenous figures in his reworking of the Nativity, much as did
native chroniclers of the colonial period.
8. See Harvey (in this volume) for a critique of the memory-history distinction.
9. Even before its publication, Los pensamientos was a source of inspiration for the native
people of Ortega, Tolima; it was only after repeatedly hearing unusual metaphoric language
uttered by the people of Ortega that editor Gonzalo Castillo-Cardenas (1987: 1-3) began to
suspect that there existed some kind of template for their cliches, resulting finally in his discovery
of Lame's manuscript.
II
MULTIPLE MEDIA:
RITUAL, WEAVING, AND
THE MAKING OF MEANINGS
5
CATHERINE J. ALLEN
When Pebbles Move Mountains

Iconicity and Symbolism in Quechua Ritual
Say it! No ideas but in things.

William Carlos Williams
B
Be Before sunrise, on the Monday before Corpus Christi, many small groups of
pilgrims climb toward the glacial sanctuary of Qoyllur Rit'i in the highlands of
southern Peru. Delegations from many communities bring small icons of crucified
Christ to "visit" Christ of Star Snow in his chapel among the powerful and sacred
mountain lords. They time themselves to end this first leg of their arduous journey
before sunrise; that way they can enter the sanctuary as the first rays of the sun strike
the snowy peaks. There are many aspects to this complex and beautiful pilgrimage,
which epitomises in many ways the contradictory social and cultural experiences of
southern Peruvian peasants.1 Here, I will focus on one small but significant ritual
practice—the "pebble game", the practice of playing with miniatures.
During their climb to the sanctuary, pilgrims stop beside the steep path at a level
place called Pukllay Pampa (the playing ground), where they rest their icons in a stone
windbreak. Michael Sallnow's rich account of the Qoyllur Rit'i pilgrimage includes
this description:
Here there was a strange interlude. .. . Groups of people .. . began to leave the
circle to engage in various kinds of foolery. All around, members of other
naciones were doing the same. Some indulged in general horseplay, leaping and
jumping around amid much shouting and laughter. Others played at games of
make-believe. One of our number bought a cow from a companion, in reality a
lump of quartz, for which he gave a handful of scraps of paper representing
73
74 MULTIPLE MEDIA
money and which he tethered with a strand of wool from his cap. He then took
his cow over to a group from another nation and proceeded to sell it to them in
turn. The original vendor's scraps of paper, meanwhile, had become seeds, which
he went off to plant. Others, meanwhile, occupied themselves privately in the
construction of elaborate miniature houses and corrals out of rocks, with stones
representing llamas, alpacas, and sheep, to indicate to the spirits of the shrine
the devotee's desires. Everyone from our party without exception joined in some
of these antics, though not all with equal enthusiasm. Participation seemed to be
mandatory. (Sallnow, 1987: 190)
I witnessed a similar scene when I accompanied pilgrims from the community of

Sonqo (Paucartambo Province) in 1975, although my companions were somewhat less
spirited than Sallnow's. I was struck by the intensity of the young couples as they
searched for stones shaped like alpacas, Hamas, sheep, and cows and then carefully
placed them in their little stone corrals.2
The next day, at the sanctuary itself, this make-believe continued. Scree-covered
slopes around the small shrine devoted to the Virgin of Fatima were bustling with
miniature house-building activity. There, most of the participants were young men in
a more urban style of dress, speaking Spanish as well as Quechua. Indeed, their little
houses reflected a more urbanised style of life and a commercial orientation. There
were fewer domestic animals and many artefactos—pebble sewing machines, refrig-
erators, televisions, and Volvo trucks. Small groups of these pilgrims carried on games
of buying and selling articles of pebble property for bits of paper representing money.
Sallnow observed that some of these compounds "had written labels identifying them
as the desired property of this or that family", adding that "the combination of this
surreal spectacle and the rarefied mountain air was dizzying" (Sallnow 1987: 193).3
When 1 returned to the Qoyllur Rit'i pilgrimage in 1980 I found that some of the
pilgrims from Sonqo had shifted their house-building activity from Pukllay Pampa to
this area around Fatima's shrine and that their little houses included the artefactos
increasingly present in Sonqo: record players and loudspeakers, sewing machines and
Colernan lanterns.
The Virgin of Fatima, or "Mama Sinakara" as she is called by some pilgrims, is
thought to be an awaq mamacha (little weaver mother, the patroness of weaving).
Women sit next to her shrine, weaving small, doll-sized samples of their handiwork
to leave as offerings. They say that as they weave, their hands "learn" from the
mamacha, and they depart more accomplished in their craft. I should add that one sees
folded pieces of paper inserted in the nooks and crannies of Fatima's shrine as well:
these are letters left by literate pilgrims, who sidestep the process of house building
by writing their desires down on paper. Meanwhile,the taytachas (little fathers, or
icons representing crucified Christ) "visit" their prototype Christ in the chapel in order
to return revitalised to their home communities.
This "play" in the heart of an exhausting and difficult pilgrimage does indeed have
a surreal quality for an outside observer. The house building looks like a child's game;
the women look as though they are making doll's clothes; and the playful buying and
selling with fake money rather resembles the game of Monopoly. It is hard to write a
description of the scene without smiling—and, indeed, the pilgrims themselves enjoy
When Pebbles Move Mountains 75
these activities and refer to them light-heartedly aspukllay 'playing'. Nevertheless, it

is obvious that there is more to these activities; they are neither child's play nor party
games. They give that impression because, out of all the pilgrimage's complex ritual
activity, they are probably the most difficult for a cultural outsider to fathom.
The activities at Qoyllur Rit'i present two different kinds of textuality side by side:
there are miniature iconic texts built of pebbles, and there are abstract symbolic texts,
letters to the saints, in alphabetic writing. Both make requests of powerful beings
(whom I will call "deities" for lack of a better word), but the ways that they com-
municate these requests operate on different assumptions about the relationship of
form, substance, and human action. In the following pages I will explore this
miniaturised "play" activity as a mode of communication between human beings and
deities. Approached as texts, the pebble households are iconic messages presented to
the mountain lords and saints to request specific material favours. I will not explore
the pilgrim's use of writing in any depth (but see Lienhard and Lund, in this volume).
Nor will I analyse a specific pebble text or corpus of such texts. Rather, I will explore
the general process of communication involved in the production of these texts. In the
process I hope to indicate some angles of approach to these textual strategies.
Much Andean textuality has proved notoriously resistent to interpretation. At-
tempts to "decode" pre-Columbian communication devices, such as the kipu (complex
knotted strings used for record keeping) and the toqapu (abstract geometric designs),
have met so far with limited success. It is not simply the extreme abstraction of these
texts that baffles us; it is their semiotics. Pre-Columbian expression operated on
premises about form, matter, action, and meaning quite distinct from those that
produced western alphabetic writing.4 At Qoyllur Rit'i, these two types of textuality
operate side by side, often employed by the same individuals. I hope to show that the
making of miniatures in pilgrimages illuminates some fundamental aspects of a
peculiarly Andean type of textuality through which ideas and values are communi-
cated in presentational forms. I hope that eventually this exploration may contribute
to our understanding of pre-Columbian communication systems like the Inka kipu; on
the other hand, it should provide some insight into the kinds of adjustments and
difficulties Andean peasants encounter as they incorporate literacy into their lives.
I will argue that these textual strategies must be understood as active, interactional
techniques for changing the lived-in world; they include synecdoche (or envelopment
of the whole as part of a larger whole) and play with dimensionality. They are premised
on a principle of consubstantiality, the assumption that all beings are intrinsically
interconnected through their sharing a matrix of animated substance. In other words,
these strategies make sense within a worldview that assumes the inseparability of mind
and matter and attributes animacy of some kind to all material objects.5
Play, skill, and reciprocity

The pebble play at Qoyllur Rit'i is one special example of the pervasive and important
role of miniatures in Andean ritual. These miniatures are viewed as tiny storehouses
of prosperity and well-being. They are carefully tended—"fed" and even on occasion
clothed—and ritually manipulated to bring their keepers the well-being they represent.
76 MULTIPLE MEDIA
Usually these miniatures are bestowed by super-human agencies—a mountain lord,

the earth, or the lightning.6 During pilgrimages such as Qoyllur Rit'i, human beings
are temporarily empowered to create the miniature texts themselves as they "play" at
the sanctuary.
Why does this childlike play—lighthearted yet intent—figure as an important,
even mandatory part of a long and difficult complex of ritual actions? For childlike it
is: the miniaturised activity at Qoyllur Rit'i does resemble the play of children back
in the pilgrims' home communities. During my fieldwork in Sonqo, for example, I
often saw small children playing with pebble "herds"; little girls' first attempts at
weaving were carried out on miniature looms, "as if for a doll".
Bruce Mannheim provides a clue to this problem in an article entitled "The
Language of Reciprocity in Southern Peruvian Quechua" where he reports that
children at play "are said to ayni with God" (1986b: 267). Ayni, a basic form of
reciprocity, has also been called the "ethos" of indigenous Andean culture (D. Nunez
del Prado 1972). Oscar Nunez del Prado comments that "life is an ayni, and must be
returned at death" (1973: 30-31). At the most abstract level, ayni is the basic
give-and-take that governs the universal circulation of vitality.7 It can be positive, as
when brothers-in-law labour in each others' fields; or it can be negative, as when the
two men quarrel and exchange insults. This circulation—be it of water or human
energy—is driven by a system of continuous reciprocal interchanges, a kind of
dialectical pumping mechanism (Allen 1988: 93). Every category of being, at every
level, participates in this cosmic circulation. Humans maintain interactive reciprocity
relationships, not only with each other but also with their animals, their houses, their
potato fields, the earth, and the sacred places in their landscape.
Participation begins during childhood play when an individual begins to acquire
the skills necessary for adult life. This learning comes not through didactic instruction
but through progressively successful efforts to master a skill (Franquemont et al.: n.d.;
Allen 1988: 50). In Sonqo, for example, every adult skill is said to have been invented
by a specific saint; a person who masters a skill—spinning, for example—is said to
be santuyuq (endowed with the attributes of the saint) for that activity. The same idea
can be phrased in terms of stars rather than saints: every skill has its istrilla (star); a
person who is good at a skill is istrillayuq (endowed with the star).8
The statement that children at play "do ayni with God" tells us that a child's first
entry into this universal world of reciprocity occurs outside the context of society as,
through play, she or he interacts directly with sacred power sources (God, saints, stars).
Gradually the person is drawn from this decontextualised play into a network of adult
interactive relationships. An adult person's basic responsibility in life is to use
acquired skills in order to participate in the web of reciprocal exchange. Throughout
adult life, the internalised saint is paid out through skilled work—producing spun
thread, woven cloth, ploughed fields, cooked food, dances, chewed coca, and so forth.
Adults consume the products of each others' skilled work; I consume my fellows'
work (through its products), while my own work is consumed by my fellows.9 Failure
to participate in this circulatory ayni leaves one unable to die properly; one becomes
a kukuchi, or condenado (condemned soul)—trapped in a rotting body and consumed
by the desire to eat one's closest relatives. The kukuchi's insatiable, cannibalistic
appetite exposes it as a kind of microcosm gone wrong and collapsed in upon itself.10
This vision of the cosmic circulation as a complex, organised, and highly con-
trolled system of mutual consumption is important for understanding the pilgrims'
activity at Qoyllur Rit'i. It reveals the significance of play as it recurs ritually in
adulthood. In pilgrimage, one comes into direct contact with the most powerful saints
and with the snow and ice of mountain lords who are normally viewed from a great
distance. It is, to the pilgrims, like walking into the face of god—a liminal state that
returns them to the decontextualised, open, and direct condition of childhood play. It
is not surprising that this ritual play occurs on a place called Pukllay Pampa. Billie
Jean Isbell has shown that the word pampa (a flat stretch of ground devoid of
landmarks) connotes openness and potentiality, a condition of being undefined, as yet
unmarked by distinguishing features.11
On Pukllay Pampa, the 'playing ground', adult pilgrims play freely in the realm
of possibility, on a space of transformation, in immediate contact with the greatest
mountain lords and the powerful shrines of saints. Their play has the effect of
reinvigorating skill and creativity. Thus women sit at the side of the awaq mamacha
and "weave with her hands". Young married couples build their households, and
young entrepreneurs play at commerce.
Pebbles and power objects

Beyond the "pebble game", miniatures abound in the highly charged ritual atmosphere
of Qoyllur Rit'i. We have already encountered one example: the little taytachas
(Christs, literally 'little fathers') brought to "visit" the great image of miraculous Christ
at the sanctuary. At the pilgrimage's end, these taytachas return home revitalised, a
process Sallnow (1974:127) likens to "recharging their batteries". Taytachas and other
images of Catholic saints are thought to store their community's well-being and
prosperity. If denied their feast days, they feel too sad and tired to support the
community.12
Among the pilgrimage dancers are formidable ukuku (bears), many of whom carry
little replicas of themselves. A widespread folktale about the bear tells how a doll helps
the ukuku to fight and subdue a ferocious damned soul (kukuchi). Ukuku dancers on
the pilgrimage face a similar trial, for they spend the night up on the glacier, amid the
hordes of cannibalistic souls thought to inhabit these frozen wastes. Reminiscent of
the folktale, the dolls seem to function as power objects.13
Outside the context of pilgrimage, miniatures and other small objects are also
ritually important. In peasant communities, persons favoured by the apukuna (moun-
tain lords) may acquire power objects of various kinds. For example, at the age of nine
years, an acquaintance of mine in Sonqo found an antique religious medallion while
playing on a hill named Picchu. She recognised it as her istrilla, the source and
storehouse of her well-being (allin kawsay). At the age of fifty, she still treasured it
as a gift from her guardian, Apu Picchu. On the eve of 1st August, she brings it out of
its hiding place and lovingly "feeds" it with coca leaves, red wine, and cane alcohol.
Another type of power object, similar to the istrilla, is the misal, which is guarded
by ritual specialists (paqo) as the source of their power and special insight. I am told
that these misales are usually quartz crystals, considered to be the residue of lightning
78 MULTIPLE MEDIA
bolts. In feet, any place or thing struck by lightning is considered to be possessed of

special and immediate potency, called qhaqha. People or animals killed by lightning
are buried on the spot, and thereafter are revered with offerings of coca and alcohol.
A person who survives being struck by lightning is himself called qhaqha, and
becomes a paqo (ritual specialist). The condition of being qhaqha, and/or the posses-
sion of a misal is what enables the paqo to perform divination—that is, to communicate
with sacred places.14
Certain sacred places are intimately connected with the welfare of domestic
animals. These apu may favour fortunate and deserving individuals with inqa or
inqaychu, small stones shaped like domestic animals. A few places in Sonqo—for
example, a confluence (tinku) of three streams, and some marshy hillsides—are
described as 'possessing inqaychus' (inqaychuyuq). There, the little inqaychus come
out to graze on 1st August, and if you listen carefully you can hear them bleating
'prettily' (munaychata). They allow people with insight to recognise them and take
them home, while the foolish or unlucky see only common stones and pass them by.15
Jorge Flores Ochoa reports that in Paratia, the animal itself enters a fog and disappears
into the ground, leaving the stone behind. He quotes a herder as explaining: "You must
look carefully and quickly throw over them either an inkuna [woven napkin for
keeping coca leaves] or a ch'uspa [woven coca bag]. If you succeed, then the animals
disappear and in their place remain white or black stones, still warm and palpitating"
(Flores Ochoa, I977b: 221; author's translation).
During the miraculous events of 1780, which are said to have given rise to the
Qoyllur Rit'i pilgrimage (Ramirez 1969, Gow 1974, Sallnow 1987), the Christ child
behaved remarkably like one of these inqaychus. One day he appeared as a mestizo
child to a shepherd boy pasturing alpacas alone in the high puna (mountain tundra).
The two children played nicely together; the herds multiplied marvelously, and the
shepherd boy was given food and beautiful clothes. When an official delegation from
Ocongate came to investigate, they were blinded by rays of light shining from his
white clothes, like light reflected from a mirror. When they approached, the Christ
child threw off his clothes and disappeared into a rock, which afterward bore his
imprint: "So in this way, in this rock, our father Jesus Christ entered naked, and from
that time our miraculous father was formed there" (Sallnow 1987:210). Like the pretty
little inqas, Christ allowed himself to be recognised by the favoured but fled from the
foolish, appearing to them as a rock. According to Flores Ochoa: "Inqa is the
generative and vital principle. It is the source and origin of happiness, well-being and
abundance. Herders say that the illa and inqaychu both 'are' and 'have' inqa" (1977b:
218; author's translation).
Arguedas relates inqa to the word inka:
INQA: (incca, according to the traditional spelling) and not "INKA" is how this
word is pronounced by the Canas Indians; and "INQA" not only signifies
emperor, "INQA" is the name for the original model of every being, according
to Quechua mythology. This concept is more commonly known by the term
"inkachu". Then "Tukuy kausaq uywakunaq INKAKUNA" should be translated
as the model or original archetype of every being (Arguedas, 1955: 74, quoted
in Gow, 1976: 199; also see Gow, 1974: 69).
The inqa does not simply provide a model, in miniature, for the living creatures it
represents; as a prototype it gives rise to the animal itself in its vitality as well as its
physical form.16 A household's prosperity is intimately connected with its inqaychus,
which store the fertility and vitality of the herds. They are described as 'caring
protectors' (khuyaqkuna) and are passed from generation to generation, divided
among a couple's children, just as the animals also are divided for inheritance
purposes. There are inqas for other aspects of the household economy as well: one
collection, for example, included smooth pebbles representing potatoes and a triangu-
lar stone described as a michiq warmi (herder woman). The collection also included
carved stone tablets representing the house compound and its fields. Sonquenos say
that these carved inqas, too, have been passed down for generations, and claim not to
know their origin.
On the eves of 1st August, Carnival, and St John's Day (24th June), a family brings
out its inqas to "graze" on offerings of coca leaf and to "drink" libations of alcohol.
While adults comment on the beauty of their little stone herd, their children fondle the
inqaychus and play with them delightedly. Chewed coca wads are called the inqas'
excrement, and are buried in the family's corral.
The inclusion of artefactos—loudspeakers, sewing machines, televisions—in
some pilgrims' pebble households at Qoyllur Rit'i can be understood simply as an
elaboration of these indigenous Andean rituals, practised to secure the well-being of
the peasant herds and crops. In rural high-altitude communities, much of a household's
wealth and local prestige is invested in its herds; peasants who migrate to towns and
cities like Cusco, on the other hand, invest much of their wealth in artefactos. They
find that saving money is pointless in conditions of staggering inflation, while
manufactured articles can be used to augment the family income and often can be
resold. Like the animal wealth of the countryside, they also provide a source of prestige
and personal satisfaction. Seen in this light, the trucks, televisions, and refrigerators
fit less jarringly into the little archetype households left on the mountainside.17
The activity of commerce, so central to many pilgrims' lives, has also been
integrated into this context. Like weavers, who leave Fatima's shrine with their hands
newly instructed, young entrepreneurs ply their commercial craft in miniature—in
miniature, not only because the material goods are represented by pebbles and the
money by bits of paper, but because the participants do it as a game.
Moments of ritual potency

In order to encapsulate personal and community well-being, power objects like inqas,
misales, and istrillas have a connection with mountains on the one hand and with
celestial light on the other. In fact, inqas may be called illa, a word referring to a ray
of light. During their ritual "meal", the inqas should be positioned facing the door
"because that is inti haykuna " (sun's entrance; commonly refers to the east) where the
first rays of light will enter the house in the morning. Interestingly, the house is treated
as a microcosm with its own orientation to the sun; the door need not be facing east.
At ritually powerful times of the year, most importantly on 1st August (near the
sun's anti-zenith passage), many Andean peasants compose a burnt offering
80 MULTIPLE MEDIA
(despacho) to "feed" the earth and sacred places.18 The offering is folded in an
envelope of paper that has a "head" and "feet", like a miniature human being. When
the despacho is burned, its head must point toward the place on the horizon where the
sun is expected to rise, again called "inti haykuna". If not properly oriented, the
offering will pass upside-down or sideways to the sacred places.
Like other ritual practices that are intended to promote prosperity, pebble games
require a proper orientation toward sources of light. The directness of contact with
sacred power sources is facilitated by both the site of the sanctuary, and the time of
year—shortly before the winter solstice when the sun is on the verge of renewal. The
rays of the sun are especially potent during this period. On the Qoyllur Rit' i pilgrimage,
too, sunrise is a moment of great potency and ritual importance. Many pilgrims time
their trek to arrive at the sanctuary at the moment of sunrise; later, at the end of an
all-night walk, the pilgrims dance on a pampa beneath Apu Ausangate at the moment
of sunrise.
In general, the shining of reflected light is felt to be creative, amplifying the realms
of possibility.19 On 24th June, the Feast of St John (formerly the Inka's Inti Raymi, or
Sun Festival), the sun is said to "dance"; stream water is thought to have powerful
medicinal qualities at the moment the sun's light first strikes them, and many people
try to draw water at this very instant. The combination of solstitial sun and glacial ice
is especially powerful. After their all-night vigil on the glacier, the ukukus chip off
chunks of ice at dawn; these are melted down as "holy water" and kept for medicinal
and ritual purposes.
Elsewhere I have written that "(the) power of the inner world is the inverse, a kind
of crystallization of power emanating from the upper world. We human beings live at
the interface, in this world, where exchange and transformation take place" (Allen,
1988: 66).20 Small power objects are particularly potent results of that transformation,
produced at times and places in which both terrestrial and celestial forces are most
powerful. They are created at moments when the configuration of pacha (the world)
is sundered: on 1st August, when the earth opens; when lightning strikes; when the
solstitial sun turns in its course.
Garcilaso de la Vega (1963 [1609], part I, bk. 2, ch. 4) seems to be employing the
same idea when he explains that the word huaca (sacred earth shrine) refers not just
to mummies and earth shrines but to anything whose appearance causes surprise or
fright.21 I was reminded of his statement when, before I left Sonqo, I asked a paqo
how I might continue to grow in insight back in my own country. He answered that I
must dream "after being frightened by the reJdmpago" (lightning bolt). To be startled
is to experience a momentary sundering of one's internal microcosm, a moment which
is potent as well as dangerous.
The pebbles at Qoyllur Rit'i, like the inqaychus on 1st August, come into human
hands at a time of opening, a kind of crack in the world's fabric. Thus, the miniature
houses with their corrals of animals and artefactos are like inqaychus—but in a
curiously reversed way. Pilgrims wilfully enter a powerful place at a moment of
opening and transformation; in their play, it is they who define the identity of the stones
by arranging them into iconic statements of their needs and desires. Then, rather than
keeping them, they leave them behind to be absorbed back into the mountainside.
Conclusion: miniatures and iconicity

We are now in a better position to understand the processes of miniaturisation and
iconic representation that take place in this potently charged playground. Let us turn
back to our opening question: what kind of communication is this, and what kind of
textuality does it involve? That the communication is iconic in nature, is clear enough:
the pebble households are little models of the things they stand for. However, there is
a deeper purpose to this iconicity beyond simple representation.
In his famous discussion of miniatures, Claude Levi-Strauss asks a similar ques-
tion: "What is the virtue of reduction either of scale or in number of properties"? (1966:
23). He concludes that miniaturisation facilitates instant apprehension of the whole
and that this produces a "very profound aesthetic emotion". Miniatures make us
"reverse the normal process of understanding", for knowledge of the whole precedes
knowledge of the parts: "More exactly, this quantitative transposition extends and
diversifies our power over a homologue of the thing, and by means of it the latter can
be grasped, assessed and apprehended at a glance" (ibid.). This analysis illuminates
the Andean miniatures and yet falls short of explaining them in some interesting ways.
Andean miniatures do more than simply delight their owners; they change the lived-in
world. The pebble texts of Qoyllur Rit'i, inscribed on a ground of potentiality and
transformation, serve a communicative and even coercive purpose.
Throughout Andean culture one encounters a pervasive tendency toward envel-
opment, or synecdoche. The Oxford English Dictionary (1971: s.v.) defines synecdo-
che as "A figure by which a more comprehensive term is used for a less comprehensive
one or vice versa; as whole for part or part for whole, genus for species, species for
genus, etc". I am using the term to refer not so much to a figure of speech as to a figure
of thought. Synecdochal thinking22 comprehends the world in terms of mutually
enveloping homologous structures that act upon each other: ayllus are contained in
ayllus; places are contained within places; every potato field contains its own vertical
ecology. Textile patterns contain other patterns in amazing arrays of symmetries and
inversions.23 The word tirakuna (from Spanish tierra plus the pluralising suffix -kuna)
refers to the whole collection of individual named places contained within the world.
The many-armed ridge called Sonqo is one of the tirakuna; but Sonqo itself contains
hundreds of named places who are themselves tirakuna. The scale of one's purview
can expand or contract endlessly.24 Every microcosm is a macrocosm, and vice versa.
In Andean ritual, synecdochal thought works on a world premised on con-
substantiality: all beings are intrinsically interconnected through their sharing a matrix
of animated substance. Thus, small and large imply each other concretely; a powerful
miniature informs the cosmos with its own form.
Pilgrims at play in the charged atmosphere of Qoyllur Rit'i are empowered with
the potential of actually shaping and directing a piece of the world according to their
own wishes and needs. Thus a pebble text actualises the prosperity—the herds, tracks,
sewing machines—it represents. When a pilgrim picks up a pebble at Qoyllur Rit'i,
he is in a kind of world-defining situation. The fact that pieces of the mountain are
presented to the mountain to communicate with the mountain is an extreme conden-
sation of the normal cycle of reciprocity in which deities provide human well-being
and are in turn sustained by that well-being.25 To pick up a piece of the mountain in
82 MULTIPLE MEDIA
this time and place of potency and transformation is to pick up the world itself and
form it in one's own terms, according to one's own desires and needs. Animals,
buildings, and artefactos are ultimately a transformation, or crystallisation, of well-
being that flows from the mountain; in that sense, what the pilgrim does is to shape a
bit of the mountain's potential energy. I believe that this is the crux of iconicity in the
ritual activity.
Epilogue: "No ideas but in things"

"The Spanish brought writing", my friend don Luis told me, just after we had finished
preparing a burnt offering for Mother Earth and the Sacred Places. "The Inkas did not
know writing; they knew stonework". He was praising the Inkas when he said this,
trying to convey to me their power and wisdom. As with many of his deceptively
simple statements, the significance of these words does not easily reveal itself. What
does it mean to "know stonework"? Why should stonework stand in opposition to
writing? And why do many Qoyllur Rit'i pilgrims—especially those who most look
to the Inkas as ancestors—communicate requests to the deities by practising stone-
work in miniature?
The contrast between the two types of textuality—symbolic versus iconic, abstract
versus concrete—brings to the surface deep prejudices in a literate society: we think
of iconic representation as crude and pictographic, a barbaric precursor of civilised
literacy. The Inkas have long stood as a puzzling challenge to this assumption. Such
a complex and sophisticated society can hardly be called "uncivilised"—but why did
they never develop anything we can recognise as writing? To explore the textuality
of stonework—in whatever dimension—is to explore the relationship between the
material world and human agency, between thought and matter—as apparently both
the Inkas and don Luis understood it. I hope that this preliminary exploration of
miniature stonework may indicate some angles of approach to the textual strategies
involved. These textual strategies are active, interactional techniques for changing the
lived-in world; they include synecdoche (or envelopment of the whole as part of a
larger whole), and play with dimensionality, and they are premised on the assumption
that all beings are intrinsically interconnected through their sharing a matrix of
animated substance.
It follows that all action is interactive because all beings are animate; activity is
dialogic and governed by reciprocity. It also follows that the world is subject to
transformation; specific life forms are transitory expressions of a single underlying
substance.26 Specific creatures exist only as long as their interactive relations with the
world maintain and support their given forms. Finally, all representations (verbal,
plastic, and graphic) are themselves active agents in this world of becoming. They do
not just encode—they embody and enact—human thought, memory, and desire.
ACKNOWLEDGMENTS Field research on which this paper is based was funded over a ten-year
period by the Doherty Foundation, the Wenner-Gren Foundation for Anthropological Research,
and George Washington University. A fellowship in the Dumbarton Oaks Center for Pre-Co-
lumbian Studies provided time in which to rethink my ideas and rewrite the paper during the
1993-94 academic year. Many colleagues have helped me through conversations or comments
on the paper, especially Rosaleen Howard-Malverde, Tom Cummins, and Brace Mannheim.
Responsibility for any errors contained here is of course my own. Finally, I extend many thanks
to the individuals in Sonqo who have been thoughtful consultants and gracious friends.
Notes
1. For further sources on the pilgrimage to Qoyllur Rit'i, see, among others, Ramirez 1969;
J. Nunez del Prado Bejar 1970; Gow 1974, 1976; Sallnow 1974, 1987; Randall 1982; Poole
1982, 1984; Getzels and Gordon 1985; Allen 1988.
2. I am using Qoyllur Rit'i as a representative example of a widespread activity common
to pilgrimages throughout the Andean region. According to Rosaleen Howard-Malverde:
The trade in home-made miniatures is a prominent feature of the pilgrimage to

the shrine of the Virgin of Copacabana, Bolivia (6th August); people buy these
from traders, who set up their stalls on the slope leading up to the calvary; then
[they] have the little lorries, houses, even degree certificates and passports
blessed both by the yatiri [ritual specialist] and by the priest. The manufacture
of mini-artifacts is virtually a cottage industry in Bolivia, (personal communi-
cation, 23rd August 1994)
3. See also David Gow 1974:71. According to Howard-Malverde, "the practice of making
artefactos from stone and trading in them also occurs in the context of funeral rites in Northern
Potosi, Bolivia" (personal communication, 23rd August 1994).
4. I am much indebted to Tom Cummins's thinking on this point. See his paper presented
at the Dumbarton Oaks conference on "Native Traditions in the Post-Conquest World" in
October 1992 (Cummins, in press).
5. I developed this argument concerning Andean animalism in my book, The Hold Life
Has: Coca and Cultural Identity in an Andean Community, Washington, D.C.: Smithsonian
Institution Press, 1988.
6. Also see Lapidus de Sager 1968; Flores Ochoa 1977a, 1977b, 1988; Aranguren 1975;
Tomoeda 1988; Allen 1988.
7. For other work on Andean reciprocity, see, among others, Alberti and Mayer 1974;
Isbell 1978; H. Skar 1982. The circulation of energy through ayni has been discussed by several
other observers in addition to myself (Allen 1988). For example, see, among others, Earls and
Silverblatt 1976; Urton 1981; Valderrama and Escalante 1988.
8. By analogy with the English term 'office holder', we might translate istrillayuq as 'star
holder' andsantuyuq as 'saint holder'.
9. Lawrence Sullivan, in his perceptive analysis of South American Indian religion, refers
to this as "an endless cycle of consumption" (1988: 69). The theme of mutual consumption is
powerfully expressed in Andean ritual through the practice of forced feeding (see Allen 1988:
169-175).
10. In the district of Paruro, kukuchi are described in terrifying terms as all-devouring
black holes (Deborah Poole, personal communication, July 1980).
11. See Isbell and Roncalla (1977). In Ayacucho, riddles begin with the phrase, "In that
pampa . . . " (Wag lawpampapi...; Isbell and Roncalla 1977: 25). See also Arnold, in this
volume, for a discussion of the term pampa in textile language.
12. As ayllu members participate in treks and dances, they pour out their energy in a
sacrifice that vitalises their patron deities. See Allen 1988, especially chapters 6-7.
84 MULTIPLE MEDIA
13. For other sources on the ukuku as a figure in folklore and ritual dance, see, among
others, Morote Best 1957; Barstow 1981; Allen 1983; Poole 1982, 1984.
14. Concerning Inka religion, Sabine MacCormack comments: "Lightning and the dead
epitomize a cosmic imbalance tending either to generation or destruction" (1991: 286).
15. Also see Randall 1990: 27.
16. Claudette Kemper Columbus's article, "Immortal Eggs: A Peruvian Geocracy;
Pariaqaqa of Huarochiri' (1990) is also relevant here.
17. Manufactured goods are thought to have a self-hood imparted by their makers (Allen
1982). Although machines—unlike hand-made ponchos, houses, and cooking pots—have no
identifiable maker, they do have a certain kind of animation when connected with their power
sources. The conceptual parallel with domestic animals was expressed, for example, by an
acquaintance from Sonqo who purchased a radio in Cusco and immediately carried it back to
my hotel room to perform its tinka, a form of ritual homage paid to domestic animals.
18. On the relation of Inka ritual activities to Inka astronomy, see the works of R. T.
Zuidema (for example, 1980, 1988).
19. See, for example, Tristan Platt, in this volume. Also see MacCormack 1991, and
Arguedas 1978, in which reflected light is a recurring motif.
20. See, among others, Earls and Silverblatt 1976; Zuidema 1980; Urton 1981; Isbell 1982;
Allen 1988; Valderrama and Escalante 1988; Salomon 1991.
21. Harrison (1989: 45-48) gives an interesting discussion of Garcilaso's definition of
huaca. Salomon (1991: 16-17) provides a good general discussion of huaca as a concept, as
does MacCormack (1991). Zuidema's work explores Inka huacas in great depth (see, for
example, Zuidema 1986).
22. The Oxford English Dictionary goes on to define synecdochism: "b. Ethnol. Belief or
practice in which a part of an object or person is taken as equivalent to the whole, so that anything
done to, or by means of, the part is said to take effect upon, or have the effect of, the whole". I
want to extend this definition beyond metonymic part-to-whole relationships and emphasise
situations in which part and whole are structurally homologous, causing metonymy and
metaphor to collapse into each other.
23. For example, see Ascher and Ascher(1981: 54-55); Cereceda 1986,1987; Franquem-
ont et al., (n.d.); also see the suggestive articles by Hams (1986), Platt (1986), and Kemper
Columbus (1990).
24. See Salomon and Urioste (1991: 14-15).
25. At Qoyllur Rit'i, the mountain is offered a part of itself, perhaps an ultimate case of
Andean ritual forced feeding (Allen 1984, 1988).
26. See, for example, Ossio 1973; Sharon 1980: 170-174; Bastien 1987: 69-74; Allen
1988: 37-66; Salomon 1991: 14-19. On transformation in the cosmologies of South American
Indians in general, see Sullivan 1988.
6
PENNY DRANSART
Cultural Transpositions
Writing about Rites in the Llama Corral
A
An n elaborate ritual event observed by individual families in Isluga, an Aymara
community in the highlands of northern Chile, has its focus of action in a certain type
of corral. It is said by the human actors to be aimed at ensuring the reproductivity and
fecundity of their animals. This ritual has an almost liturgical character since each
family performs the events in a very similar manner, although no sacred text exists.
Incorporated in the ritual are songs, gestures, actions, and material objects (some of
which are used in everyday contexts, but which now take on highly charged roles).
The contextual meaning of these events is not easily translated into words; instead,
the significance for the participants should perhaps be seen in the insertion of the ritual
into a fuller temporal and spatial setting—that is, in its position in the annual ritual
cycle and in its topographical siting within an animated landscape.
If we accept that texts may occur in verbal form and also as the products of other
generative channels, then the events that take place in certain types of llama corral
may be considered to constitute texts. In the Andes, the kancha, or corral, is a site or
space where humans and animals produce words, noises, gestures, and actions. Colour
and shape in the form of material objects (some of which are used in everyday contexts)
now take on highly charged roles and are incorporated into the activities. The emphasis
here is on the unfolding of action or, in other words, the production of a meaningful
ceremony, meaningful at least to the human participants.
Roland Barthes uses textual analysis as a way of understanding a text as the means
for producing what he calls signifiance (moving play of signifiers), and the text in
question is not treated as "philological object, custodian of the Letter" (Barthes 1977:
85
86 MULTIPLE MEDIA
26). Although Barthes is careful not to regard texts as objectified words, the curious
capital L of the phrase "custodian of the Letter" should be noted. It is as though a
residual notion concerning the dominance of verbal over non-verbal performance is
being dragged along with the volatile but generative play in which meaning is
produced.
Catherine Allen (in this volume) argues that pre-Hispanic non-verbal communi-
cation operated on a different set of premises concerning "form, matter, action, and
meaning", which may be contrasted with that which underlies Western alphabetic
writing. She demonstrates how "playing" with pebbles as part of the rituals performed
on a pilgrimage undertaken by the Quechua of Sonqo also shares this different
"relationship of form, substance, and human action". It is my purpose here to
contextualise in particular the visual and tactile aspects of Aymara ritual observance.
I wish to emphasise that it is the very manipulation of colour, texture, and sound that
enables the participants of the ritual to articulate various parts of the ceremony in a
meaningful manner.
The Aymara of Isluga have developed an ability to compartmentalise Western
alphabetic writing from other areas of experience. To Western educationists, most
Isluga people would be regarded as functional illiterates. Writing is part of school
learning, and school is an institution from city life imposed and controlled from
beyond the boundaries of Isluga territory. As an institution, the school and its teachers
are perceived as promoting Chilean city values which, all too often, are seen as being
in opposition to Isluga Aymara values, but to which parents must send their children.
This intrusion from city life is grudgingly accepted in Isluga. All too often, people
gradually lose their reading and writing skills after leaving school.
In contrast, the events in the kancha corral take place in an arena that is inserted
into an Aymara temporal and spatial setting. In other words, they have a temporal
location in the annual ritual cycle of Isluga, and a topographical siting within a
landscape with which Isluga people are intensely involved through their herding
activities. Moreover, they consider the landscape to possess animated qualities. The
events considered here are not underwritten by an inscribed discourse.
The ceremonies observed in corrals are ritual acts of remembrance undertaken by
different individual families. During these celebrations, the greater part of one or two
days is spent within the llama corrals. As the event unfolds, there is disorder, chaos,
and drunkenness. Confusion reigns. Panic-stricken llamas and alpacas jump over and
dislodge precarious dry stone walls of the corrals. All the same, the ceremonies to
which I was invited in January, February, and March 1987, displayed a liturgical
character, for although no sacred inscribed text existed, each family conducted its
individual ceremony in a very similar manner. When the ritual is not being performed,
all that remains is the potential and the willingness in people to perform it, along with
a carefully wrapped bundle of items that is stored away in a dark corner of a house.
It is worth mentioning that Leach maintains we engage in rituals in order to
transmit messages to ourselves (1976: 45), but I would add that the verbal content is
not necessarily apparent or dominant. Aijmer says that "language and culture are quite
different sorts of codes and there is no easy and immediate way of translating from
one into the other" (1987:4). We should, he says, understand non-verbal codes in their
own right, because ritual is the expression of linguistically irretrievable information
Cultural Transpositions 87
(ibid., 12). Therefore, the written text that offered here constitutes a transposition of
activities in the llama corrals into words.1 To advocates of textual theory, all meaning
is contextual. The context of these events is the landscape itself, and in Isluga, people
understand their landscape through their own close association with their llamas,
alpacas, and sheep, which frequent the pasture grounds and water sources.
An Aymara landscape contextualised

The Isluga livestock ceremony reveals two sets of relationships: first, the relationship
between humans and their herded animals on the one hand and the landscape and its
pasture on the other; second, that between human and animal lineages. The former
consists of a married couple and their children, and the latter of herded animals and
their progeny. To a certain extent, human and animal lineages are seen as being
parallel. Both herded animals and humans are associated with mythical points of origin
in the landscape. The former are associated with juturi, hollows considered to lead
from the Inner World, from which camelids or sheep emerge. Each village recognises
several juturi, and there are different ones for camelids and sheep. These juturi are
uywiri2 (herders or providers) which are spatially located in the landscape as springs
or dry places. However, they are inconspicuous and do not constitute imposing
geographical features such as the hills and the awesome volcano that dominates much
of the landscape in Isluga.
In contrast, the generation of humans who lived in the previous world age are
associated with t"aqsu, which are located on ground with sparse vegetation at the
juncture between the gently sloping arid pampa and the rocky, steep lava flows at the
foot of the volcano in the village where I did my fieldwork. T'aqsu are said to be like
small corrals, but people say, emphatically, that they were never used to shelter
animals.
The land—that is, the body of the "Wirjin Tayka", as the Pachamama, or earth
divinity, is referred to in Isluga—is the site from which herded animal and human
lineages spring. Moreover, soil and water produce pasture, which is eaten by llamas
and alpacas who produce fleece. The livestock ceremony symbolises the metaphorical
transformation of pasture into fleece, which is an important component of the material
conditions of Isluga social life. The waynu3 as it is known in Isluga, or the uywa
k 'illpana (the marking ceremony of the herded animals) presents itself as the culmi-
nation of the herding cycle. It takes place in Isluga during the rainy season, and it is a
celebration of the herded animals held by a nuclear family (mother, father, and
children) in honour of the Wirjin Tayka and the uy wiris. The ceremonies are addressed
to the Wirjin and the uywiris by the human supplicants, and husband and wife
respectfully kneel down to ask for permission to proceed at various stages of the long
event.
Because of the complexity and richness of the event, I will not consider all aspects
of the ceremony in detail. Therefore, my text is incomplete. However, I wish to point
out that the combination of components is meaningful to the people who conduct the
events. Seen in another perspective, it is the use of matter derived from the land
itself—products such as fleece, feathers, grass, fragrant firewood, maize, and stone—
88 MULTIPLE MEDIA
that contributes to the liturgical character of the ceremony. The articulation of colour,
texture, and aural effects are the means through which the people of Isluga produce
this particular text.
As an example of how substance is manipulated in Isluga, the mother of the family
offers smoke from fragrant embers to the Virgin (the Christian Virgin is implied),
while both mother and father offer coca leaves to the Wirjin Tayka. The father grinds
black and white maize kernels, and the flour is mixed in separate cups with water.
These mixtures of maize flour suspended in liquid are known as ch 'uwa; the black
mixture is offered to the T'alla (the female uywiris) and the white to the Mallku (the
male uywiris). Thus the human participants ask the powerful Wirjin Tayka and the
uywiris to grant them an audience with them and their animals.
Such is the awe and respect with which these beings are held in Isluga that
participants, especially the mother and father of the family observing the waynu, are
expected to achieve a state of ritual drunkenness. Thus the long all-night vigil is an
important part of the ceremony which precedes the colourful events that take place in
the corral. The mournful sounds produced during the vigil and in the corral form an
essential complement to the strong colours (red, pink, and orange and tonal gradations
of green, red, and orange).4 Traditionally, Isluga people avoid the excessive use of
colour in daily life, employing it only in certain contexts (Dransart 1988:44-48). Also,
alcohol is not normally consumed on non-festive occasions. Hence the abundance of
strong colour and the need for humans directly to address the uywiris, with their higher
status, means that a state of drunkenness is enjoined upon the participants.
The waynu ceremony in Isluga

Isluga families usually celebrate two waynu ceremonies, one for their camelids, which
takes place between New Year's Day and Carnival, whenever that may fall, and
another for their sheep, if they possess any, which usually takes place immediately
after Carnival. Here I will concentrate on the former, the waynu of the llamas and
alpacas. The ceremony takes the form of a symbolic investiture of the animals, and
the ritual "dressing" of the animals is accompanied by songs and mandolin music,
which are designed to ensure the fertility of the herds. Without this music, it is said,
the ceremonies would have little effect. Each family has its own date, which must fall
on a Thursday or Saturday. These days of the week are known as Dia Compadre
(godfather's day) and Dia Comadre (godmother's day) respectively; the former refers
to the day of the Mallku and the latter to the day of the T'alla.
The waynu ceremony requires a considerable investment of labour in preparing
for the event, and some families may celebrate it every other year. Alternatively, they
may conduct an abbreviated version known as ch 'allta one year, and the waynu proper
the following year. Individual families have some flexibility in deciding how often
and when to observe the ceremonies. The health and well-being of the animals is a
decisive factor, for if they suffer ill health or meet with misfortune, a family may well
decide to change its customary date.
Preparations begin in advance with the making of chicha, a maize beer, and of the
sarsillu, chimpu, and wistalla with which the animals will be "dressed". Sarsillu are
ear ornaments, in the form of tassels of coloured yarn, in which red predominates.
Four different types are made for different categories of llamas and alpacas: tilantir
sarsillu for the guide animals, often older females with no offspring to accompany
them; tama for the female animals of the herd; wantilurita or "flags" for the male
animals; and sombreros, pompons that sit on top of the ear, also for the males. Chimpu
is unspun llama or alpaca fleece, dyed red, pink, or orange; yellow is sometimes used
as well. It is tied to the backs of the llamas and alpacas. In the case of female animals,
chimpu will be tied in a cluster over the withers, but in male animals it is tied lineally
along the length of the spine. The third item, wistalla, is a neck piece which consists
of a series of coloured yarns in tones of pink, green, and orange, which hang from a
thick cord that will be tied round the neck of the camelids (the word "wistalla" is also
used for a small bag used by women and men for holding coca leaves). It is important
that all these colours be strong and clear; any faded colours are rejected as being useless
for this ceremony. The responsibility for the making of the maize beer and the tassels,
neckpieces, and chimpu lies with the women of the household, but the men often make
some of the items, especially the sarsillu called "wantilurita" and "sombreros" (which
will be worn by the male animals), and the wistalla. Neighbours may lend a hand with
these preparations.
The ceremony begins with an all-night vigil during which the father and mother
of the household may go to visit the juturi and pasture grounds that their herds are
accustomed to frequent. The juturi, which are said to be the owners of the animals of
both husband and wife, must be properly acknowledged. Offerings to the juturi may
include flamingo feathers which are burnt in the hope that the camelids will behave
like flamingos, which habitually cluster together in cohesive flocks. The feathers are
burned so that the llamas will not wander, but will stay close together. In the meantime,
the rest of the family and other relatives and friends sing mournful songs inside a
house, where they make libations of chicha and alcohol, and drink. As dawn begins,
the people are still inside and maintain the vigil. It seems like a wake.
Early in the morning, the llamas and alpacas are released from a large corral where
they have spent the night. As they leave, their owners sprinkle the ground with liquid
libations. A young woman and a young man herd them out onto the wet bofedal
pastures and beyond to the dry scrub pasture grounds. Later, about midday, they will
bring them back to the village, this time to a kancha corral, to where the focus of
attention turns.
First of all, the wife enters the kancha, moving in an anti-clockwise direction round
the central stone and wafting the smoke of fragrant-smelling embers to each of the
four corners. The kancha is an enclosed space in which ceremonies designed to
encourage the reproductive vitality of male and female animals will take place, and
from which the animals with their ritually enhanced reproductive powers will emerge
in due course.
The llamas and alpacas enter the corral. An animal is selected for slaughter, but
first its back is sprinkled with chicha, maize flour, and alcohol. Brightly coloured
chimpu fleece is tied to its back. Then a neighbour or a relative cuts the jugular of the
sacrificial llama, and spoonfuls of the blood of the dying animal are scattered over the
ground of the kancha by the husband, who directs the rest of the blood into a hollow
dug into the ground, known as a wina. All persons present lie down, head to the ground,
90 MULTIPLE MEDIA
in an act of obeisance to the uywiri. This silent act of remembrance is usually

performed along the rear wall of the corral.
Later on, the animals are marked. The marking takes two forms. First, llamas and
alpacas are "dressed" with chimpu (the dyed fleece; this word really means 'mark'),
ear adornments, and neck pieces of coloured yarns. Second, animals that are approach-
ing sexual maturity—that is, animals born in the previous birth season—have an
owner's mark cut into one or both ears. The ear tassels which are pierced through the
ears of the animals cause the ear to bleed slightly, but the ears of the one-year-old
animals are made to bleed profusely. The ceremony is therefore a rite of passage for
the animals at the time of their initiation into sexual maturity. It parallels the ear
piercing ceremonies and the ritual investiture of the breechcloth which are recorded
for Inka noble boys (held, admittedly, at another time of the year, as part of the Qapaq
Ray mi celebrations at the time of the winter solstice). The initiation of the llamas and
alpacas thus becomes a celebration of the social identity of the herded animals, and
the recognition of one of the stages in the process of achieving "animalhood"
experienced by all herded animals in Isluga.
This part of the ceremony, the marking, will probably not finish until the following
day. After the female animals are marked, they are allowed to go free. The next
morning, the ritual misa table is transferred inside the corral. Then the male animals
are marked and finally allowed to join the females. In the intervening evening, a group
of men go off to make a symbolic journey to the valley. They return, playing the role
of a herder and a caravan of llamas. As they enter the village, the herder attempts to
keep his "llamas" under control, but they escape to chase young women as they
endeavour to indulge in sexual buffoonery by mimicking the copulation of llamas.
The mood and pace of events changes, and the apparently dirge-like and repetitive
music that took place inside the corral yields to dancing and cheerful music. A
communal meal to which all are invited is served. This is called a comilona (feast) or
boda (wedding).
The llama corral as site and source of

regenerative vital ity
Having presented an account of the waynu ceremonies, let us focus our attention on
the kancha, with its creative inner space. This, after all, acts as the nexus for the
ceremonies, within the wider context of the landscape. At first sight, the kancha seems
like a simple rectangle of casually built dry stone walling, which is often in a state of
disrepair. However, it is designed to suit the needs of the waynu ceremony, and it is
furnished with some significant features (figure 6-1). The majority of the kancha in
Isluga are constructed so that the entrance faces east, toward the rising sun like most
of the houses. Some, however, have an entrance facing north, looking toward the
mighty, multiple summits of Mallku Kawaraya. Most are furnished with a small, more
or less square stone, which is placed in the interior space of the kancha. They all have
a large, flat stone placed outside, to the left of the entrance as one looks in, but on the
right of the entrance from the point of view of someone sitting behind the stone, with
Key
Wina Hollow for containing the sacrificial blood of a llama
Qihwa A tree which grows at high altitudes in the Andes
T'uju A niche in the corral wall at ground level
Mamaqullu misa The ritual table which is laid out on a large flat stone for the part of
the ceremony during which the female animals are symbolically dressed
Tataqullu misa The ritual table which is laid out on the ground inside the corral for
the part of the ceremony during which the male animals are ritually dressed
Ch'uwa A series of dishes filled with the so-called ch'uwa libations
Figure 6.1 The design of the kancha (llama corral) in Isluga.
one's back to the kancha wall. This large stone serves as the mamaqullu misa, the ritual
table for the female animals.
Some kancha are provided with a niche (t 'uju) in the middle of the rear wall, low
down on the ground, directly opposite the entrance. This is the place where the ch 'uwa
libations are placed in a double row on the ground. Before the llama or alpaca is
sacrificed, a hollow is dug in the ground between the central stone, on which embers
are placed. The ch'uwa dishes are arranged in front of the t'uju. It is between this
blood-filled hollow and the rear wall where material objects are placed that the people
lie on the ground in silence, withholding speech and song. In this instance, the material
objects consist of clay dishes and enamel cups containing ground substances mixed
with water. These dishes hold the substances that will be libated to the earth and hills;
92 MULTIPLE MEDIA
they are evidently positioned to play a mediating role between humans and supernat-
ural beings. Once again, the human supplicants hesitate to address the powerful Wirjin
and the uy wiris; instead, they prepare to proffer material substances that will be ritually
sprinkled on the ground. For the purposes of the ceremony, branches of qinwa trees
are placed on both sides of the entrance to the corral, which is closed when the animals
are inside by means of a rope held taut across the walling, over which a blanket is
draped. Qinwa is a high-altitude tree which grows in sheltered places; locally it can
be obtained from a great valley on the volcano above the village.
As an outsider who has read the work of Santacruz Pachacuti Yamqui
Salcamaygua, an early seventeenth-century author of native Andean descent from the
Lake Titicaca area, a striking visual analogy came to my mind. This is, of course, not
a connection which the people of Isluga would make, although they do recognise
certain archaeological sites as being the work of the Inkas. The view of the entrance
to the Isluga kancha, flanked by boughs of trees, with the interior t'uju niche behind,
resembles the central part of the early seventeenth century drawing by Santacruz
Pachacuti of the three exits from the inner world through the Paqaritampu (the house,
or inn, of origin) from which the first Inkas crawled into this world (figure 6-2). The
central part of the drawing depicts a rectangular geometric design flanked either side
by a tree, representing Apo Tambo and Pachamamaachi, according to the written
account. The ancestors were perceived as the progeny of the earth itself, that is, of the
Pachamama, and the lineages were expressed visually as rooted and branching trees.
On the lower left and right are two more rectangles of simpler design. The three
openings or "windows" are named by Santacruz Pachacuti as Tampottoco,
Marasttoco, and Sutittoco, belonging respectively to the uncles, maternal grandpar-
ents, and paternal grandparents of the first Inka ruler, Manku Qhapaq, who was said
to have emerged from the cave with his three brothers and four sisters.
It would seem that the three exits were the points of origin for the royal panaka
lineages of the Inka rulers.5 The drawing by Santacruz Pachacuti refers to human
lineages. Yet the word ttoco as written by this author is the same as the modern "t'uju".
The focus of the elaborate wayfiu ceremony in Isluga turns on the generative interior
space of the kancha, where songs are sung, bright colours are used, libations are
offered, and the symbolic dressing of the animals takes place to ensure that lineages
of camelids multiply. The qinwa tree, besides providing a genealogical metaphor, is
also a symbol of the source of life and of longevity, and, according to Sherbondy
(1988: 112), as a strong tree which is resistant to the cold, it was probably the symbol
of Imperial Cusco.
With this emphasis on the importance of lineage, the root of the word "wina", the
hollow in the corral which is filled with red blood and green coca leaves, should be
explained more fully. Winaya is the Aymara and winay the Quechua word listed in
early-seventeenth-century dictionaries as 'always, eternal, forever'. The Quechua
dictionary of Goncalez Holgum (1952 [1608]: 352) adds 'the generations and descent',
while Bertonio translates vinayana vinayapa into the Latin saecula saeculorum
'generations of generations' (Bertonio 1984 [1612] bk II: 388). Since saecula implies
'successive generations [of people or animals]' (Simpson 1973: 529), my understand-
ing of the wina hollow is that of a source of vitality which is regenerated in the llamas
and alpacas through the ritual performance of the waynu ceremony.
The very design of the kancha reinforces the idea that the creative interior is the
space from which ever-multiplying lineages of camelids will emerge. The fact that the
ceremonies are undertaken by a husband and wife in partnership suggests that the
animal lineages are considered to be parallel to human ones. Llamas and alpacas
require constant attention, as do human children, and herders worry constantly about
their herds lest they go hungry, fall into a river and drown, or fall prey to predators.
However, imagery manipulated by the people of Isluga is multi-faceted, since the ritual
term by which llamas and alpacas are designated during the waynu is p "aqalli 'flower'.
A herder may say, "these are my flowers".
The simple idea of a bud forming and blossoming forth appears to be replete with
meaning that has lasted over the centuries through changing social contexts in the
Andes. A pair of Inka ceramic vessels with a long flaring, flower-like neck are painted
with a row of stylised women holding birds, above which are flowers (Purin 1990,
2:193). We should recall that some of the Inka empresses depicted in the early
seventeenth century by Martin de Murua (Ossio 1985) and Guaman Poma (1980) show
Figure 6-2 The Paqaritampu (the house of origin) of the Inkas, drawn by Santacruz
Pachacuti Yamqui Salcamaygua, 1615. Reproduced by permission, Biblioteca
Nacional, Madrid, ms. 3169.
94 MULTIPLE MEDIA
a woman standing holding a flower. It is clear from the text that Guaman Poma is
interested in the importance of lineages. However, his drawings are more suggestive
than the text, for in some cases—for example Mama Huaco and Mama Anahuarque—
the empress is shown inside a room with two windows, one at each side of the monarch
(figures 6-3 and 6-4). My interpretation of the scene is that Guaman Poma is presenting
in visual form the theme of the procreation of lineages, which originally in Inka myth
emerged from the body of the Pachamama through the cave of Paqaritampu. The room
stands for the cave, and the windows are the two lateral entrances, while the body of
the empress contains the vital inner space from where the lineage or panaka emerged
into this world.
It is worth recalling that Gabriel Martinez (1976:271) pointed out that the Aymara
verb amtana and the Spanish recordar are both used as ritual terms in what he calls a
liturgical act which is full of meaning. He cites the work of Rodolfo Kusch, who draws
a parallel between the root amu, which appears in Bertonio's dictionary as amu,
meaning 'boton dela flor' (flower bud) and amutaatha 'acordar' (to remember, to
recollect) (Bertonio 1984, bk II: 17). This clearly indicates that the acts of remem-
brance honouring the Wirjin Tayka, uywiris, Mallku, and T'alla that I have analysed
are not just an acknowledgment of their existence. When a supplicant lists a litany of
names of the places frequented by his or her llamas, and then adds "Uywir mallku
amtata jan amtata, Uywir t'alla amtata jan amtata" (Male uywiris remembered and not
Figure 6-3 The first Queen Quya of the Inkas, Mama Huaco, drawn by GuamanPoma
de Ayala.
Figure 6-4 The ninth Queen Quya of the Inkas, Mama Anahuarque, drawn by Guaman
Poma de Ayala.
remembered, female uywiris remembered and not remembered), it indicates that the
person who has uttered these words is actively involved in feeding and nourishing
these beings, just as the Wirjin Tayka and the uywiris are held to nourish fortunate
supplicants. Equally, the fact that families invited me to accompany them in their
wayfiu involved active participation. One cannot be an observing bystander in Isluga
without being made to feel intensely uncomfortable.
Acts of remembrance, acts of negation

The events in the llama corral are not, as I said, underwritten by an inscribed text, and
yet I appealed to some visual parallels from early seventeenth-century records. This
would seem to imply that certain formal, visual aspects have persisted over some time.
However, the seventeenth-century Quechua context and the Aymara context of the
late 1980s are discontinuous in time and space. One of the reasons for proposing this
analogy is that visual manifestations may be endowed with a disquieting character
which admits of multiple meanings. If they are considered to constitute cultural texts,
then they do not serve to fix essentialistic meaning. Yet, in Isluga, the use of the
material objects serves to underpin the articulation of sensuous phenomena as the
ceremonies proceed. This would explain the apparent longevity of the imagery and
also account for my observation that the different families who invited me to their
waynu all performed the ceremonies in a very similar manner.
96 MULTIPLE MEDIA
Since I was first introduced to Isluga in 1986,1 have no long-term perspective on

how the ceremonies might change through time. On a return visit to the village in
1989,1 found that some of the families of the village where I did my fieldwork no
longer have the willingness to perform these ceremonies. They have become evangel-
ical Protestants, a cult that shuns the taking of alcohol and dancing, and which tells
its membership to forget the Wirjin Tayka and the uywiris. I am using the verb 'to
forget' in the sense of positive rejection, in contrast to the Aymara amtana, to
remember.
The events I observed in 1987 in the llama corral took place against a Catholic
background which, through its inability to intervene in village life, was tolerant of
non-Christian Andean beliefs and practices. Throughout my main period of fieldwork
from 1986 to 1988, the village maintained a staunch anti-Protestant stand of which it
was proud. Indeed, the mother of one of the families which later converted to
Protestantism was most insistent that outsiders from the city (for example, the district
nurse and government agency officials) should attend her waynu; she seemed so proud
of being able to show to them that this was her Aymara-ness and that she was in her
domain. And it was clear that the outsiders were reluctant to accept the proffered copita
of alcohol; they only stayed to take one gulp, barely hiding the grimace, before they
left to return to their domain.
The full implication of the co-existence of those who remember the Wirjin Tayka
and those who do not remains to be explored in Isluga.
Conclusion
To return to my opening remarks regarding textual theory, I wish to offer some
comments. Proponents of textual theory such as Julia Kristeva concentrate on the
speaking subject as producer of verbal utterances, where language is considered to be
a complex signifying process rather than a monolithic system. Similarly, for Barthes,
signifiance is a process in which the text is read as "a moving play of signifiers". Thus,
oral practice results in new combinations and transformations of what went before.
Yet the waynu ceremony, with its remembrances and songs, seems to revolve round
non-verbal imagery, for some of which parallels can be found in seventeenth-century
sources.
There is an apparent contradiction here where the potentiality for verbal perfor-
mance to change is matched by invariance in certain aspects of non-verbal perfor-
mance. In examining the relationship between words and images, let us consider the
well-known essay entitled "Rhetoric of the Image" by Roland Barthes. Barthes is
moved, when confronted by a visual image, to suggest that "the image is felt to be
weak in respect of meaning" (1977: 32) while dismissing the question whether
"signification cannot exhaust the image's ineffable richness". The image he is consid-
ering is extremely banal: it is, in fact, an advertisement for Italian pasta. Yet, in a
footnote, he reveals that he is worried about the banalisation of the verbal content, not
that of the visual content (p. 40). Barthes wonders if there is always textual matter
under or around the image, and he recommends "go[ing] back to partially illiterate
societies, to a sort of pictographic state of the image" (p. 37, emphasis added). To talk
in such terms is to prioritise the written word, for Barthes claims that the linguistic
message is present in every image, and that writing and speech constitute the full terms
of informational structure (p. 38). I seriously doubt that the verbal context is as
dominant as Barthes suggests (how else would I be moved by paintings by Mark
Rothko—which I am—without feeling the need to read the title underneath?). My
experience of Isluga life is that people have very acute and developed visual
awareness, and they are producers of visual phenomena that cannot easily be
translated into words. The wayftu ceremony is a vehicle of symbolic thought which
brings together images and sounds that are constellations of significant signs display-
ing varying degrees of abstraction. This thought pervades all other areas of life and
all aspects of herding technology, linked, as it is, to the use and knowledge of the
animated landscape. Until recent developments in visual anthropology, the discipline
of social anthropology has been hampered by its reliance on mere words. I wish to
stress the power of imagery. And on this imperfect text which I have written, I rest
my case.
ACKNOWLEDGMENTS I gratefully acknowledge the receipt of scholarships from the Pirie-Reid

Fund, and from the Emslie Horniman Fund, which enabled me to undertake fieldwork in
1986-88, and for travel expenses from the British Council in Santiago in 1989.1 would also
like to thank most warmly the people of Isluga who made me feel at home with them during my
fieldwork. This account is based on ceremonies observed in a village belonging to Araxsaya,
which is the Upper Moiety of Isluga in the highlands east of Iquique, Chile.
Notes
1. In this article 1 am concentrating on the whole ceremony, especially on the ritual
action and the space where it takes place. Songs with a verbal content are, of course, an
important component (see note 3). At the time of the marking of the animals, different songs
to the guide animals are sung by the people in a constantly repeated refrain. However, on
this occasion I wish to emphasise the over-arching non-verbal content of the livestock-
marking ceremony.
It is also worth pointing out that the non-human participants of the ceremony contribute
their utterances, in the form of a whinny or, in quieter moments, a mournful "y-y-y" (as written
in Spanish; the English equivalent would be an 'e-e-e') sound. While I was herding, an Isluga
woman said her llamas were crying when making this noise. This would be in keeping with the
dirge-like songs sung by the humans in the llama corrals during the ceremony. It is all very
reminiscent of Guaman Poma's drawing of the Inka emperor and empress singing in a similar
fashion to their red llama (Guaman Poma 1980: 320).
2. In Spanish they are also known as aviador, which has been translated as 'el o ella que
provee' (he or she who provides) (Martinez 1976: 281).
3. Strictly speaking, waynu is the term for 'song'. However, it is also the word adopted by
Isluga people for referring to the livestock-marking ritual discussed in this article. Sound and
colour are the two important motivating elements, without which the events would have no
effect. My concern here is to examine the visual and tactile aspects of the ceremony. For a
corresponding consideration of the role of music in the equivalent ritual observed by the northern
neighbours of the Isluga people, see Manuel Mamani Mamani (1990).
98 MULTIPLE MEDIA
4. See the next section for a description of the colourful ear tassels, neckpieces, and dyed
fleece with which the animals will be "dressed". The use of colour during the wayfiu and its
relationship to woven textiles is discussed more fully in Dransart (1991).
5. Panaka were royal lineages that have been interpreted by R. T. Zuidema as exogamous,
matrilineal groups, in which a man belongs to his sister's group, hence his children cannot
classify themselves with him; this is discussed further by Rostworowski (1986: 138-141).
7
DENISE ARNOLD
Making Men in Her Own Image

Gender, Text, and Textile in Qaqachaka
A
A number of recent ethnographies, though distinct in time and space, have compared
and contrasted the weaving of cloth, as women's work, with men's writing of texts,
as two hierarchically related orders of gendered activity.1 Moreover, this gendered
distinction is often implicit in the wider debate about orality and literacy, in the
supposed superiority of "literate" societies as opposed to those with more "oral"
cultures. This chapter will challenge such a hierarchical viewing of these gendered
activities in the Andean case, where even a distinction between weaving and the
writing of texts is problematic. In Andean studies, the existence and status of "written
texts", both before the Spanish conquest and immediately afterward, is a subject of
continuing debate (Murra 1975a and b, 1989; Adorno 1986; Platt 1992 and this
volume; Arnold 1993; Ferrell, 1994). Moreover, the increasing interest in the pre-Co-
lumbian mnemonic technique of recording information as diverse as bureaucratic
records, genealogies of the noble Inkas, and tales of their exploits on the knotted
threads called kipu in Quechua, or chinu in Aymara, has shown that this characteris-
tically Andean art of memory bridges any supposed divide between weaving and the
writing of texts (Cereceda 1987; Zuidema 1989; Arnold and Yapita 1992; Parssinen
1992).
When we examine the same proposition in relation to the contemporary men and
women of Qaqachaka, an ayllu commmunity in highland Bolivia (department of
Oruro), the case is just as problematic. In Qaqachaka, both men and women weave
cloth, although the cloth that women weave technically is the most complex, and with
the introduction of compulsory education in 1952, both girls and boys have had the
99
100 MULTIPLE MEDIA
opportunity to learn to read and write. But even though historical evidence suggests
that women's weaving and men's writing of texts have been different orders of
gendered activity in the past, there is no evidence that the two activities were related
hierarchically; rather, they were viewed as parallel domains of gendered activity.
I shall suggest that these different orders of gendered activity may derive instead
from some perceived metaphysical differences in memory and intelligence, whereby
men are believed to think "with their heads" and women "with their hearts" (Arnold
and Yapita 1994: 53ff). It is a common expression in Qaqachaka, for instance, that
men wrote their historical texts with plumed pens and blood ink, using the force of
spirit in their heads, whereas women "wrote" their textiles with their weaving picks,
using the memories and inspiration held in their hearts.
I shall examine how these particular gendered differences in intelligence are
evident not only in such tasks as weaving and writing, but even, for example, in certain
garments which men and women still make for themselves. However, the main
argument here challenges the idea of any hierarchy between women's weaving and
men's writing of texts by showing how contemporary women weavers in Qaqachaka
mediate any such hierarchy simply by using their weavings as a form of text. In
weaving certain gendered garments for themselves and for their menfolk, in creatively
manipulating their woven texts and the symbolic language embedded within them,
women are able to order and define, within the symbolic domain, the relative powers
of production and the relative generative powers of reproduction of each gender, and
to decide on their relative hierarchical value. A woman weaver is thus able to define
in practice the limits and obligations of both genders as social constructions in
Qaqachaka society.
In the first section, I examine the activity of weaving in its broader social and
cultural context. Here I shall show how common ideologies of gender, and ideas about
creation, transformation, production, and reproduction, are not only embedded in the
finished cloth as a kind of "text", as Cereceda (1978) and others have shown, but in
turn form the basis of the spoken texts: descriptions, narratives, myths, songlines,
sayings, and so on, which accompany such practical tasks and relate them to other
associated tasks. In the second and third sections, I focus on some specific represen-
tations of gender in two distinct woven garments: the woman's over-mantle, called
awayu in Aymara, and the man's poncho, a term borrowed into Aymara aspunchu.
Aspects of gender will be compared in the two garments as items of dress. In particular,
two characteristic patterns of weaving from these finished textile garments will be
examined as mediators of a symbolic language about gender, production, and repro-
duction. In these final sections, I shall pay particular attention to the women weavers'
own discourse about the use and significance of these gendered garments.
The social and cultural context of weaving

The arts of the hand and the arts of the voice
Weaving as an activity is related to other practical tasks, and these interrelated
activities give rise in turn to the interrelated "texts" that accompany them: whether in
the more formal performance of songs or tales, or simply in the multilevelled analogies
Making Men in Her Own Image 101
expressed in the more everyday comments and sayings that interweave the different
tasks in a complex "intertextuality", which others have described (see, for example,
Tedlock and Tedlock 1985; Howard-Malverde 1989; Hanks 1993 after Kristeva
1980).
The Tedlocks, most particularly, have noted how the practical arts of the hand are
necessarily interrelated to the arts of the voice. They criticise a Saussurian semiotics
as describing a closed code that exists prior to any of its particular manifestations in
the material world, disengaging language and the verbal arts from hand and tool and,
ultimately, from face and voice. Semioticians, they argue, merely follow codes across
the boundaries of various genres, and human action from this point of view is merely
reducible to the terms of the code. They prefer instead the work of Bakhtin who
examines "interrelationships" rather than "codes", or the work of Julia Kristeva on
"intertextuality". The Tedlocks use the notion of intertextuality to explore the rela-
tionships between, and among, various arts of the Quiche Maya of Guatemala,
pointing out that rarely is an adult specialised exclusively in just one activity. Rather,
they practice and collaborate on several of them, interrelating the various levels of
their performance.
LEARNING TO WEAVE, LEARNING TO WRITE

In Qaqachaka, too, this crossover between different activities occurs. I have mentioned
elsewhere how the building of a house is compared to the setting up of a loom, and
how, in a much broader sense, certain weaving designs and their modes of repetition—
in reflections, serial repeats, staggerings, and bilateral symmetries—are repeated
throughout distinct media. Thus, homologous configurations can be found in the
different garments of rustic woven and knitted cloth and in sophisticated modern
factory-made jumpers, in braided slings and ropes, in dance choreography and the
patterns of song verses, or in the decoration of simple pots and the decorative elements
of urban architecture (Arnold 1992a and b, 1994).
It is not surprising, then, that such analogies are made between women's activity
of weaving and men's activity of writing. Moreover, a parallel manner of learning
these two different activities is acknowledged. Before 1952, there was a distinct
gendered division of labour, whereby boys, but not girls, tended to be sent to school
to learn to read and write. It was thought that boys had greater need of these skills than
girls in their journeys to the cities and, even more importantly, in order to read and
defend their title documents to the land. Moreover, a girl's labour in herding was
considered to be too important for her to have time to go to school. Her knowledge
was focussed instead in her weaving and her recognised ability to "read" the ayllu
textiles. In a parallel manner, too, boys were believed to learn to read in school by
exercising the power of spirit (ispiritu) in their heads, like hens—all birds are
associated with spirit—pecking seeds of knowledge from the ground, whereas girls
were believed to learn to weave by exercising the inspiration and memory (amta) held
in their hearts. And in the parallel processes of learning, boys learnt to read and write,
first of all, letters, then syllables and words, and finally sentences and whole pages of
script, enabling them to understand and interact with the world outside the immediate
ayllu, above all with the urban domain and the offices of city bureaucrats. A girl, on
102 MULTIPLE MEDIA
the other hand, learnt first of all to weave small garments, such as skirt and belt
fasteners, manipulating just a limited number of two to five warp-threads (chinu) to
make easy designs on her first simple loom. Then, as she grew older, she learnt to
weave more complex cloth, moving from one garment to another until she reached
the highest degree of complexity in her mantle, with designs of up to one hundred
warp threads. At the same time, the range of meanings of her designs, and her own
discourse about them developed to increasing degrees of complexity (see also
Rodriguez 1994).
In Qaqachaka, girls initially learn their weaving designs from the women who
surround them—from their mothers, aunts, and grandmothers. Girls often form a
special apprenticeship to an older weaver, who shares her weaving skills in return for
help in pasturing her animals, and she often recalls this relationship between weaving
and pasturing in her first song performances in public, later in her adolescence. They
also learn from contemporary apprentices, girls of the same age, whom they respect
as weavers, as they are pasturing their animals in the hills. Young girls begin to spin
wool from the time they are large enough to hold and spin a spindle. The first weaving
venture is usually undertaken when a girl is about eight or nine, when she weaves her
first skirt-fastener or hat-band. Both of these weavings, small as they are, introduce
the basic understanding of many more complex weaving techniques that they will
require later on. Afterward, a girl rapidly moves on to a belt, a small food bag, and
then to her first mantle (awayu). However, she weaves her first poncho only when she
is married or has acquired a permanent male partner. The weaving techniques and
designs that a girl has learnt in each of these early textile ventures are still evident in
her awayu when she is a mature woman, married and having set up her own household.
The design section of the awayu continues to incorporate the small design bands,
reminiscent of her early skirt-fasteners and first belt designs. And she often carries
some of her own mother's designs into her mature mantles.
This is not to say that a boy does not develop skills comparable to a girl's weaving.
His musical ability develops, in the playing of various flutes and pan-pipes, incorpo-
rating musical elements that are often compared to weaving designs (see Stobart
1987). And young boys quickly learn to imitate their father's technical ability to knit
the fine and brilliant knitted hats known throughout the region, and to braid ropes and
slings. They, too, begin with simple designs of just four or eight threads, developing
their technical abilities until, in the most complex braided ropes and slings, grown
men are able to manipulate complicated designs of bilateral symmetry based on a
minimal scheme of sixteen component threads, working up to seventy-two-thread
designs.
THE GENDERED DIVISION OF LABOUR IN WEAVING

In this way, the cultural rules that differentiate activities by gender give rise to different
gendered domains, social realms with their own distinct lexicons, utilised more by men
working with other men or by women working with other women. No doubt, too, this
contemporary gendered division of work in weaving has been influenced historically
by Inka and later colonial demands on household labour. For example, the lexical dif-
ference between the term for a mature woman, warmi, a woman who can weave as op-
posed to a woman who cannot weave, called yajiru, may be the result of centuries of
household obligations toward the state during the Inka and colonial periods, whereby
a married woman was required to weave a prescribed number of garments for state use
each year (see Murra 1975a and b, 1989). Similarly, married men as the named heads
of the household, registered in the state census rolls, were required to weave a certain
yardage of coarse homespun cloth in both historical periods.
Nowadays in Qaqachaka, men weave on the upright loom introduced by the
Spanish. They weave homespun cloth made from sheep's wool, in two thicknesses;
this cloth is called wayita, a term Aymarised from the Spanish bayeta. From a thick
yarn, they weave various clothes: men's trousers, men's shirts, also women's dresses
(allmilld), blouses, and skirts. From a finer yarn, they weave men's jackets and
women's mantillas. Fathers still make their children's trousers, skirts, and blouses
from wayita. In the gendered division of labour related to weaving activities, men spin
sheep's wool for the coarse homespun cloth, but they also spin the llama wool
necessary for the food sacks (kustdld), although the sacks are woven by women. Men
spin the strong warp threads for most weavings. They weave belts (wak'a), and in the
recent past they wove the large blankets called apichusi. In addition to weaving, they
twist thick hanks of wool, plaiting it into ropes (wiska) and slings (q 'urawi). Men also
knit the characteristic finely detailed woollen caps of the region (ch 'ulu), on five wire
knitting needles, using modern acrylic wools. It is in their knitted woollen caps, rather
than in their coarse homespun garments, that men practise the most technically
complex of their weaving art, including the technique of 'taking out designs' (apsu
salta).
Until a couple of generations ago, the women of Qaqachaka wove larger garments,
such as the mantle and poncho, on a backstrap loom which they called a 'loom on
poles' (lawat sawuna), usually set up in a sheltered spot in the courtyard of the house.
Nowadays, however, they weave them on a horizontal loom (ch 'akur sawuna) formed
from four stakes set into the earth, and fastened back to other wooden ties with ropes,
reserving a miniature backstrap loom for small items such as belts and fasteners. Older
women comment that the weaving of these larger garments was much finer in the past,
'right bald' (suma q 'ara), since the tension was much easier to adjust by a slight body
movement, rocking backward on your waist. But although there are disadvantages
with the modern horizontal loom, it has more flexibility. For instance, you can roll up
the working weaving and take it with you to weave up in the hills while you are
pasturing the animals. Weaving is usually done in a sheltered spot, often in a walled
corral near to where the animals are grazing.
In the gendered division of labour in weaving, women weave the smaller modern
blankets (p"irsara) from sheep's wool, women's mantles (awayu) and overskirts
(urk"u), and men's ponchos, in a mixture of alpaca and sheep's wool (see figure 7-1).
They also weave belts (wak'a), skirt-fasteners (t'isnu), babies swathing-bands (also
called wak'a), as well as the food sacks woven from llama wool. Women and young
girls weave the decorative hat-bands (irsipilla) for fiestas. And women plait various
forms of ties and fasteners, as well as the edge decorations of their finished weavings.
In a typical year, a woman might weave two blankets, two mantles, and a variety of
other smaller weavings.
104 MULTIPLE MEDIA
Figure 7-1 Marfa Ayka Colque weaves an awayu on a horizontal loom as she sits in her
own field.
GENDERED ASPECTS OF TEXTILE LANGUAGE

Gendered distinctions are also evident in textile language. Recent studies of the
relationship between textile language and speech are beginning to suggest ways of
understanding the linguistic parallels between the formal structure of textile designs
and the syntax and discourse organisation of spoken Andean languages (Arnold and
Yapita 1992; Parssinen 1992). For example, particular spatial configurations of textile
layout may refer us analogically to the different levels of ayllu organisation, as they
are perceived according to gender relations.
In this way, women's mantles may be worn as items of dress, hanging down their
backs, although the precise configuration depends on a woman's age and standing in
any one community. A young woman usually wears the patterned area of her mantle
hung horizontally in the style called k'illp"a, the top two corners fastened at her chest
with a large sewing needle or, alternatively, with a large safety pin. Curiously enough,
this horizontal configuration is explicitly compared to the horizontal web of female
blood ties, both human and animal, which interconnect the more localised patrilines
across space, since women usually move after marriage with their inheritance of
animals to the household of their new husband. This same idea is enacted ritually
through the spilling of blood in the maternal line during the important annual animal
earmarking ceremony called by the same name: "k'illp"a" (see Arnold, 1988: 229ff.).
Or, more commonly for married and older women, the mantle is folded in half, doubled
up, like the married couple, and then draped over their shoulders and fastened at their
chests. If a married woman wears her mantle single, people might ask her: "Do you
want to become a widow"?
One cannot overemphasise the importance of weavings in the Andes as containers.
Mantles are used as carrying cloths, containing anything from raw food products
collected from the fields to cooked lunch snacks, from skeins or balls of wool to
rolled-up weavings in progress, and not least of all small babies. The mantle as a
container is worn horizontally across the back of the shoulders, and the two outer
corners are pinned together at the chest.
Young men sometimes use a small awayu as a carrying cloth, although they wear
them differently from women. The load is worn behind, bundled up behind their waists,
and the two outer corners of the mantle are tied at the front, rather than fastened with a
pin. Young men often wear colourful and decorative mantles this way in fiestas, to show
off their girlfriends' or young wives' weaving skills. If older men should use a mantle
for carrying, once again it is worn in a distinctive fashion, slung diagonally behind one
shoulder and then down around the waist, where the two outer corners are tied.
Men use their ponchos as women use their mantles, as warm overgarments. They
are worn draped over their shoulders with their head through the opening in the central
seam, called its 'mouth' (laka), but with the patterned areas hung vertically. Or they
use them as carrying cloths, to carry anything from extra clothing to heavy burdens of
firewood, over their shoulders with the two outer ends of the poncho tied together at
the chest or waist.2
INTERTEXTS BETWEEN AGRICULTURE AND WEAVING
Weaving as a gendered activity has analogies with other such transformatory activi-
ties, and the play of analogies between them, as "intertexts", continually negotiates
gender status in the different domains. This can be amply illustrated in the multiple
analogies between men's ploughing of the land and women's weaving of textiles,
viewed as gendered forms of work, or in those between women's songs as they weave
and display their textiles, and men's music making as they admire the young women
and their skill as weavers. The Andean equivalence between weaving and food
production is evident historically in pre-Inka and Inka obligations between the
peasantry and the state (Murra 1989: 285-286), and modern analogies between these
tasks still lie at the heart of traditional systems of exchange of gendered gifts between
men and women and in the songlines that accompany them.
In contemporary vernacular discourse, the female activity of weaving as a task
still has many direct analogies with men's work in ploughing and planting the earth.
The motion of weaving the weft threads into and out of the warp threads on her
horizontal loom is considered to be like a man's action with the tip of his ploughshare,
ploughing furrows into the earth. A man's action ploughing the land opens the ground
preparing it for the placing of the food seeds into the earth, while a woman's action in
weaving a mantle opens and closes the working mouth of her loom by the heddle
movements, leaving behind woven seeds embedded in the textile designs. Incidentally,
both activities are, in turn, related to writing. The written lines on the white pages of
106 MULTIPLE MEDIA
my field notebooks, for example, were frequently compared by visiting women to

furrows in the earth, and the movement of my pen on the page to that of a ploughshare
or to a weaving shuttle.
Although these analogies between agriculture and weaving—and writing—are
also present in relation to men's weaving activities, they are much more muted. Men's
weaving of coarse homespun cloth and the simpler motion of the Spanish upright loom
is much simpler technically than the women's weaving of fine cloth, and does not
seem to merit the same layering of analogies and metaphors related to the complex
picking and heddle movements of the horizontal loom. Only in relation to men's
knitting of their woollen caps may we find a weaving discourse that is at all comparable
with the women's weaving discourse described here.
Qaqa women, then, overtly counterpose the language of agriculture with their own
weaving discourse. In the formal layout of the mantle design, they acknowledge that
the pampa, the wide expanses of a single colour, are those that are newly under
cultivation. The patterned areas of the mantle are called in Aymara salta 'to stand up',
a term with double significance: referring to the complexity and movement of the
weaving technique in this zone of the fabric whereby the salta designs "stand up", and
marking the stage of growth of the food crops as they sprout up from the newly worked
earth. The salta—or its Quechua equivalent pallay (gather up)—can thus be compared
to the collection of food products from the harvest piled up at the sides of the zone
under cultivation, the agricultural field called "pampa". These intertexts between
weaving and agricultural activities and the discourse that surrounds them are also
evident in other regions of the altiplano. In the region of Pacajes, for example, the
cultivated fields are described in drinking language as Qipa Mama 'Weft Mother', the
people of the region are compared to the loom, and everything they produce with the
weft (Arnold and Yapita with Apaza, in press).
THE ORIGINS OF, AND INSPIRATION FOR, WEAVING

It should not surprise us either that both weaving and writing share the same spoken
origin myth, which alludes to the transitional interface between the prehistoric dark
ages of Qaqachaka and its more Christian present. The Qaqas themselves generally
agree that the ancient chullpa, the ancestral occupants of the chullpa houses whose
bone remains from mummy bundles can be found throughout the ayllu, did not know
how to weave. The origins and inspiration for weaving does not come, then, from the
ancestral chullpa in their age of darkness. They come, rather, from the dawning of
another age of light, and the oral histories concerning these origins narrate the common
myth about the destruction of the chullpa with the birth of a new sun and the burning
of its resplendent light. Some authors have argued that this is a myth concerning the
Spanish conquest. However, the same myth also appears to be related to an Inka
presence in the area, for the Qaqas acknowledge that weaving—and writing—came
into existence with the birth of the Inka sun god, Inti Tala, and that with these two
skills the Inkas were able to dominate the chullpa peoples.
Nevertheless, the same chullpa mummy bundles, but recast as the god-saints of
the new Christian age, are acknowledged as the source of inspiration for weaving. The
various mamitas such as Mama Tulurisa, Mama Pitunisa, Mama Kantilayra, Mama
Making Men in Her Own [mage 107
Waralupi, and Mama Qupakawana are called 'the weavers', sawuri, and women chew
coca and pray to them when they weave "for a good hand, a good pulse, and a good eye":
Mamita Waralupi Little mother Guadelupe,

amparam churitanta. you will give me your hand.
Sawu Tawaqu umt'anani Let's drink to the Young Girl of the Loom,
Waralup tayksataki for Guadalupe, our mother,
amparamataki, wich'unataki, ma for your hand, for a weaving-pick, for a pulse,
wulsutaki.
Depending on the day on which you weave, you may pray to a dead ancestress who
was a good weaver on a Monday, the day when the dead are remembered; you may not
weave on a Tuesday, which is a devilish day. You may pray to one of the mamitas as
you weave on a Wednesday or a Saturday, or pray to a male god- saint, Killak Tatala or
Panakach Tatala, as you weave on a Thursday, but you may not weave on a Friday,
another devilish day. Sunday is a day of rest. Catherine Allen (1988 and this volume)
also mentions the importance of female saints to women weavers in Sonqo, Peru.
Manifestations of the Inka ancestral spirits make a seasonal appearance when
they become an important inspiration for weaving. During the entire period of the
rainy season, from the Feast of the Dead in November until the following Carnival,
the jira mayku and jira t'alla 'the spiralling lords and ladies', often compared to
Inkas, are present in the ayllu. They are a part of the realm of the mountain spirits
and are believed to come from the distant ocean, the rain clouds, and the inner
world. They are the spirits of the dead, ambiguously a dangerous and deathly force,
but one that may also generate fertility (see also Harris 1982; Arnold 1992b).
Women are inspired by them to weave from the time of their arrival in the ayllu
from the western ocean at the beginning of the rainy season at the Feast of the
Dead to their departure back to the ocean at Carnival with the end of the rains. A
weaving that a woman is working on before each of these feasts, the Immaculate
Conception of the Virgin and Carnival, must definitely be completed by the date
of the feast; otherwise they believe that it will never be finished, the woman will
become lethargic, and the strong earth will grasp her. At these times of the year,
there is the most frantic desire to weave.
Apart from these wet and diabolic sources of inspiration or their more Christian
counterparts, women also acknowledge a metaphysical source for their individual
inspiration, which they say comes from their hearts, chuyma. The term "chuyma"
describes literally the chest organs, including the physical heart (lluqu) lungs (sama
sama), and liver (k'iwch"a), and the mass of blood and breath located there. But in its
more metaphorical sense, it translates more as "conscience", or perhaps even better
as the old Nordic "pluck". The woman agree among themselves that certain women
have special gifts for weaving, and these gifted women weavers they call 'lifted hearts'
alax chuyma, intelligent women who just observe a design once and are able to copy
it down from memory. Most women chew coca as they contemplate their weaving
forms, colours, and designs. At the precise moment that their heart opens like a bud,
then spirit flows and they are inspired to weave, grasping the designs from the blood
mass in their hearts.
108 MULTIPLE MEDIA
Inspiration for weaving also comes from the world that women observe around
them as they weave. Above all, it comes from the natural world of the hills and the
wild animals, birds, and plants that dwell there. Women are inspired most in their
creative tasks of weaving during the most fertile times of the year, especially the rainy
season when the earth is green and flowering. The Aymara verb p"anchayana
describes both the opening of a bud and the opening of the heart, at the moment when
it is lifted and inspired to weave. As Dransart observes (in this volume), this powerful
notion of "flowering" at once interrelates the activity of remembering in the heart as
the female locus of memory and inspiration with the female capacity for the perpetual
generation of lineage, both human and animal. "Heart, make me flower", chuyma
p"aqartatistanta, they say. Then their animals are happy and they are happy. Women
do not like to weave in the dry season, when it is cold. There is no vegetation or food.
The animals walk about disconsolately, and so do the women. The dry dust blows
about, making the new weaving on the loom dirty, whereas with the rains the air is
clean and fresh, and so are the weavings.
Steps in the process of weaving

I shall now consider in stages the different processes of transformation involved in the
conversion of the woollen fleeces of the animal herds, through specific tasks of work,
into woven cloth. At each stage, the precise relations of a gendered division of labour,
not only in the human but also in the animal domain, give rise to a series of gendered
analogies between people and their animals, expressed in the vernacular texts that
accompany their work tasks.
Wool from male and female animals are used for different weaving tasks, associ-
ated with the different genders. Wool from female animals is said to be softer. It is
used for making skeins, and women weave with it for their mantles. Wool from male
animals is said to be coarser, and men use it to weave the coarse homespun cloth. Wool
from both male and female llamas is used for making food sacks and the smaller food
bags. Woven textiles that use dyed wool are generally woven from female animals,
as the wool is softer and easier to manage. The main gendered division of labour, then,
is between women using wool from female animals to weave fine cloth of dyed colours
and men using wool from male animals to weave coarse homespun cloth.
In Qaqachaka, as they shear their animals, weavers and herders also mutter this
direct comparison between the woollen fleece of a given animal and the weavings they
will make from it after the wool has passed through various stages of transformation.
The men murmur:
Punchusim, narak puncht'asi". Take off your poncho, I shall put on a poncho,
too.
while the women murmur:
Awayusim, narak awayt'asT. Take off your mantle, I shall put on a mantle,
too.
The animals, too, are addressed in humanised and gendered terms as they are sheared.
Alpacas are left with special clumps of wool dangling from their coats, which are
called their "coca-cloths" (inkund) if they are females or "coca-bags" (wallqipu) if
they are males:
T'arwa inkunyanaw jisk'ita, The wool is for making a coca-cloth, a little one,
warmipiniw, taqpach warminak she's really a woman, and all women have coca-
inkunani. cloths.
NATURAL HUES AND COLOURED DYES

The wools used for weaving come in the raw natural hues of tan (paqu), grey (uqi),
black (ch 'iyara), white (janq'u), coffee, and chestnut (ch 'umpi), and a variety of
combination in tones between them. The llamas traditionally reared in Qaqachaka,
called ch "uwa, have fleeces of one single colour, and these are the most favoured
animals for the yarns of natural colours used in weaving. It is these natural colours,
particularly black and chestnut, that are used for the broad plain-coloured areas of
everyday garments, the pampa of women's mantles, and the saya of men's ponchos.
The prized 'painted llamas' (pinta qarwd) introduced into the ayllu more recently,
with their broad patches of mixed colours, tend to be used only for the coarse ropes
and slings where mixed coloured yarns may be used.
Or the wools may be dyed. Each colour set has its distinct set of uses. However,
the weavings that are considered to be the most pretty (k'achitu) are those that have
dyed colours, above all the colour scarlet, or blood-red wila,3 used with other colours
that "grasp" each other well in combination. It is a predominantly red colour, for
example, which is used for the salta figurative area of the woman's mantle, contrasting
with the plain and naturally coloured pampa. Ideally, the variety of colours used will
be somewhat like the rainbow, but not too like the rainbow which is dangerous. As
opposed to the contrast between the reddish salta and the plain-coloured pampa in
normal everyday attire, however, festive garments are infused with the colour red all
over. Even the pampa becomes infused with red.
DYEING THE WOOL

Dyeing the wool entails making it cook (q"atiyana). The plied wool, wound into
skeins, is immersed in boiling water containing the dyes. In the past, these were made
from special herbs, roots, and stones, which are still recorded and remembered in the
words of the women's songs. Specialist dyers provided the dyes and dyed the wool.
Nowadays brighter aniline dyes are preferred as they are more colourfast, and the
Aymara term for the modern dyeing technique, tiniha, is a direct borrowing from the
Spanish tenir. There is even a move toward factory-made acrylic wool, whose bright
and clear colours, with their characteristic of longevity, have brought them to the
height of fashion among the younger women of the ayllu.
Qaqa women explain that making the skeins of wool cook over fire "introduces
light" into the wool, making it brighter in colour, thus reminding us of yet other
associated narratives about the civilising function of cooking-fire as well as of the
110 MULTIPLE MEDIA
resplendent sunrise at the very origin of weaving as an activity. However, while blood
red may be the bright and cooked colour par excellence, the Qaqas' definition of tone
and intensity in the dyed colours of woven cloth depends on the amount of white
or black added to the primary colour. Lighter colours, with white added, are
regarded as drier colours, whereas darker colours, with black added, are regarded
as more moist.
WEAVING A MANTLE
Now we focus on the weaving and design of the woman's mantle, called awayu,
examined here as a kind of text. I concentrate on the ethnocategories used by the
weavers themselves (see Gavilan and Ulloa 1992), but in a wider sense. In order to
weave her awayu, a woman first sets up her loom in a sheltered and quiet spot in the
hills, beating the four corner stakes into the earth and speaking her request to the earth
virgin to receive her offering as she does so. She ties the stakes back to other smaller
stakes in the earth with ropes and then begins the process of setting out the warp
threads. With a partner, she throws the warp threads (chinu) to and fro, looping them
over the timber poles of the loom, the double threads of one complete loop forming
one chinu of the cloth.4 She already has in mind the entire layout of her finished cloth,
the threads of which she will select in a process called apsu. She sets out the first half
of her mantle, with its broad divisions into areas of plain or "ordinary" weave (inaki)
in plain dark-coloured natural hues in the centre, the pampa, with the narrow decorated
outside edge, to be used for fastening the mantle at her chest, to one side. The wider
decorated area of cloth, called 'large salta' (jach 'a salta), is set out to the other side,
with its mixture of dyed-coloured stripes in plain weave, its checker patterns (k 'utu),
also in plain weave, with the most detailed patterned areas, called 'small salta' (jisk'a
salta), formed in a complementary warp-faced weave using three warp threads and two
wefts. The three warps form the three layers of cloth in this complex figurative zone (see
figure 7-2).
The small salta may be set out in boxed units, called tika (as in figure 7-2), or in
a continuous design, depending on the time the weaver has at hand. The boxed pattern
is repetitive and easiest to weave, and is used mainly at the higher levels of the ayllu
where women have larger flocks of animals and less time to concentrate on their
weaving. The continuous patterns that require constant attention are found at the lower
levels of the ayllu where the flocks are smaller, and even there only among the best
weavers. The salta designs are selected and organised by the process of counting the
warp threads, called chinu, and each woman knows a certain range of figures that she
may weave in any precise number of chinu (see figure 7-3).
The finished mantle as a text: the pampa and salta

Various authors have commented how Andean cloth, as a kind of text, is a woven
expression of ayllu territory (for example, Platt 1978; Cereceda 1978; Gisbert et al.
1987; Torrico 1989). Many of these studies have been inspired by taxonomic or
semiotic readings of the cloth, rather than by a study of the weaving techniques
involved or of the language that the weavers themselves use about the component
Figure 7-2 The weaving terminology of a woman's typical everyday awayu.
elements of their work. The weaving terms for the distinct zones of the cloth have been
analysed semiotically, sometimes in a structuralist pairing of binary opposites, partic-
ularly with reference to the supposed distinction between categories associated with
"culture" and those associated with "nature".
A definition of the broad areas of one single colour, called "pampa" in the woman's
mantle and "saya" in the man's poncho, has been particularly problematic in weaving
studies to date. It is often concluded that the pampa area is related to the large expanses
of flat uncultivated land, outside the main cultivated zone (see, for example, Cereceda
1978; Bouysse Cassagne 1987: 190-91). It has been observed that the thin coloured
stripes (of just one or two chinus) that border the pampa areas and make up the salta
figurative areas have names related to the boundary zones between cultivated fields,
the stone walls (jalja) that divide communities and agricultural fields, the walls that
surround small cultivated plots of land (uyu and tini), and the streams (jalaqa) that
flow down the mountains, some of which also form boundary zones (Arnold 1988;
Torrico 1989). Usually these distinctions between the two principal zones of the cloth
112 MULTIPLE MEDIA
Figure 7-3 Typical everyday awayu with the salta enlarged. Note the box-like tika figures
of horses and tigers alternating. Note also the striped end detail called patt'ata and
the final figure of a bird.
have been drawn in terms of nature and culture. The pampa is inevitably associated
with the category of nature, the wilder zone outside inhabited and cultivated space,
while the figurative design, the salta, is associated with the cultivated and domesti-
cated space around the immediate hamlet.
In Qaqachaka, these structuralist distinctions do not seem to be appropriate.
Although there is a coding of elements in its designs, the mantle is still essentially
figurative. The wide expanse of a single colour in the mantle, called pampa, is directly
associated by the Qaqas themselves with an area of flat land periodically brought under
cultivation, land where the animals may also graze, on occasion. Pampa as a textile
zone is related then to an agricultural field which may be cultivated. As dona Lucia
Quispe Cheque, one elderly Qaqa weaver, explained:
Yapuchan pamp sataw purapa. Both are called agricultural fields.

Manq'an pampat jaqi manq'i, It's a field where people eat from,
manq'an pampapinij sha, it's really a field for food-crops,
ukat ukan sawupt"a. and it's where we weave.
Sawun ch'utun sawupt"a, jisk'a We weave in the weaving places, which are little
pampanaka. fields.
The raw and the cooked: the mother and her offspring
A further clue to the nature of the textile term pampa is given in its alternative name,
tayka, meaning 'mother'. The entire zone of cloth called pampa may also be contrasted
with the central and decorated part of the cloth, called salta. Salta is the weaving term
for the whole expanse of the patterned and coloured area, as well as for its distinct
"figures". The weaving term salta is probably derived from the Aymara verb saltana,
in turn a borrowing from the Spanish saltar, meaning 'to stand, or jump up'. These
two principal textile zones with their different weaving techniques and uses of colour,
the pampa and the salta, are contrasted by the Qaqa women themselves as 'raw'
(ch 'uqi) and 'cooked' (q "ati) zones. Their own categories echo the technical processes
they have carried out in the stages of weaving. The unpatterned and monochrome
pampa of an everyday mantle, as we have seen, is more usually woven from undyed
wool in natural hues, usually of black or chestnut-brown, while the patterned and
coloured salta is literally "cooked", being woven from the cooked and dyed yarns of
various bright colours.
But the relationship between pampa and salta designs are also phrased by the Qaqa
women weavers themselves in other ways. The salta designs are described as the
'offspring' (qallu) of the pampa, or using the pampa's alternative name of "mother",
as "the young one of the mother". Sometimes the pampa is related to the salta as the
'large mother' (jach'a tayka) to the 'little mother' (Jisk'a tayka). Then the salta
patterned area is further subdivided, so that the "little mother" salta has her many
"offspring", which are the smaller figurative designs within the overall patterned area
of cloth. Weavers, like dona Lucia, say that the "large mother" is broad, whereas the
"little mother" designs, her offspring, are "sprinkled about" inside her.
This formal organisation and taxonomy of textile zones into "mother" and "off-
spring" units has been observed in other textile genres: in Cereceda's study of the small
food bag called "wayaqa" (Cereceda 1978), in Torrico's study of the large food sack
called "kustala" (Torrico 1989), and in Zorn's study of the coca-cloth called
"unkhuna" (cf. "inkuna" earlier in this chapter; Zorn 1986). As to the woman's mantle,
called "awayu", however, its very technical complexity appears to give rise to even
more layers of meaning concerning this relationship between mother and offspring.
There are, for example, some further clues to their possible formulations in the
differential emphasis given to these two textile zones by older and younger women.
Young women, showing off their skill as weavers, and thus their marriageability, make
a point of weaving the largest saltas that current fashions will allow. They have
introduced more and more complex weaving techniques into the salta designs, using
up to five large sets of heddles in the most complex saltas of all, the t 'isnu salta. Their
pampas, on the other hand, are in relation shrunken in size. The scale of the pampa
114 MULTIPLE MEDIA
space withdraws in relation to the salta space. The Aymara verb used for this process,
irkatana, emphasises this notion of contest between the red and the black parts of the
textile; they say that the red part, wilani, "rises against the black". Thus, the saltas of
the young women, as a textile zone, become shamelessly large, so much so that the
sheer scale of the young women's saltas shocks the older women of the ayllu, those
who were brought up on modestly scaled saltas which were not even to be seen until
they were completed.
Older women sometimes suggest that this generational gap in the scale of the salta
designs is due to the modern influence of schooling. They say that now is the 'time of
reading' (liyi timpu) and that younger women, with their reading and writing abilities,
are able to just 'pick out' (p 'itana) designs without the necessity of using the design
models that the older women were accustomed to copy from in their youth. Interest-
ingly, the Aymara verb they use here for weaving, "p'itana", has the more specific
meaning of 'picking out' the design with a pointed object, like a weaving pick, or a
plumed pen, or a modern pencil. It is also the same verb used when men knit their
woollen hats with pointed knitting needles, or when hens pick out seeds from the
ground with their pointed beaks. Are the literate young women "writing" their designs
into their woven texts? (See figure 7-4).
There are other functional reasons for this difference in scale between pampa and
salta. Older women, in contrast to the younger women, have less time for weaving.
With their failing eyesight, they weave much less elaborate saltas and much broader
pampas. The "large mother" weaves less salta design-space than her "little mother"
daughter. The only exception to this general rule is dona Maria Ayka Llanque, an older
woman in her seventies, whom we affectionately call "Punk" on account of her youth-
Figure 7-4 Dona Juana picking out her salta designs.

fulness and her propensity to wear numerous safety pins here and there on her mantle.
Dona Maria is generally acknowledged in Qaqachaka to be one of the best and most
prolific weavers, in spite of her age. She regards herself as an "eternal maiden", saying
that she weaves "just how she likes"! She is an exception that proves the rule.
Learning the salta designs

Are there other reasons for this difference in age and modesty in relation to the salta
designs of most other women of the ayllu? Further clues to this difference in approach
between generations emerge when we study how women learn the salta designs.
Qaqachaka as a whole has its own characteristic style in the organisation of its
textile zones and use of colour, above all in the proportion between its pampa and
salta and the number of patterned bands, the 'little saltas', which the large salta
incoiporates. These factors make Qaqa cloth recognisable to insiders and outsiders
alike in the region, marking Qaqachaka ethnic identity and history, recording and
reproducing it in cloth. Nevertheless, each individual weaver has her own distinct style
in weaving, her own idiosyncratic discourse about it, and she is considered to be the
personal owner of her textile designs. In relation to the complex and patterned salta
designs, a woman says that her inspiration for them comes from her 'design heart'
salta chuyma), but more generally her love for the saltas. Women explain that their
designs are already "just inside their hearts", allowing them to select whichever salta
they wish:
Chuyma manq"sanki, They are just inside our hearts,

saltas kuna palltanasa. letting us select whatever salta design.
They say that they just contemplate the designs and then "draw them out".
In the process of weaving a mantle, the finished cloth is usually formed from two
identical but separate halves. Each of these sections is called mak"allu, ma meaning
'one', and k"allu meaning 'half-cooked'. In Qaqachaka, there is a clearly aspirated /kh/
in the naming of this component of the finished textile, and it cannot be confused with
the term qallu 'offspring' for the same component that Zorn (1986) suggests is used
in the region of Macusani in Peru. The weaving term qallu is reserved for other
component parts of the mantle in Qaqachaka, as we have already seen. Furthermore,
the term k"allu 'half-cooked' evokes another stage in the transformatory process of
converting raw wool to a cooked and finished cloth. Finally, the two component halves
of the cloth are sewn together. The generative central seam is called chuyma ch 'uku
'heart-seam', since this central seam is regarded as the "heart" of the animate mantle.
The pathways of the salta designs

Then, once the salta design of a weaving is in progress, a woman must follow its unique
pathway (t"ak"i). She tries not to get lost or to move from the salta pathway, for then
she will make a mistake. In following the pathways of her designs, she is also
analogously following the pathways of her territory, for the textile paths are compared
to "the veins of the earth". Thus, she follows the pathways of her animals, and in this
116 MULTIPLE MEDIA
way she incorporates the pathways she treads with her animal flocks in her daily task
pasturing the animals, moving from the mam pueblo or her hamlet to the hillside
houses and their surrounding pastures, from corral to corral, into the textile designs.
The seeds that fall accidentally with the animal droppings onto the daily pathways
form another design feature, which a weaver integrates into her cloth. Other pathways
of growth woven into the cloth are those of the wild plants and shrubs that she is
familiar with as she pastures her animals in the hills, the patterns of which she has
copied in her designs.
The most common salta designs are inspired by the wild beasts of the hillsides,
the condors and falcons, partridges and quails, vizcachas and foxes, lizards, toads, and
snakes, as well as the water birds that frequent the lakes and watering places in the
upper levels of the aySlu. These wild beasts and plants are described by the women as
the beasts, the shrubs, and plants of the virgin earth. Each salta is regarded as animate.
It has a beginning, an end, and a centre, its "navel".
A woman's capacity to weave thus expresses her transformative skills; weaving
activity is regarded as vita! and magical, in weaving the pathways of her salta designs,
of her animal herds and her cultivated fields, a woman not only copies them from the
memories held in her heart but also manipulates them, working on these miniature
elements in microcosm to transform them and bring their larger counterparts into being
in the ayllu macrocosm. Through weaving, she maps and records ayllu history in a
more figurative way, and by embedding the seeds of future production in her designs
she provides the reproductive matrix that guarantees this future production (Arnold
1988; Arnold et al. 1991; Allen, this volume).
Analogies betweenpampa, salta, and the earth

The relationship between the two principal zones of the mantle, between the plain
pampa and the patterned salta, also follows the yearly sequence of the agricultural
cycle. The women weavers say that the pampa is the earth resting. They call her the
'black mother' ch 'iyar tayka, "She has not been worked", they say, meaning she has
not been ploughed. The pampa, like the earth, is black and unturned, she is still lazy
and unwilling to procreate. She; is strong, and her blackness indicates that she will
bring forth, but not yet. Qaqa women, like dona Asunta Arias, another weaver,
emphasise yet again the relationship between cloth and production when they observe
how older women wore black dresses and overskirts, when the earth produced well,
"Whereas nowadays", she says, "women wear just any colour and production in the
ayllu is declining".
The salta, by contrast, has been churned up well: ch 'ata. They say that it is as if she
has been amply ploughed and then planted. "Ploughing thus, like the fields that we
plough, thus it is then, thus it could be, the salta designs", explains dona Asunta. Ch 'ata, the
churned-up area, is another term for the plough handle that does the turning.
The salta, with its bright dyed colours, also has to do with the coming of the rains,
and the greening of the vegetation. The rains, like the devil spirits which bring them,
are believed to come from the western ocean. When they are late, people complain,
using the verb saltana:
Salt'amay carajo, Get up then, damn you,

purxatpanay carajo. oh, that it would rain over the earth, damn it.5
Weaving terms for the narrow stripes that border the little saltas echo this preoccupa-
tion with water. They may be called taniqa 'to flow down'. Or they may be called 'ir-
rigation canals' (larqa), and the little salta designs that they supply "are like the fields".
Sometimes the entire mother salta is called 'large river' (jack 'a jawira) and the small
saltas are called 'little rivers' (jisk 'ajawira). It should not surprise us, then, that the wild
birds and beasts which animate the salta domain are most specifically those "that announce
the rains": the toad or birds flying in groups, the animals of the virgin earth.
However, the precise nature of the difference between the plain pampa and the col-
oured and figurative salta zones of the woven cloth becomes most explicit in relation
to an alternative term for the salta. This key term emerged one day in a casual conver-
sation about weaving with dona Juana Ayka Colque. She called it wilpa, incorporating
the root-term wila 'blood' and the third-person possessive suffix -pa-, a term we might
gloss as 'her redness', or 'her blood'. This "redness" is most evident in the edge zone of
the salta, where it borders the pampa, although, as we have seen, it may flood the entire
salta and even overflow into the pampa zone in a festive mantle. Women say that the
salta is predominantly red in colour "because the earth has her blood and is fertile". The
rains have run into the earth and made the soil turn red. Dona Asunta explained:
Wilaw sas It is blood, saying,
jumintukichix uraqix, ch'ata laq'a, the earth is damp, the trampled earth,
uka wilpaw sapx nanakax .. . this is her blood we say ...
"Thus she produces her fruits, her babies, the wawas, the food products", she added,
"the blood of the earth is her fruit".
Is it for this reason, then, that young women exaggerate the scale of their salta into
outrageous sizes, and shrink back their pampas? In doing so, are the young women
exaggerating their fertility, their blood flow, their womanliness, and their marriage-
ability? Are the older women content to shrink back their saltas to a more modest scale,
to acknowledge the passing of their blood flow and fertility, and their return toward
an ample pampa and a time of rest?
Other textual connections can be made here. It could be said, for example, that it
is precisely a woman's blood flow at her first menstruation that gives her the
inspiration for developing the most complex techniques of her weaving career in the
beautiful large salta designs, those that will become a visual expression of her new
fertility and her potential reproductive powers. This blood flow is the first signal of
her potential capacity to form a new human lineage with her future husband, as well
as relating her to the equivalent stage in an animal's reproductive life marked
analogically by blood at the important ear-cutting ceremony called "k'illp"a", and to
the local birth of the Inka sun in the region. Then, in the following reproductive years
of her life, the rhythm of her cyclical blood flow and her increasing powers as a weaver
are continually nourished by the mass of blood in her heart, providing a continued
source of inspiration for her textiles throughout the middle years of her life with their
large and impressive designs. Only in her final years, when her blood flow has ceased,
is she more modest again in her weaving designs.
118 MULTIPLE MEDIA
In a similar vein, this key symbol of blood flow is also evident in relation to men's
creative powers, but in a quite different manner. While a young man uses the powers of
spirit that reside in his head, in the past, at least, certain "literate" men used the matrix
of sacrificial blood from the animal herds to provide the blood ink that expressed their
developing powers of writing on the page. And it is this same sacrificial blood that a man
still offers in a ritual context to the Inka sun god in a more exclusively male domain.
The pampa and the salta can be compared in other ways. The pampa in everyday
garments, whether in a woman's mantle or a man's poncho, is usually of a dark hue
and natural colour. It is without fire and raw (ch'uqi). It is unproductive like the dry
season. The salta, by contrast, is cooked (q"ati), with firelight glowing from within,
making its colours clear and bright. It is fertile, irrigated, and damp, like the earth in
the rainy season. The two zones of cloth could even perhaps be opposed as shadow to
light, as night and day, or as dark rnoon to full moon; the opposing poles of light pulsate
the cyclical rhythm that brings into being female blood flow. But it is important to
point out that Andean weavers do not, like structuralists, experience a binary reality
of either/or. Even within the most complex of the salta designs themselves, the t'isnu
salta, the purposeful juxtaposition of dark and light colours in the three layers of cloth
i s manipulated in order to contrast the two sides of the cloth in a complementary weave,
like night and day. Both co-exist, as a weaver looking at the upper face of the cloth
understands the other below.
The poncho as a text

We continue here to examine the differences in meaning represented by the unpat-
terned pampa and the patterned salta areas of cloth, but in a comparative study of the
formal organisation of these textile zones in two gendered garments, in the man's
poncho, on the one hand, and the woman's awayu, on the other. In this way, I shall
attempt to offer some final observations about gender and its ideological manipulation
through weaving in Qaqachaka. First, though, I shall outline the formal design of the
man's poncho.
Like women's mantles, men's ponchos are woven in two identical halves, joined
at the centre by a sewn seam. However, there is a difference between the poncho and
the mantle in the zones of cloth that are joined by this central seam. In the poncho, it
is the centra! unpatterned zone of the textile that incorporates this central seam,
whereas in the mantle the central seam joins the two halves of the patterned salta. The
broad design of each half of a typical everyday poncho in Qaqachaka is generally of
two wide bands of patterned areas made up from coloured stripes of varying widths,
called lista, located between three wide bands of unpatterned textile, called saya,
woven in the same natural colour (see Figure 7-5).
The naming of parts of the poncho

The weaving term "lista" recurs at several different levels in the formal organisation
of the patterned areas of cloth of the man's poncho. One entire striped band of the
poncho, comparable to the salta of the woman's mantle, is called "lista". This wide
coloured band, which we might call the "maximal" lista of the poncho, is then further
subdivided into a series of smaller vertical stripes, also called "lista". These smaller
stripes are organised into identical formal groupings, once again called "lista", the
"minimal" lista groupings of the poncho. A typically everyday poncho usually has up
to five minimal lista groupings in each maximal lista band, whereas a festive poncho
may have up to ten. (see Figures 7-5 and 7-6).
Each minimal lista group is usually composed of four coloured stripes, also
called lista, commonly of darker and lighter hues of the same tone, or, alternatively,
of two complementary colours. In the poncho, the four coloured stripes of the two
outermost minimal lista bands, those that border the unpatterned saya on each side
of each maximal lista, are usually red in colour. This is like the same reddened
zone of the salta in the mantle as it borders the pampa. Each of the four coloured
stripes of the minimal lista, usually of between three to four warp threads, has
other alternative terms. They may be called 'group' (qutu) or pampa. In the minimal
lista group, the four wider stripes of colour are sub-divided into two pairs by even
narrower stripes of just one warp thread, the narrowest lista of the entire poncho
(see Figures 7-7 and 7-8).
The weaving terminology of these narrowest stripes depends on their precise
position in relation to the minimal lista grouping as a whole. Each minimal lista group
is symmetrical around a central narrow stripe, flanked on each side by a pair of wider
stripes, which are then, in turn, bordered by two outer narrow stripes. These inner and
outer narrow stripes have distinct names. The outermost narrow stripes are called
taniqa, meaning 'to run down', the directional suffix -qa- denoting the direction
Figure 7-5 The weaving terminology of a man's festive poncho with a red saya.
120 MULTIPLE MEDIA
Figure 7-6 A man's poncho in use.
downward. An alternative name for them is larqa, a term that was also used for the
narrow stripes bordering the smallest salta design of the woman's mantle and having
the meaning in both garments of 'irrigation canal'.
The narrow stripes that form the central 'heart' (chuyma) of the minimal lista
grouping, on the other hand, are most commonly called jalsu, a weaving term
derived from the Aymara verb jalsuna meaning 'to go or to run outward', the
directional suffix -su- marking its direction of flow outward. Weavers compare
the jalsu to a spring of water bubbling up from the earth. An alternative term for
the inner stripes is tansu 'to run outward' using the same directional suffix. And
sometimes these inner stripes may be called 'eyes' (layra).
Besides these differences in weaving terminology, there is also a difference in the
use of colour between these inner and outer sets of narrow stripes. The taniqa
outermost stripes may be black, but they are more usually white. They are not dyed,
but natural hues of wool, whereas the jalsu inner stripes should ideally be dyed red in
Figure 7-7 A maximal lista group in the poncho.
colour. As a rule, even if not all the jalsu stripes in a maximal lista band are red, at
least some should be red. Table 7-1 summarises these differences.
Differences between the woman's mantle and the man's poncho

There are many pertinent differences in the weaving techniques and weaving termi-
nology of the woman's mantle and the man's poncho. I shall examine some further
differences between them, and then attempt to relate these differences to aspects of
social and gender organisation in Qaqachaka as a whole. Finally, I shall suggest how
the differential gender relations of Qaqachaka society are manipulated by women
through a means of cultural or ideological production of cloth.
Table 7-1 Differences in terminology of the narrowest listas.

Outer narrow stripe Inner narrow stripe
taniqa jalsu
-qa- 'to go downward' -su- 'to go outward'
'irrigation canals' 'eyes' or 'heart'
black or white red
natural hues dyed colour
122 MULTIPLE MEDIA
Figure 7-8 A minimal lista group in the poncho.
A primary difference between the patterned zone, or salta, of the woman's mantle
and the saya of the man's poncho is that the mantle's salta is essentially figurative,
whereas the man's poncho comprises a much simpler pattern of stripes, the lista. A
further gendered difference that is immediately noticeable to any observer is in the
way in which the respective patterned section is worn by each sex. Young women
wear their mantles with the patterned salta hung horizontally, in the style called
"k'illp"a", whether the garment is worn hung down the back or doubled up and hung over
the shoulders, whereas men usually wear their ponchos with its lista stripes hung
vertically.6
There is also an immediate difference in the weaving techniques of the patterned
and unpatterned zones in the woman's mantle and the man's poncho. In the woman's
mantle, the unpatterned pampa is woven in a plain weave of a single natural colour,
which contrasts with the more elaborate complementary warp-faced weave of its
patterned salta, with its variety of figures in natural and dyed colours, whether in the
most complex multi-layered warps of the t'isnu salta or, alternatively, the simpler apsu
salta. In the woman's mantle, then, the flat pampa and the "standing up" salta are
differentiated with regard to both weaving technique, non-figurative and figurative
designs, and the use of colour. In the man's poncho, by contrast, both zones are woven
in a flat plain weave (llanu). The difference between the patterned and unpatterned
zones in the man's poncho is not marked, as in the woman's mantle, by a difference
in weaving technique, but by the use of colour and pattern. The unpatterned saya of
Making Men in Her Own image 123
an everyday poncho is woven in natural hues of a single colour, whereas the patterned
lista is woven mainly with dyed colours in a striped pattern.
Further differences become apparent in the weaving terminology of each garment.
Although there are some common weaving terms for the parts of each garment,
particularly in relation to the two main textile zones, there are also distinct differences.
As in the woman's mantle, the unpatterned zones of cloth in the man's poncho can be
called pampa, meaning 'flat place', whereas the patterned and coloured areas of cloth
can be called salta, meaning 'standing up'. The pampas of the man's poncho can even
be called 'mothers' (taykd), as in the women's mantles. However, there is more usually
a distinct gendered difference in the weaving terminology of the two primary zones
of cloth in each garment. The unpatterned zones of the man's poncho are most
commonly called by their weavers "sayas", another weaving term with the inter
changeable meaning of 'standing up'.7 And instead of using the term salta, reserved
more for the woman's mantle, the patterned zones of cloth of the man's poncho are
more usually called lista, the weaving term meaning 'stripe'. (I shall return to other
important dimensions of the significance of the weaving term "lista" later on). The
terms pampa and salta, then, are reserved more for use in relation to the woman's
mantle, whereas the terms say a and lista are used in relation to the man's poncho. Like
the differences in orientation of the patterned area of each gendered garment as they
are generally worn, mentioned earlier, these differences in weaving terminology
similarly define a dominant orientation of the unpatterned areas of cloth according to
gender. The female pampa is a "flat" and horizontal place, whereas the male saya is
"standing up" vertically. These weaving terms and gendered modes of dress thus
reinforce the common notions that women as a gender are associated with the flat
places, the pampa or t 'alla, and men with their respective higher mountain peaks. Table
7-2 summarises these differences.
Mantle pampa and salta, poncho saya and lista compared

With these similarities and differences in weaving terminology, technique, design, and
styles of dress of the two gendered garments in mind, we might venture a comparative
study of their respective parts. Despite their superficial dissimilarity, it seems that,
certain elements of the woman's mantle and the man's poncho deal with similar
semantic domains, and a comparison between them helps to clarify the function of the
puzzling range of stripes, called lista, in the poncho. For example, the variety of figures
in the coloured and patterned salta of the woman's mantle appears to be in many ways
a female equivalent of the repetitive sequence of minimal lista stripes in the man's
Table 7-2 Different terminology and use of the mantle and poncho.
patterned zone patterned zone
WOMAN'S MANTLE pampa salta
'flat place' worn horizontally
MAN'S PONCHO saya lista

'standing up' worn vertically
124 MULTIPLE MEDIA
poncho. Similarly, the monochrome pampa of the woman's mantle appears to be a

female equivalent of the unpatterned zones of the man's poncho, whether they are
called in weaving terminology pampa, tayka, or saya. Thus, the mantle's salta and the
poncho's lista, on the one hand, and the mantle's pampa and the poncho's saya, on the
other, appear to treat essentially the same semantic domains. Perhaps we can be
more precise, then, about a comparable meaning of the lista stripes in the man's
poncho?
If the maximal lista group of the poncho is like the patterned salta of the mantle,
then does the minimal lista grouping of the man's poncho express the same kind of
significance as the minimal small salta of the woman's mantle? There is distinct
evidence for this. For example, we can note the same preoccupation in the weaving
terminology of the minimal lista grouping with the idea of flowing liquids, in its use
of such terms as jalsuna 'to flow out' and taniqana 'to flow down', as there was in the
figurative salta of the woman's mantle, with its "irrigation canals" and "streams that
flow down" the hillside. An alternative term for the minimal lista grouping is, once
again, taniqa 'to flow downward', and weavers say that these minimal striped bands
"flow down from the plain saya" of the poncho, like the salta in the woman's mantle
"comes down from its mother field". Even the basic weaving term salta, in the
woman's mantle, as we have seen, referred in the wedding songs to a rnoist and
female-gendered parcel of land, a salta q"uchi. Within the microcosm of the minimal
lista grouping of the man's poncho there is also the same organisational logic as there
is in the salta of the mantle. There are the similarly wider stripes of colour, which may
be called pampa, divided at the centre by the narrow, usually red, stripe called "jalsu",
just as the broad plain-weave parnpa of the woman's mantle is divided at the centre
by the predominantly red salta (see again Table 7-2).
The relationship between pampa and salta in the woman's mantle, and their
respective gendered equivalents, the saya and lista in the man's poncho, also have to
do more explicitly with the gendered social formations in Qaqachaka. The various
hamlets, formed around patri-, viri-local households are conjoined across space by a
network of female ties. Male ties of semen are passed down vertically through the seed
or semen lines of the male lineages called chacha kasta, most often localised in space
through generations, whereas the female ties of blood are passed down through the
bloodlines called wila kasta, connecting the patrilocal groups horizontally across
space "like a web" in each generation.
In each gendered garment, too, it is as if the patterned and coloured salta of the
mantle or the striped lista of the poncho has to do predominantly with the respective
generative powers of women and men. Thus, the difference in orientation of the salta
or lista in each respective garment as it is worn is analogous to these primary gendered
social formations in the ayllu. The man's poncho hangs with its coloured lista stripes
vertically, like the vertically organised male kindreds, whereas the young woman's
mantle hangs with its figurative and predominantly red salta horizontally, in the
k'illp"a style, like the female (both human and animal) web of blood ties connecting
the patrilineal groups horizontally across space.8 The weaving terminology of pampa
'flat-place' for the unpatterned monochrome zone of the woman's mantle and saya
'standing up' for the equivalent zone of the man's poncho further reinforces these
proposed analogies. The only exceptions to this general rule are those that seem at the
same time to prove it—namely, older women who may wear their mantles with the
salta hung vertically (in the style called pulu) like the man's poncho stripes. They
themselves say that this style of dress "keeps them warm", but they also insist that it
confirms their status as older women "who have lived in the same place for many
years". "Like men", they say, they are "standing vertically in one place".
We must remember, however, that it is a woman weaver in Qaqachaka who weaves
both of these garments, her mantle for herself (or her daughter) and a poncho for her
husband (or her son). In doing so, it is as if she skews both of these gendered social
formations of female bloodline and male semen line toward the female side. It is as
if, in her mantle, a woman weaver displays two different but associated aspects of
autogenesis. On the one hand, she displays the power of her human bloodline of female
ties as revealed in her predominantly blood-red "large salta", and, on the other, the
generative capacity of her household and ayllu, the wild birds and beasts of the hills,
the domesticated animals reared there, and the food products that the ayllu may grow
within its boundaries, as revealed in the "little salta" designs. In the festive mantles,
these female blood ties have even greater predominance. In this case, even the
monochrome pampa of the woman's mantle is flooded with a dyed-colour red, as if
to display during the fiesta the moments of greatest creativity of a woman, her
household and ayllu, mediated through redness. It is as if the passage of the fiesta
condenses the most fertile moments of the year into this single hue.
Even in a man's poncho, a woman weaver displays the same preoccupation with
marking a reddened bloodline in the central space of the minimal lista grouping, the
heart-place where it springs forth like water. It is as if she wishes to emphasise that a
man's own blood relationship to his mother and maternal kin is the primary one. Then,
in the narrow outer stripes of the minimal lista grouping, those which are most often
white, and "semen-like", according to dona Asunta, a woman weaver marks what
appear to be the connections of male ties of semen, the semen lines or chacha kasta,
between social groups. But then, in the festive ponchos, with their dyed-red say a, these
male ties of semen are once more mediated through female ties, ties of blood. What a
woman has done in both examples, then, is to place female ties of blood at the centre
of the weaving design and to relegate male ties of semen to the periphery. And in doing
so, she has re-organised and re-defined the gender relations and social formations of
Qaqachaka for both women and men.
The meaning of the stripes called lista

I shall conclude with a final hypothesis about a difference in signification between the
man's poncho and the woman's mantle as gendered garments. The hypothesis has to
do with further observations on gender relations in Qaqachaka and the way that they
are represented in the textile zones, form, and colour of these two woven items. A vital
clue to this difference lies in the use of the weaving term "lista", a term that we have
noted is used most often in relation to the coloured striping pattern of men's ponchos,
I argue here that although the term "lista" has a range of meanings associated with
agricultural and herding activities, these meanings in a woman weaver's hands are
deliberately muted in the man's poncho, as compared to their raised significance in
126 MULTIPLE MEDIA
their own mantles. Women are thus able to emphasise through the manipulation of
their weavings, as texts, the relation of men with more productive tasks and themselves
with more reproductive and regenerative powers.
The use of the weaving term "lista" in the poncho of Qaqachaka is puzzling.
"Lista" as a term can also be used in the women's mantles, but its use is restricted to
the narrow bands of colour in plain weave that decorate the outer fastening edge of
the cloth, as well as the narrow bands of colour that decorate the large salta area of
the cloth, those which generally border the small saltas. As a weaving term, "lista" is
rather underplayed in the terminology of the mantle. It is in the man's poncho that lista
are predominant. Why should this be?
"Lista" appears to be derived from the Spanish term for a stripe, fringe, or narrow
band. When we study its etymology, the introduction of the Spanish term "lista" into
Aymara weaving terminology is first noted in Bertonio's Spanish-Aymara vocabulary
of 1612, collected in the region of Juli on the southern edge of Lake Titicaca. Bertonio
mentions several kinds of stripe called "lista", and he notes what appear to be a number
of contemporary Aymara terms for these stripes still in use in the region in the
seventeenth century, such as hattu, koli, and caruma (Bertonio 1984 [1612], 1:293;
11:37, 56). These archaic Aymara weaving terms are not known in Qaqachaka and do
not seem to be used in Bertonio's region today, either. Bertonio does, however,
mention an equivalence between the Spanish verb listarand an Aymara verb still used
by weavers around the southern edge of Titicaca, sukochatha, and he implies that the
adjective 'striped' may be translated into Aymara as suko suko. Suk'u suk'u, but with
a glottalised /k'/, is still a weaving term in the lakeside region to the present day for a
narrow stripe in a man's poncho. Our puzzle is why, in Qaqachaka, the Aymara term
for a poncho stripe of colour has disappeared and the Spanish term, "lista", has been
introduced to the exclusion of any other?
Bertonio also relates as synonymous the Spanish verbs listarand baretear 'to open
trenches with a bar or pick' with the Aymara verb sukochatha, whose modern meaning
is 'to open furrows with a plough', thus drawing the meaning of the weaving term
"listar" into the realm of agricultural production (Bertonio 1984 [1612J, 1:293). The
concept of a ploughed furrow may have been foreign at first to an Andean population
used to using the hand plough and only turning over clods of earth in order to place
the food-seeds inside, and the Spanish term for furrow, surco, is borrowed directly
into Aymara as suku. The close equivalence between the terms for furrows, stripes,
and the action of breaking open trenches does nevertheless imply an analogy between
the narrow lista stripes in a woven poncho and each furrow in the turned field of the
modern salta design. As long as anyone can remember, the Spanish foot plough has
been used in Qaqachaka to plough the land into furrows. Only the ayllu elders remark
how the sharp point of the Spanish ploughshare injures the virgin earth more than the
older indigenous foot plough, and they request her forgiveness before the ploughing
begins.
There are associations of the term "lista" with herding activities as well as with
agricultural ones. According to Cassandra Torrico, the term "lista" among the Macha
of northern Potosi, neighbours of the Qaqas, is used for the coloured woollen earrings
placed in the notched ears of the animals during the marking ceremony or on their
saint's day (Torrico 1989). "Lista", in this context in Macha, is thus comparable with
the Quechua terms t'ika or chimpu elsewhere, or p"aqara 'flower' in the Aymara
terminology of Qaqachaka. When the animals are marked in Qaqachaka, they are
regarded as being "flowered''. In this notion of "flowering", there are further sugges-
tions of maternal ties. In the marking ceremony of the animals, the woollen earring is
placed in the notched ear of an animal, and the marking by notching, as we have shown,
draws forth a flow of blood marking a continuous bloodline of animals, since the
marking passes from mother animal to offspring (Arnold 1988: 229ff.). Torrico also
notes the association of certain colours of wool for the lista earrings in Madia with
distinct family lines of animals.
The more common definition of "lista", however, that which survives into modern
Spanish, is as a 'list', a roll call or register, such as a bill of fare, a college roll call,
and so on. It is this modern definition that I shall pursue here. The poncho designs of
Qaqachaka have changed significantly in the past fifty years, and it is possible that
these modern changes might have been decisive in the implementation of the use of
the weaving term "lista" for the poncho stripes, to the exclusion of other garments.
There have been particularly dramatic changes in poncho design since the Bolivian
Revolution and the Agrarian Reform of 1952-53, design changes that have been
accompanied by major social and cultural changes, such as the introduction of
mechanised transport, public education, and greater access to the external market
system. Could it be that the stripes of the poncho may have taken over a vestigial
function, one that was no longer necessary after this period of major change? Could
this function have been in a listing of items which are no longer displayed or counted
in this manner?
The poncho stripes as a mnemonic device

Silverman-Proust, in her study of the striped wayaqa food bag in Q'ero in Peru (1988),
suggests that the lista stripes of the food bag, with their various different widths and
colours, were homologous in many ways with the older kipu system of counting. She
gives evidence that the broader stripes of the food bag signified larger quantities,
previously marked by the wide knots of the kipu cords, while the narrower stripes of
the food bag, like the smaller knots of the kipu, marked smaller quantities. She also
elaborates a possible coded system of colours for the products represented in the
stripes, as the various varieties of maize and potatoes produced in the ayllu of Q'ero,
as well as the different colours of llama fleece reared there. She even notes the use of
certain colours in the coded system of the food bag as associated with family lines.
She confirms, too, that the information recorded nowadays on the food bags was
previously recorded on the kipu cords within the living memory of the older residents
of Q'ero. This information, concerning ayllu production, was presented by the cord-
keepers of Q'ero to the outside authorities who visited them on occasion.
Might I tentatively suggest that the poncho stripes, called "lista" in Qaqachaka,
may also have expressed similar domains of knowledge to those annotated on the
historical kipu orchinu cords? This would presumably have occurred during the period
of cultural change, from an older mnemonic system to the introduction of modern
writing in the programmes of enforced literacy in the newly built schools of
Qaqachaka. There are clues in Qaqachaka, as there are in Macha, that family lines are
128 MULTIPLE MEDIA
marked by certain colours in the stripes, as they were once in the various kipu cords.
There are also clues that certain products of the ayllu may have been colour-coded
into the stripes of the poncho, in the same way that they are always recited in a certain
sequence in the food libations, as modern intertexts of the older kipu. Dona Lucia
confirmed to us how:
List ukat, aka maq'anaspa, This stripe could be the food products,
aka ch'uqi'spa, this one could be potatoes,
u k a t akaru t u n q i i s p , tirijusp, then this one could be maize, wheat, beans, then
jawasasp, ukat yaranusp, barley,
uk"amasjipi, thus it is, then,
surtinans uk"ampini. even in the libations it is so.
As yet, however, we have not found such a clearcut relationship between the lista
stripes of the man's poncho in Qaqachaka and the older system of counting by kipu
that Silverman-Proust has encountered in Q'ero in Peru.9 It is not common practice in
Qaqachaka, either, to extend the chinu threads of a poncho into a fringed border, like
those of neighbouring regions, to suggest even more pertinently, visually at least, the
appearance of the coloured pendulant cords of a kipu device. Such fringes do still
figure, however, on the special ponchos used by the ayllu authorities of Qaqachaka
when they take office, those who might have once presented such coded information
to the external authorities.
The poncho and mantle compared as gendered garments

At a more general level, however, it seems likely that the preponderance of lista stripes
in the man's poncho and their underplaying in the woman's mantle is related to
important gendered differences to do with their respective functions: on the one hand,
of representing a coded system of information on ayllu production for outer consump-
tion, and, on the other, of representing the inner resources of the ayllu as a more
figurative form of beauty and aesthetics. Might we tease out this distinction even
further, to suggest that an essential gendered difference between the two garments
might lie in a differential emphasis in the man's poncho on "production" and in the
woman's mantle on "reproduction"? Qaqas themselves do suggest such a distinction
when they differentiate between men who, like silver, enter into circulation and
participate in the productive cycles of external exchange, whereas, women, like gold,
embody the wealth that grows within, in the bowels of the earth of their ayllu. Dona
Lucia also points out that since the poncho is "flat" it can only represent food as the
result of men's agricultural work. Whereas, she says, the woman's mantle with its
"standing up" salta design represents in more general terms "all of the blessings that
the earth provides, in animals and food products, and even in birds".
The men's ponchos with their lista stripes do seem to mediate a more abstract and
codified system of representing the elements of ayllu production than the women's
mantles, in both agricultural and herding domains. This information is in turn dis-
played by men, who since colonial times have been the named heads of households,
the named mediators between the inner ayllu and the outer world of the external market
system. Historical evidence suggests that male kipu-keepers, the kipukamayuq of the
region, once presented their knotted cords, with their detailed inventory of ayllu
production, to an external system of authority. Perhaps, in a similar way, modern Qaqa
men might still be displaying in their poncho garments "lists" of items for both internal
and external consumption. That being so, a man also readily acknowledges that his
productive work in farming agricultural produce and earning money from the market-
place is, finally and inevitably, given up to his wife, to deposit in the household stores,
where she will control it.
Besides their powers of production, men's powers of reproduction and auto-
genesis are also represented in the ponchos that they wear. It seems possible, for
example, that more personalised patrilineal family lines, the chacha kasta, are
revealed by their colour-coding as owning certain sets of ayllu produce. In the
colour-coding of the stripes, too, there seems to be embedded information about
the seminal nature of male reproductive and generative powers. But I would argue
that representations of these overtly male powers are underplayed in the poncho
as compared to the colourful display of women's reproductive and generative
powers in the excessively red and blood-like salta designs of the mantle. The
modern association of the stripes in a man's poncho with the historical kipu cords
does remind us of the historical association of men with the writing of such texts,
but it is a woman weaver who is defining this association, and she does not give
it a high status. Overall, I would argue that, in her woven poncho, a woman weaver
is defining men in her own image.
There are other important reasons why a woman weaver may underplay male
reproductive powers in the poncho as a garment. According to a gendered system of
metaphysics, women are believed to think "with their hearts". As the female locus of
thought, intelligence, and memory, the heart is believed to be the source of their songs
and their weaving designs. The heart is viewed as a mass of concentrated blood and
breath and is a locus of transformation. It is the essence of their femaleness. And it is
the mantle as a garment of dress that wraps that part of their bodies which encloses
their hearts. By contrast, a man is believed to think "with his head". As each individual
woman's first blood flow initiated her greater creative powers of weaving, so her
continuing blood flow, like her later salta designs, marks the memory of the birth of
the new Inka sun god in the region, the memories of which are held in her heart.
Similarly, a man's equivalent powers are believed to reside in the spirit in his head,
only mediated through the sacrifical blood ink with which he once wrote, in memory
of the same Inka sun god. His poncho as a garment of dress wraps only his body. It is,
in fact, only when men knit headgear for themselves that we find any similar degree
of expression of their respective powers of virility, but this will be the subject of
another essay.
Conclusion
Weaving in Qaqachaka is a necessary form of productive activity for both men and
women. It is functional in the productive domain of household tasks, as well as in the
social and cultural obligations of a married couple toward each other and toward their
130 MULTIPLE MEDIA
kin and affines for the rest of their married lives. However, I have argued here that the
wider interpretation of the significance of weaving within Andean cultures must take
into account not only the various aspects of weaving as a gendered activity but also
weaving as a part of a multi-layered and intertextual discourse which is also gendered.
I have therefore argued against the interpretation of the proprietorship of land and its
products or of the animal herds, the means and relations of production that make
weaving possible, according to an exclusively male discourse or one that is external
to the culture. In listening instead to the alternative layers of gendered discourse in
Qaqachaka, most especially to what women say about their work as weavers and its
significance, we become aware of other levels of action and meaning: weaving as
work or as exchange, weaving as production or as a part of reproductive and
transformatory processes, weaving as an art of the hand with its own aesthetic and
information system embedded in it—each forming a separate but interrelated part of
a gendered vernacular discourse.
When we return to the main issue here, the gendering of such activities as weaving
and the writing of texts according to a supposed hierarchy, then we cannot ignore the
evidence from Andean history, nor can we ignore the Andean present. However, the
most challenging evidence that confronts us emerges from women's own discourse
about their weavings. In weaving garments such as the mantle for herself and the
poncho for her menfolk, it is a woman weaver who defines through her work the limits
and obligations of each gender as a social construction. She does so through a
symbolic language, a means of cultural and ideological production continually
re-invented from generation to generation, which defines her own place of power
in the inner and domestic domain of the ayllu, and her husband's place of power
on the periphery and outside of it. In her weavings, a woman not only invents the
symbolic language to express the metaphysical postulates of Qaqa society, but, by
manipulating the woven texts of such gendered garments, she is able to counter-
pose the two different orders of gendered activity and their relative powers of
production, reproduction, and creative intelligence. If any hierarchy exists at all
within this symbolic domain controlled by women, then surely it is one in which
women define for themselves the central place of power, and only then do they
make men in their own image.
ACKNOWLEDGMENTS This essay emerged from a paper given originally to the anthropology
workshop on "Genderrelations, work and proprietorship among the indigenous peoples of South
America" held on 7th-8th December 1990, at the Institute of Latin American Studies, University
of London, and from a departmental seminar given to the Department of Anthropology at the
University of St Andrews, Scotland, in February 1991. Thanks to Nicole Bourque, Tristan Platt,
Penny Dransart, Peter Gow, Joanna Overing, and Rosaleen Howard-Malverde for their com-
ments. Of all the weavers of Qaqachaka whose words I remember here, special thanks to dona
Guadalupe Colque Quispe, dona Bernaldita Quispe Colque, dona Juana Ayka Colque, dona
Nicolasa Ayka, dona Mauricia Mamani Cheque, and most especially dona Lucia Quispe
Cheque, dona Asunta Arias Tarki, and dona Maria Ayka Llanque. Thanks also to Juan de Dios
Yapita, for the fieldwork we did together and the discussions we had afterward about it. The
paper would not have been written at all without him.
Notes
1. See, for example, Watson Franke 1974 on this distinction in the Guajiro culture of
Colombia; March 1983 on the Buddhist Tamang of Tibet; Niessen 1985 on the Toba Batak of
Indonesia; Messick 1987 on the Islamic Berbers of Azrou, Morocco; and Bergren 1983 on
ancient Greece.
2. Historically, the poncho as an item of dress was in use by the Spanish in the seventeenth
century, although the older Andean tunic (urk"u), or short poncho (q"awa), continued to be
worn underneath (Money 1966, cited in Barriga Barahona 1981). The poncho's origin is usually
attributed to the Araucanians, or Mapuche, who are said to have invented it after the introduction
of the horse. I should like to thank Penny Dransart for drawing my attention to this reference.
3. I have used the translation 'blood red' for the Aymara wila, a term that denotes both
blood and its scarlet colour. There are, however, several other colour terms that refer to the
distinct colours of different types of blood: the blood at birth, menstrual blood, and so on. There
are also taboos on using these colour terms in certain contexts.
4. For a more technical definition of chinu, see Gavilan and Ulloa 1992: 123ff.
5. Literally: 'Get up then, damn you, it should arrive over (the earth) then, damn it'.
6. Sophie Desrosiers (1988) discusses the horizontality of women's garments and the
vertically of men's garments.
7. Saya in everyday parlance may also refer to the two moiety-like configurations of
Qaqachaka,patrasoya, meaning 'upper' moiety andmanq"asaya, meaning 'lower' moiety. The
term "saya" is also used in relation to the ritual battles once held in the main plaza, when the
dancing groups of warriors entering the four corners of the plaza were referred to as four "saya"
of warriors.
8. For a historical perspective on the gendered use of horizontal and vertical bands of
colour in styles of dress, see the description of the succullu ritual in Bouysse-Cassagne 1987:
233ff.
9. However, in a comparative study, in the lakeside community of Chukinapi in the
department of La Paz, we have found similar analogies between the food products and the coding
of colours in the textile stripes (see Arnold et al., in press).
Ill
INSIDE THE TEXT:

PROCESSES OF
TEXTUALISATION IN WRITTEN
QUECHUA DISCOURSE
8
WILLEM F. H. ADELAAR
Spatial Reference and

Speaker Orientation in
Early Colonial Quechua
TA he introduction of writing constitutes a major landmark in the history of a

language. When language is transmitted through writing, the mechanisms govern-
ing the production and the interpretation of speech are different from those that
obtain in an oral communication process. If the introduction of writing is success-
ful, two separate traditions may arise, one of which is the direct continuation of
the existing practice of oral transmission. For the second tradition, that of written
speech transmission, to develop, a series of adjustments is required in the language
system. These adjustments occur both at the morphological and at the lexical
levels. Subsequently, the interaction of the two traditions may result in a number
of new varieties combining elements of both. Such complex configurations of
several co-existing and interrelated traditions are found in the national languages
of today's modern nations.
In written communication, knowledge of the situation surrounding the speech
event can be absent. The identity of the participants, in particular of the addressee but
also of the speaker-writer, is not self-evident, as it usually is in oral speech events. In
spoken language, the categories space and time are identified in relation to the place
of speaking and the moment of speaking. When language is written, information about
space and time has to be presented in a different, more precise way. A producer of
written speech not only must refer to time or location in the situation described, he
135
136 INSIDE THE TEXT
may also feel compelled to refer to locations in the text. The latter tendency is less
prominent in oral speech.
The revolutionary character of the introduction of writing must be mitigated some-
what. Traditional societies usually entertain a practice of oral transmission of narrative
discourse. Narratives are often presented in a standardised way and must be under-
stood in a context unrelated to the place and the time of speaking. As in the case of
written text, some adjustments with respect to plain oral communication are in order,
and do, in fact, occur. It is not unusual for an incipient writing tradition to inherit some
of the mechanisms that separate narrative discourse from plain oral communication.1
Incipient literary tradition andnd

the Huarochiri document nt
In the specific cases of Quechua and other languages of the Andean region, writing
was introduced abruptly with the arrival of the Spanish conquerors. A tradition of oral
transmission of texts was in existence. Discursive features of this oral tradition may
have set the path for changes in the language that later accompanied the shaping of a
written linguistic practice. A strong literary tradition never developed in any of the
Andean languages, most written texts being more or less direct reflections of oral
speech.2 Nevertheless, a written tradition in its incipient stage can be observed in one
of the longest and oldest running texts available in Quechua, a document originating
from the province of Huarochiri, not far from Lima. It was written on behalf of one
of the most zealous of Peru's idolatry crusaders, Francisco de Avila.3 The Huarochiri
document has been dated 1608 (Acosta 1987).
The authorship of the Huarochiri document remains in doubt. Apparently, the
author or authors deliberately chose to remain anonymous. Except for a marginal note
on one of the pages of the manuscript, any reference to a possible authorship is
avoided. Whoever wrote the text frequently uses the inclusive plural "we" form,
dealing with his audience of readers, one might say, as if they were his co-authors.4
In chapters 20 and 21 of the Huarochiri document, a principal role is played by
one of Avila's most efficient assistants, don Cristobal Choquecaxa. The vivid descrip-
tion of his adventures is full of detail from personal experience and would make him
a good candidate for the authorship of the document.5 If he were indeed the author,
Choquecaxa would have made efforts deliberately to hide his identity. The stories
concerning him were all written in the third person and in the hearsay form.6 In normal
speech, use of the hearsay form indicates that the speaker disclaims having witnessed
the events he narrates.
The role of the speaker in Quechua discourse

In a Quechua discourse, speakers are compelled to express certain aspects of their
relationship to the events they describe. The grammatical markers used to such effects
cannot be suppressed from the utterance without bringing about substantial modifica-
tions in meaning, if the utterance remains acceptable at all. It follows from the above
that Quechua grammar requires a specification of the speaker-event relation, even
Spatial Reference and Speaker Orientation 137
when the speaker's involvement in the event or his identity become less important or
even irrelevant. For a language with an oral tradition this is not surprising. Since there
is always a speaker who pronounces the discourse or a narrator who tells the story,
there is no reason not to take note of the speaker's or narrator's attitude toward the
event described, his involvement in the action referred to, and so on. But as soon as
the oral tradition is replaced by written text, the need may be felt to eliminate all
reference to the speaker's identity. One may wish to leave unspecified the speaker's
feelings or his participation in the activity described in order to give the text a general
truth value. The compulsory use of grammatical markers referring to such categories
may then be experienced as superfluous or even bothersome.
When a written tradition is about to develop, grammatical markers referring to the
speaker's involvement in the event are either likely to disappear, or to become merely
decorative and fixed. Alternatively, their functions may change. There are at the least
two aspects in which the identity of the speaker and his experience play an essential
role in the make-up of a Quechua sentence. One of them has to do with the source of
the information supplied, which can be from personal experience or observation, from
hearsay or inference, or simply be a matter of conjecture. Since the grammatical
markers used to these effects are found in virtually every single sentence, the speaker
is constantly compelled to specify the degree of certainty and the source of his
contentions.7 A second speaker-related element that must be explicitly indicated in
the discourse is his location in space with regard to the events described. In Quechua,
events have to be specified for certain spatial features for which the position of the
speaker at the moment of utterance serves as a point of orientation. In what follows I
will briefly discuss the different ways in which this phenomenon is dealt with in the
Huarochiri document. Two categories, one syntactic and one morphological, will
receive particular attention.
The role of the proximal demonstrative pronoun

Two demonstrative pronouns, cay and chay, occur in the language of the Huarochiri
document. They correspond roughly to English 'this' or 'this place' and 'that' or 'that
place', respectively.8 There are no separate roots for spatial adverbs such as 'here',
'there', and 'over there'. Instead, the demonstratives are used in combination with the
locative (caypi 'here', chaypi 'there') or directional case markers (cayman 'hither',
chayman 'thither'; caymanta 'hence', chaymanta 'thence').
As a first approximation, cay can be characterised as "proximal" ("place of the
speaker", "object near the speaker"), whereas chay is "non-proximal" ("any place
but that of the speaker", "object not near the speaker"). The proximal demonstrative
cay does not constitute an obligatory element in the sense that a grammatical
marker would be. It can be replaced by the more neutral chay, or be omitted
altogether. As a deictic element, however, it is essential for reference to the location
of the speaker.
The Huarochiri document contains many passages that are direct quotations from
a grammatical point of view. In Andean languages, such as Quechua, the use of quota-
tions is not a stylistic option but a grammatical necessity. Verbs are sub-categorised for
138 INSIDE THE TEXT
the type of complement they require.9 In the language of the Huarochiri document, as
in most conservative dialects of Quechua, the verb ni- 'to say; to consider; to intend'
requires a direct quotation as its complement. Semantically, ni- refers not just to verbal
communication but to mental activities as well. The content of such a mental activity
is presented as a quotation that has the same shape as would have a stretch of speech
actually pronounced by the subject ofni-. The following example illustrates this:
(1) chaysi quiquillantac huc huata chica nononhuan causachircan "pip churinh
cayca" nispa. (chap. 2, sec. 12)10
[For about a year she kept him alive with her own breast, while saying: "Whose
son could this be"? Or: For about a year she kept him alive with her own breast,
wondering whose son he could be.]
In quotations of thoughts, as illustrated in (1), the place-of-speaker function of the

proximal pronoun cay is preserved in the sense that the narrator chooses as a point of
reference the location of the speaker he quotes. Embedded quotations are treated as
independant utterances in their own right, reflecting the structure of such full sentences
in which the narrator would describe his own experiences. No relevance is accorded
to the fact that the quoted speaker does not actually speak.
The following example refers to actual speech uttered by two distinct personages.
As is often the case in Quechua, one direct quotation is embedded within the other:
(2) chaysi yayantaca"yaya, caypim huc huaccha 'yayaiquicta alliyachisac' nimuan"

nispa villarcan. (chap. 5, sec. 48)
[She told her father: "Father, there is a poor man here who says to me: 'I shall
cure your father' ". |
In portions of the text that are not quotations, the speaker orientation of cay is preserved
when there is a reference to the geographical point at which the writing of the text
occurs. In some cases, the place of writing is explicitly mentioned; in others, it is not.
See, for instance, example (3):
(3) chaysi chay quintecunaca anchatac cay checacunacta chicnircan. (chap. 11, sec.
153)
[Those Quintis held these Checas in extreme contempt. Or: The Quintis there
held the Checas here in extreme contempt.]
We know from other sources that the text was probably written in the reduction
village of San Damian de Checas, where the survivors of the Checa nation had
been brought together. At the moment of writing, the author was presumably
surrounded by Checas. Whence his use of the expression "these Checas". The
Quintis, on their side, lived near the town of Huarochirf, a considerable distance
away. A rather remarkable example in which knowledge of the place of writing
is a prerequisite is (4):
(4) caycunaca riman nispa "hunanchanchic 'ychach ari chay' ". (chap. 4, sec. 35)
[These ones speak of it saying: 'We believe that such may actually be the case'.
Or. The local people here comment on it, saying that they believe that such may
actually be the case.]
In examples such as the above, it is not always evident which area functions
as the domain of speaker orientation. It requires imagination to determine the
location of the narrator and find out which people surrounded him at the
moment of writing, a difficulty that will not normally arise in a situation of
oral communication.
Proximal cay is frequently found in expressions referring to extensive areas that
include the place of writing. This fact is illustrated in (5) and (6):
(5) chaysi cay quitipi ancha hunay puricorcan ancha ahca llactacunacta runactapas
llollachispa. (chap. 2, sec. 28)
[For a long time he wandered about in this province deceiving many gods and
people.]
(6) cay llactacunaman natac miticamurcan. (chap. 2, sec. 28)
[He fled towards these communities.]
In these cases, the area is not specified. The writer is confident that his readers will
understand which area is meant. Cay quiti 'this region' and cay llactacuna 'these
communities', 'this region', refer to highland provinces, or simply to the high Andean
region in general, as opposed to the lowlands near the ocean. Even when the narrator
maintains a certain distance from what he is writing, he is constantly aware of the place
where he finds himself at the moment of his writing.
Beside this geographically motivated use of the proximal pronoun, we can also
observe a text-adapted version of it. The demonstrative cay may, in fact, mean 'this
place in the text', as in (7) and (8):
(7) caypim churani cay huc yayayuc guarocheri niscap machoncunap causascanta.
(preface,sec. 2)
[Here (in the text) I put (write down) the traditions of the forefathers of the people
of Huarochiri, who descend from a common ancestor.]
(8) coymantam natac pariacacap pacarimuscan callarinca. (chap. 5, title)
[From here (in the text) will start (the story of) Pariacaca's beginnings.]
The demonstrative cay frequently refers to what follows in the text. Illustrative
examples are (9) and (10). Example (9) is a recurrent expression in the document of
Huarochiri:
(9) chay simiri caymi. (chap. 3, sec. 29)

[That story is this. Or: That story is as follows. Or: This is how the story goes.]
(10) coytam coysaua quillcasson. (chap. 20, sec. 259)
[This we shall write down hereafter.]
140 INSIDE THE TEXT
The demonstrative cay may also refer to the preceding section of the text, 'what
we have just been talking about', 'what we have just explained', as in (11):
(11) cay isconnin capitulopi . . . (chap. 10, sec. 141)

[In this ninth chapter (just finished) . . . ]
(12) cay coycunacta villanacuptinsi chay huatyacuri nisca huyarcan. (chap. 5, sec.
46)
| While they told each other all these things, Huatyacuri was listening.]
In cases of recurrent reference to several persons or entities of more or less equal

importance, cay can be used in order to keep the attention fixed on the principal topic
of the account. Digressions or asides are characterised by the presence of chay, the
main thread of the account by cay:
(13) chaymantas pariaeacaca cay huallallo caruinchocta atipac rispas . . . (chap. 8,

sec. 98)
[Then, when Pariacaca set out in order to defeat this Huallallo Caruincho ... 1
The use of cay is rather consistent. Apart from the original deictic place-of-speaker
reference, we find several uses that fit well in a narrative tradition, either oral or
written. In some cases, cay refers explicitly to a place in the text. This fact may be
interpreted as a first adjustment to the needs of written language. The demonstrative
cay is more marked semantically than chay and less frequent. The function of chay
varies from a clearly deictic one to something near to a definite article in English.
The role of the verbal derivational marker -mu/-

The one grammatical marker that implies reference to the location of the speaker is
the verbal suffix -mu-. Like the other verbal affixes traditionally described as deriva-
tional,11 it is inserted between the verbal root and the endings referring to tense, mood,
and person. The suffix -mu- is extremely frequent. Although Quechua dialects may
differ substantially in their morphology, the suffix -mu- is not known to be absent from
any of them. Also, its use is quite constant throughout the dialects.
Verbs that imply motion in a particular direction are obligatorily marked with -mu-
if the motion to which they refer is directed toward the speaker. Alternatively, a
narrator may choose a point of reference different from his own location to which
motion denoted by a verb with -mu- is directed. In an unmarked situation, the location
of the speaker will be automatically interpreted as the point of reference for -mu-.12
The absence of -mu- from a verb referring to a change of location implies a
movement away from the speaker or a movement that is neutral with respect to the
location of the speaker. The following elementary examples illustrate the way move-
ment toward the speaker and its opposite are indicated in present-day Quechua
dialects:
(14) apamuy apay

'Bring it (to me) ! ' 'Take it away (from me) ! '
(15) apamuy kayman *apay kayman

'Bring it here (to me)!' (ungrammatical expression)
The presence of a noun phrase specifying direction toward the speaker (e.g., kayman
'hither', kay llaqtaman 'to this village or town') may accompany a verb with -mu-, but
can never replace it. This is shown in (15), where the expression kayman 'hither' can
be combined with apamuy but not with apay.
The semantic contrast emerging from the absence or presence of -mu- is compa-
rable to the difference in meaning that separates the members of the lexical pair
ri-/hamu-. The verbs ri- and hamu- are used in the southern Quechua dialects and in
Huarochiri Quechua with the meanings 'to go' and 'to come', respectively.13 The verb
pairs ri-/hamu- and apa-/apamu- are reminiscent of the Spanish lexical pairs ir/venir
and llevar/traer. Both in Quechua and in Spanish, the semantic contrast is mainly
based on the criterion of whether or not there is a motion toward the speaker.
In Quechua, the use of -mu- in its function of indicating direction towards the
speaker is not restricted to events referring to real movements. It can also manifest
itself in events that imply some sort of a psychological approach to the speaker. As a
result of such an event, the subject of the verb that contains -mu- becomes more visible
or otherwise perceptible to the speaker than it was before the event took place. Verb stems
that are frequently used in this way in the Huarochiri document are listed under (16):
(16) non-existing to existing pacarimu- 'come into existence' (chap.

3, sec. 36)
not living to living causarimu- 'revive' (chap. 1, sec. 4)
yurimu- 'be born' (chap. 5, sec. 51)
not active to active pucorimu- 'start to blow (of wind)'
(chap. 18, sec. 225)
invisible to visible llocsichimu- 'cause to come to light';
'cause to come out' (chap. 6, sec.
83)
ricurimu- 'appear' (chap. 13, sec. 175)
silent to speaking rimarimu- 'speak up (in a meeting or
after silence)' (chap. 3., sec. 30)
rolled-up to spread out pahyamu- 'explode', 'breakout' (chap.
31, sec. 435)
few to numerous mirarimu- 'increase in number' (chap.
3, sec. 34)
A second, but optional use of -mu- is found with verbs that do not primarily refer
to motion in a particular direction. With such verbs, -mu- refers to a definite place at
a substantial distance from the speaker, usually involving a previous movement from
the speaker's location to that place. At the same time, there is the suggestion of a
circular movement, as the effects of the action referred to are expected somehow to
142 INSIDE THE TEXT
revert to the speaker. The suffix -mu- in these cases can be translated as 'there', 'over
there", or 'where 1 am telling you'.
This second function of -mu- is less frequent and not obligatory. If a noun phrase
referring to a place indicated by -mu- is added to the verb, the presence of the suffix
is no longer required. It is a different situation from that which obtains when -mu-
occurs in its 'direction toward the speaker' function. As we have seen, agreement with
a possible directional noun phrase is obligatory in the latter.14
In what follows, several characteristic instances of -mu- as it occurs in the
Huarochiri document will be discussed. In them, the use of the marker -mu- is quite
similar to that in present-day dialects of Quechua. It is best preserved in direct
quotations. Examples (17) and (18) illustrate the function of speaker-directed motion
of -mu-\ (17) also contains the lexical counterpart of -mu-, the verb hamu-:
(17) huc atoc anas huarminhuan hamunca, chaymi huc huchoylla porongollapi
ashuanta apamunca. (chap. 5, see. 58-59)
[A fox and his wife, a skunk, will come this way. They will bring 'chicha' in a
small jar.]
(18) pana cauellaca, cayman cahuaycumuay. (chap. 2, see. 17)
[Sister Cauillaca, please look at me here!]
In (18), cahuaycu- 'to look into a particular direction' is treated as a verb of motion
as a result of the presence of the suffix -ycu-. It should be observed that there is no
strict sub-categorisation of the verb class based on the motion vs. non-motion distinc-
tion. Some verbs, such as chaya- 'to reach, to arrive, to hit' can be used both as motion
and as non-motion verbs. Cahua- 'to look' becomes a motion verb through the
extension of the stem by means of the suffix -ycu-.
The use of -mu- in the verbs listed in (16), which refer to an increase in percepti-
bility, are frequently found in the Huarochiri document. As we have seen, such verbs
can denote change, appearance or sudden diffusion. They occur in direct quotations,
as well as in the embedding narrative:
(19) cayaminchallam yurimunca. (chap. 5, sec. 51)

[One of these days his birth will take place.]
(20) chay llamaca runa yna rimarimuspa nispa nircan . . . (chap. 3, see. 30)
[The llama suddenly began to speak like a human being and said ... ]
Example sentences (21) to (23) from the Huarochiri document illustrate the use
of -mu- referring to an action that occurs at an indicated place at a distance from the
speaker ("action over there"). It occurs both in quotations and in the main narrative:
(21) cayallatacmi chaypi chasquechimuson. (chap. 9, sec. 128)
[There tomorrow we shall make him receive them.]
(22) huatyacori sutioc pariacacap churin niscatac naupaclla ricumorca yachamurca.
(chap. 5, sec. 38)
[Huatyacuri, Pariacaca's son was the first to (go and) see him there and learn
about him there.]
(23) na cay rarcacta pichayta puchocamuspari runacunacta anchapunis cornbidac

carcan saracta purotucta. (chap. 7, sec. 94)
[After having finished this canal-cleaning there, they would invite the people to
lots of maize and beans.]
The following example (24) consists of two correlative clauses, one of which
functions as an accessory background description to the other one. Both verbs
contain -mu- in its "action over there" meaning, even though the accessory verb
is a verb of motion. It illustrates the absence of a clear-cut lexical division between
motion verbs and non-motion verbs in Quechua (both verbs behaving like non-
motion verbs):
(24) maypacham sieneguillaman huraycomunchic chay horcucta, chaypis pono

murca. (chap. 5, sec. 42)
[Where there is a mountain which we pass when we go down to Cieneguilla, that
is where he slept.]
Example (25) is a clear illustration of the circular movement associated with the
"action over there" reading. It is also found in present-day Quechua dialects and can
be glossed as 'to go and act over there and then come back':16
(25) rarcactam pichamuni, padre, taquecusac upiacusac. (chap. 7, sec. 95)

[We have been to clean the irrigation canal, father. Now we shall dance and
drink.]
Of greater importance for the concept of speaker orientation and its fate in a written
environment are those instances of -mu- (and hamu-) that indicate 'direction toward
the speaker' in passages that are not part of a direct quotation. Logically, in such cases
-mu- should refer to the place of writing, but this is not always the case because in
narratives the orientation point for -mu- can be transferred to the principal site of
action—that is, the place where the speaker wants his audience and himself to imagine
themselves as being. Nevertheless, in the Huarochiri document, the author is still very
conscious about his "place of writing" and seems confused as to which place should
be chosen as an orientation point in individual instances. The use of -mu- in relation
to shifting orientation points is in fact quite inconsistent and unpredictable. This is
symptomatic of an incipient literary tradition in which the role of the narrator cannot
yet be duly disposed of.
Writer orientation has as its domain the community of San Damian de Checas as
opposed to other communities in the area, or it may be centred in the Huarochiri
highlands in contrast to the coastal lowlands or the Anti (Huanca) region further inland.
There is, however, an additional complication which originates from the mythical
character of most of the Huarochiri narrative. Mythical places may be less well defined
than concrete, geographically defined locations frequented by the author.
In chapter 2, the myth of Cuniraya, who follows Cauillaca from the meeting
at Anchicocha to the ocean shore, contains no instances of -mu- until the ocean is
reached. Anchicocha, situated in the highlands like San Damian, is used as the
point of orientation. An alternative explanation would be that the author treats
144 INSIDE THE TEXT
Anchicocha and San Damian in the same way because both places are at equal
distances from the sea. Sentences (26a-d) describe Cuniraya's arrival at the ocean
shore and what happened subsequently. Instances of -mu- suddenly become fre-
quent:
(26a) cocha patapi chayaspaca . . . pachacamacnicman cutimurcan. (chap. 2, sec. 27)

[Having reached the seashore, he turned back toward Pachacamac (writer's
direction).]
(26b) chaymantas chay cuniraya viracocha niscaca miticamurca cocha patanicta.
(chap. 2, sec. 27)
[Then Cuniraya Viracocha fled along the seashore . . . ]
(26c) ancha pinaspa catimurcan. (chap. 2, sec. 27)
[Being very angry she came running after him. (as 26b). |
(26d) cay llactacunaman natac miticamurcan. (cf. 6). (chap. 2, sec. 28).
[He fled toward these communities . , . |
The inland movements exemplified in (26a) and (26d) have -mu-, whereas
movements toward the sea remain unmarked. However, in the passage referring
to the movements of the protagonists along the seashore away from the sanctuary
of Pachacamac the author also uses -mu-. It appears as if, in search of a new
orientation point, he imagined that he were looking at Pachacamac from a distance,
seeing the heroes coming toward him. The "action over there" interpretation can
be excluded because there is no reference to a definite location, no previous
movement toward it, and no circularity.
In chapter 5, we observe a sudden change in the point of reference for speaker
orientation, which is clearly selected on the basis of a shift in the focus of the
narrative. Speaker-directedness is indicated by the lexical verbal expression hamu-,
whose semantic value is similar to that of the affix -mu-. In (27a), the verb hamu-
has the connotation of 'to appear on the scene' (invisible to visible), a semantic
interpretation parallel to that of the affix -mu- in (16). First, people are said to
come from different places to visit the semi-deity Tamtanamca at Anchicocha. The
orientation point is Anchicocha, not San Damian:
(27a) cay runacta chica alli causascanta ricuspas tucoy hinantin llactacunamanta
hamuspa. yupaicharcan mucharcan. (chap. 5, sec. 40)
[Because they saw how well this man lived, people came from all communities
to honour and worship him.]
Then, Huatyacuri approaches the site coming from the sea:

(27b) chaysi cay huatyacurica chay pacha ura cochanicmanta hamuspa . . . (chap. 5, sec.
42)
[At that time, Huatyacuri was coming from the lowlands near the sea and . . . ]
And, eventually, the orientation point is transferred to the spot where Huatyacuri is
resting. Two foxes approach from either side:
(27c) huc hatocca uramanta amusca hocri hanacmanta amuscatac, chaysi chaypi
pactalla tincuspas . .. (chap. 5, sec. 42)
A fox came from below and another one came from above. They met each other
there halfway and .. .
Chapter 6 describes Pariacaca's visit to the town of Huarochiri. In it, the author
takes care to indicate that Huarochiri is not his (the author's) location. Obviously, his
perspective is from San Damian, the place of writing. All motion toward Huarochiri
and, subsequently, to the village of the Cuparanear San Lorenzo de Quinti is expressed
by means of verbs that do not contain -mu-. Interestingly enough, the avoidance of
-mu-is even maintained in the description of an irrigation canal that emerges from the
mountain at San Lorenzo de Quinti:
(28a) cay pucyum carian san lorenco hanacnin aton orcomanta llocsircan. (chap. 6,
sec. 81)
[This well now emerged from a high mountain above San Lorenzo.]
The verb llocsircan 'it emerged' is associated with an act of appearance, "invisible
to visible", by itself a sufficient ground for -mu- assignment. Apparently, the author
wishes to stress that the canal and its source are miles away from his own location. He
cannot see them from where he writes. Therefore, he leaves -mu- aside. On the
other hand, the author seems to know the place from his own observation, a fact
shown by his use of the sentential suffix -m (in pucyum). Its presence indicates first-hand
information.
Later on, the author tells us how the same canal is dug out by animals. At that
moment, his observation point shifts to the foot of the mountain from which the canal
takes its source. He now uses -mu- in every verb of motion in function of his new
imaginary perspective. The following sentences describe the fox's part in the digging
of the canal:
(28b) yna naupamuspa na chaupicta san lorenco hanacnin hurcucta sequemuptinsi

concayllapi huc yutuca "pisc, pisc" nispa pahuarimurcan, chaysi chay atocca
"huac" nispa muspaspa huraman hurmamurcan. (chap. 6, sec. 87)
[As he [the fox] was leading onward and had proceeded halfway up the
mountain above San Lorenzo, marking out the course (of the canal), a partridge
suddenly flew up and said "pisk"! Then, the fox said "wak"!, lost his foothold
and fell to a lower level.]
A very unexpected change in the observation point of the narrator can be seen in
the following examples, taken from a passage in chapter 8. It tells the story of the
Pariacaca brothers meeting a man on his way to sacrifice his son to Huallallo
Caruincho. At first, the orientation point is the mythical place of the meeting, where
the Pariacacas are playing:
(29a) chaysi chay pacha huc runaca huacacuspa hamurean huc churinta apaspa. (chap.
8, sec. 99)
[At that time, a man came along crying, carrying one of his sons.)
146 INSIDE THE TEXT
(29V) mancharispa tucuy apamuscanta corcan. (chap. 8, sec. 101)

[Being scared, he gave them everything he had brought with him.]
Then, the orientation point is transferred to the village where the man lives (29c). The
reason for this may be that the author feels a closer geographic association between
that man's village and his own place than between the latter and the location of the
mythical scene high up in the mountains to which he is referring. The association may
be one of altitude or climatological conditions:
(29c) chaymantas chay runaca cutimorcan churinta apacospa. (chap. 8, sec. 101)
[Then that man returned home (literally, came back), carrying his child.]
Subsequently, the man returns to the place where he had previously met the Pariacaca
brothers. The verb is no longer marked with -mu-, showing that the real orientation
point (place of speaker/writer) has ultimately prevailed over the mythical orientation
point (place of main scene in narrative):
(29d)ynaspas natac pihcantin ponchaomanta natac chay pariacacap simincama

cutircan. (chap. 8, sec. 102)
[And so he returned (to that place) following the orders of Pariacaca. 1
Conclusion
The way speaker orientation is dealt with in a text such as the Huarochiri document
provides evidence of incipient adjustments to the requirements of writing. In the
case of the proximal demonstrative pronoun, there are clear instances where
proximity is established in terms of a location within the text. The use of the
pronoun seems to obey a particular standard both for unmarked situations (location
of the speaker/writer) and for specific text situations (location in the text, location
in the thread of the narrative). On the other hand, the factors that dominate -mu-
assignment to verbs of motion are much less well defined. There seems to be a
conflict of tendencies between different perspectives that can be used. Either the
place of writing may determine the presence of -mu- in a verb of motion or an
imaginary observation point chosen by the author and liable to be abandoned at
any moment may do so. Sometimes the author seems to be well aware of his
present location and individual personality; at other times he forgets these and
travels along with the heroes of his narrative to some place far away. It would not
be correct to say that the orientation point for -mu- assignment in a narrative is
always the location of the event in focus. Such may be the case for a moment,
but there is a tendency for the author to shift his orientation point back to the
place of writing. Since this can hardly be called functional in an anonymously
written text, not intended to be place-bound nor person-bound, we may hold that
-mu- assignment was still far from being well adjusted to the requirements of a
written text. On the contrary, the contextual exigencies of actual speaking, in which
the place of speaking is often highly relevant, seem to prevail. There is a
discrepancy, then, between the degree of adaptation to text requirements of the

pronominal system and direction marking in the verb. It is in the latter that the
insecurity felt by the author becomes most conspicuous. That the pronominal
system shows more advanced adjustments may be explained by the supposition
that it is easier to manipulate and assign new meanings to lexical elements than
to morphological categories.
To recapitulate, we may say that in early-seventeenth-century Quechua, a clear
strategy for handling distinctions of speaker orientation in texts was not yet well
developed. Yet, there existed tendencies that may have been indicative of possible
future developments in case the language were to move toward a written tradition.
One of these tendencies concerns the use of demonstratives for referring to parts of
the text itself or to the thread of the narrative. Tentatively, another such tendency may
have been the practice of temporarily replacing the speaker's location by the place of
the event in focus as an orientation point for marking direction on the verb.
Notes
1. For a discussion of Quechua narrative discourse and some of its characteristics, see
Howard-Malverde (1988, 1994) and Gutmann (1994).
2. It is not the case, however, of the theatre plays written in the seventeenth and eighteenth
centuries (see Meneses 1983 and Mannheim 1991).
3. The document of Huarochiri has been published with an annotated English translation
by Salomon and Urioste (1991). For a Spanish translation with a discussion of the historical
background of the document, see Taylor (1987a) and Acosta (1987). The examples in this article
were translated into English from the Dutch edition (Adelaar 1988).
4. In our view, the Andean inclusive plural must not be defined primarily in terms of its
opposition with the exclusive first-person plural. More accurately, it refers to a sort of general
group belonging as does on in French or "people" in English (Adelaar 1993).
5. For an analysis of the role played by Choquecaxa, see Kemper Columbus (1994).
6. In Quechua, hearsay is indicated by the presence of a sentential suffix -s or -si.
7. For a detailed discussion of evidentiality in the Huarochiri document, see Dedenbach-
Salazar Saenz (in this volume).
8. Huarochiri Quechua agrees with the northern Peruvian and Ecuadorian varieties of
Quechua (corresponding to the Quechua IIB division in Torero's classification) in that it
distinguishes only two degrees of distance in the demonstrative system (cf. Adelaar 1994d).
Central and southern varieties of Peruvian Quechua distinguish three degrees, comparable to
Spanish este, ese, and aquel. The demonstrative system of Pacaraos Quechua (Huaral, Lima)
also encodes differences in altitude (Adelaar 1982).
9. For a more detailed treatment of quotations in Quechua and in other Andean languages,
see Adelaar (1990).
10. Chapter 2, section 12: the section subdivision follows Salomon and Urioste's 1991
translation of the Huarochiri document.
11. The qualification of -mu- as a derivational affix follows an established usage of the
linguistic literature on Quechua. From a strictly linguistic point of view, -mu- does not exhibit
the characteristics of a derivational affix. For a discussion of this matter, see Adelaar (1994c).
12. The suffix -mu- and its functions have been discussed by Bills (1972) and by Van de
Kerke and Muysken (1990). Virtually all reference grammars of specific Quechua dialects
contain a section on -mu-. The most extensive one is found in Weber's (1989) analysis of
148 INSIDE THE TEXT
Huallaga Quechua. For a slightly different approach in terms of actant orientation and the
centripetal-centrifugal opposition (-mu- being interpreted as centripetal), see Taylor (1990:
309-311; 1994:60-61, 140),
13. In Huarochiri Quechua, the actual form is amu- (often written hamu-). In this respect,
Huarochiri differs from most of the northern Peruvian and Ecuadorian varieties, which have
shamu-.
14. The labels "cislocative" and "translocative" are frequently found in the linguistic
literature on Quechua for the purpose of referring to the different uses of-mu-. Another possible
label for the cislocative function would be "ventive", which is commonly used in Africanist
studies.
15. -mo is an orthographic variant of -mu-.
\ 6. At this point, not all occurrences of -mu- have yet been accounted for. The suffix -mu-
may occur, for instance, with the verbs ni- 'to say', as in nimuan 'he says to me' (along with
nihuan, same meaning), or with cu-/co- 'to give', as in cumumm 'he gave it to him' and
camuaspa 'because (you) have given it to me'. It is not quite clear whether these instances of
-mu- may be interpreted as 'action over there' or are due to lexicalised combinations of a
particular suffix with a particular root. The combination of suffixes -mua- (~mu~ + -wa-) could
be an emphatic alternative for -wa- 'to me'. None of these possible explanations seems to account
for all the attested cases.
9
SABINE DEDENBACH-SALAZAR SAENZ
Point of View and Evidentiality in the

Huarochiri Texts (Peru, 17th Century)
I In this chapter, I shall look at some texts from the Peruvian seventeenth-century
Huarochiri Quechua manuscript Tradiciones de Huarochiri (c. 1608). It is probable
that these texts were narrated and written down by one or several Quechua speakers
who had contact with Francisco de Avila or persons who knew him, but it is also
possible that Avila himself copied and/or redacted these texts.1 As we do not know
much about the circumstances in which the texts came to be written down, an analysis
of the discourse structure should help us get closer to the Quechua dialect used, to the
narrators involved in their (re-)production, and to the quality of the texts as a collection
of narratives and "ethnographic" and "testimonial" documents that originated in a
non-literate society and yet became fixed in the written medium.
As the oldest known source on Andean culture written in an Andean language, the
Huarochiri texts provide a starting point for the analysis of Andean verbal art. They
date from a time when people first began to use the new medium of writing in order
to express themselves and to convey information about their culture. As the author of
these texts is unknown, the author-narrator-audience relationship has to be explored
without a situational performative context, purely on the basis of the texts themselves.
Although at first sight they appear unique, these texts are in some ways typical of
Andean traditional narrative: their content refers to ancient Andean beliefs, but already
situated in a colonial setting; this is reflected in how language itself is used,2 and in
the very fact that the, texts are written down. Thus these texts have methodological
implications for the analysis and understanding of traditional Andean literature,
149
150 INSIDE THE TEXT
Here, I will concentrate on how the author-narrator(s) present(s) a certain point of

view through the usage of evidential devices. Before entering into the discussion, I
should like to look briefly at the state of research and clarify the concepts I use. With
regard to the concept of "point of view", 1 have found Lanser's approach very
stimulating, and her definition of the concept, as follows, useful for the present study:
If we understand point of view to concern the relations between narrating

subjects and the literary system which is the text-in-context, then we confront a
complex network of interactions between author, narrator(s), characters, and
audiences, both real and implied. At the very least, the notion of point of view
subsumes those aspects of narrative structure that concern the mode of presenting
and representing speech, perception, and event; the identities of those who speak
and perceive; their relationships with one another and with the recipients of their
discourses; their attitudes, statuses, personalities, and beliefs. Moreover, because
narrative has the capacity to be multi-discursive . . . works of narrative often
involve numerous subject/system relationships. In such cases, the textual per-
spective may become a superstructural synthesis of the various voices and
perspectives—points of view ---encoded in the discourse. (Lanser 1981: 13-14)
With respect to the relationship among the participants in the literary discourse,
Lanser distinguishes "status", i.e., the relationship of the author with the literary act;
"contact", i.e., the interaction between the narrator and the addressee; and "stance",
i.e., the attitudes of the characters of the text toward the represented world.-3
Whilst with respect to the Huarochiri texts virtually nothing is known about the
status of the narrator(s), it is possible to study them as far as the contact between
narrator and addressee and the position of the narrator is concerned. Besides style and
content of the narrative, there are also certain morpho-syntactic devices that give
information about the point of view. Thus the usage of the evidential suffixes, together
with other morphological devices, such as tense and space markers, indicate the point
of view adopted by the narrator. They also reveal something about the subjectivity of
the information, the narrator's view, and his attitude toward the narrated account, as
well as the position of the narrator in the cultural context.
Research carried out in the field of evidentials can best be judged on the basis of
the bookEvidentiality. In their introduction, the editors (Chafe and Nichols 1986: vii)
say that evidentiality finds itself at a "stage of exploration", and although there have
been several studies over the past years with reference to evidentiality in Andean
languages, no new attempt at summarising and systematising has as yet been made.
In order to mark evidentiality, different linguistic means are used in different
languages, as for example morphological affixes or lexical predicates. As to their exact
function, some languages are described as having markers that exclusively refer to the
expression of evidence; others show forms that have other functions at the same time.
Anderson suggests the following criteria for the identification of evidentials:
(3 a) Evidentials show the kind of justification for a factual claim which is

available to the person making that claim, whether direct evidence plus
observation (no inference needed); evidence plus inference; inference (ev-
Point of View and Evidentiality 151
idence unspecified); reasoned expectation from logic and other facts; and
whether the evidence is auditory, or visual, etc.
(3b) Evidentials are not themselves the main predication of the clause, but are
rather a specification added to a factual claim about something else.
(3c) Evidentials have the indication of evidence as in (a) their primary mean-
ing, not only as a pragmatic inference.
(3d) Morphologically, evidentials are inflections, clitics, or other free syntactic
elements (not compounds or derivational forms). (1986: 274-275)
In two Andean language families, Quechua and Jaqi, evidentiality is marked

grammatically. Verbal markers are essential for indicating an experienced or non-ex-
perienced action, and there is an interplay of these with discourse markers (tradition-
ally called comment suffixes in Quechua and sentence suffixes in Aymara), which are
also crucial for giving evidence of personal or hearsay knowledge (Hardman 1986;
Weber 1986; Briggs 1994; Calvo Lopez 1994; and others).
Weber (1986), in his essay on the so-called evidential suffixes in Quechua, points
out, through examples from different dialects, that the suffixes which indicate the
perspective on the information of a sentence are to be considered as evidential rather
than validational—that is, they express how speakers came by their information, and
not primarily the commitment speakers have toward the truth of what they say, Weber
gives the different usages of the reportative suffix -si, one of which is its use when
direct experience is unlikely and another is when the speaker wants to escape the
implication of direct experience.
In her study "Talking about the Past: Tense and Testimonials in Quechua Narrative
Discourse", Howard-Malverde (1988) describes the complex use of the suffixes that
indicate the past tense in their interrelationship with the evidential suffixes in a
Quechua dialect of the Peruvian central highlands. She distinguishes a "non-per-
sonalised" mode of description of events from a "personalised" mode that shows the
participation of the speaker's subjectivity (Howard-Malverde 1988:127-128,133-134).
With similar intent I shall look at how the evidential reportative suffix -si, in
comparison with the affirmative -mi, and the suffix of conjecture -cha are used in the
Huarochiri manuscript.4 I will focus—by means of analysing some chapters of the
manuscript—on the usage of the evidential or commentary suffixes, the inclusive/ex-
clusive verb plural, as well as the narrator's explicitly expressed opinion on the
narrated account, and on how far he can be said to be an eyewitness of the narrated
events. The interplay of these devices should help us in answering the following
questions:
What was the situational context when the stories were told?
What is the attitude of the narrator toward the narrated text and toward the
addressee?
Are there different text "genres" present in the manuscript?
Two genres easily distinguishable in the Huarochiri manuscript are narratives, on

the one hand, and what I have termed "ethnographic descriptions", on the other, which
I shall examine in turn.
152 INSIDE THE TEXT
Narratives
As one would expect in a narrative about events that occurred in a period of time
that can be considered mythical (in Quechua marked by the time adverbials nawpa
pacha 'in former times' and chay pacha 'in those times'), the narrator uses the
suffix -si to mark the narrated account as not witnessed by himself, combined with
either the simple or the habitual past, sometimes changing to the present tense.
The combination of the narrative past with the reportative suffix -si is widely
documented in southern Peruvian Quechua narrative. However, in the case of the
Huarochiri texts, there is a closer similarity to the pattern found in San Martin
Quechua, insofar as consistent usage of the simple past with -rka- is found in the
narrative structures of both.5 Thus, in the chapters that relate a coherent narrative,
usually of mythical character, we find the combination of the reportative evidential
suffix -57 and the -rka- past tense:6
(1) § nawpa pachas inti wanurkan /

chaysi chay wanuskanmanta pichka punchaw tutayarkan /
chaysi rumikunaka paypura waktanakurkan /
chaymantas kay mortero muhkakunari chaymarita kay maraykunapas runakta
mikuyta kallarirkan /
llama urkukunari hinatak7 —runakta na katirirten /
(Comment:)
kaytam kanan nukanchik christianokuna unanchanchik jesu christo apunchik-
pak8 —wanuskanpi tutayaskantah /
kaykunaka riman
nispa "unanchanchik ichach ari chay" / (chap. 4, fol. 66v)
[It is said that in ancient times the sun died.
Then it is said-—after its death it became dark for five days.
Then—it is said—the stones banged against each other.
After that -it is said—this mortar, the grinding stones and after that also these
quern-stones began to eat the people.
And in the same way the llama and the mountains began to drive the people.
(Comment:)
This—I know—now we Christians understand as the probable darkness at the
death of our lord Jesus Christ.
These people, however, say: "We believe that maybe, it was that (what has been
narrated)".]
But the reportative mode9 is not only used in narratives with mythical character;
it also serves to relate events that happened in a not very remote past connected with
historical persons, but which were not witnessed by the narrator. Thus Chapter 20
starts with the account of the appearance of a huaca10 in the past (-rka- and -si, chaysi
[ 1 ]), then changes to the immediate historical past and to the present (-rka-/unmarked
tense, -mi, kaymi [2]) in order to introduce the person who experienced what the
narrator then tells in the reportative mode (-rka- and -si [3]) as he has been told about
these events:
(2a) kay waka rikurimuskantas, huk warmi lantichumpi sutiyuk alaysatpa ayllu
tarirkan
chakrakta, uryakuspa /
chaysi
huk mita nawpak aspispaka
"imah. kayka" nispas
pachallanpitak wischurkan / (chap. 20, fol. 84-85r)
(2b) kaymi [don xpistobal] huk mita kay supay11 llucllayhuancupa niskanchikta
rikurkan nawinwan
paypas yayan wanuskanmantaka na niskanchik mana alli supay
machukunap llullaykuskan kaspa /
kay simiri kay hinam l / (chap. 20, fol. 86r)
(2c) huk tutas don xpistobal, chay llocllayhuancupap wasinman rirkan,
chaypi sipasnin kaptin,
chay wakaktaka, na, hakispa
mana na asllapas chaytaka, yuyaspa, /
chaysi,
na chay wasipi chayaspa
ispaykuypak
chay huchuylla rakay wasillaman yaykurkan, / (chap. 20, fol. 86r)
(2a) [It is said that a woman called Lantichumpi of the Alaysatpa ayllu discovered
this huaca that had appeared when she was working the field.
Then—it is said—once, when she was digging it out for the first time, saying—it
is said—" What could this be", she just threw it to the ground again.]
(2b) [This one [don Cristobal]—/ know (it/him?)—once saw this "supay" 11 called
Llucllayhuancupa with his eyes, being, after his father's death, also deceived
by the ancestors' bad supay we have already mentioned.
This story is like this—I know it.]
(2c) [It is said that one night don Cristobal went to that Llocllayhuancupa's house—
where his girl was—having already abandoned that huaca, almost not remem-
bering it anymore.
Then—it is said—after having arrived at that house, in order to urinate, he entered
into that tiny derelict house.]
We can thus define the usage of -si in narratives as marking a temporal and often also
spatial distance of the narrator from the narrated events.
Whenever the narrator takes an explicit position in a story—that is, when he
comments on what he has told, he changes from the reportative -si to the affirmative
-mi, usually accompanied by the inclusive plural -nchik, maybe implying the wish to
include the addressee in his statements, which is understandable when he sets himself
off from the "pagan" content of the narrative, as, for example, in (1) above, that tells
about the rebellion of animals and objects against man after the death of the sun
(unanchanchik 'we understand'; apunchikpak 'of our lord'). It is possible that this
commentator is not identical with the narrator, and at the current stage of my research
this hypothesis seems to be substantiated by the structure of the sentences and also by
the organisation of the chapters within the whole of the text corpus. 12
154 INSIDE THE TEXT
The narrator also uses -mi when he refers to preceding or following chapters:
(3) kaytam kay kipanpi churasun atinakuskantawan /

nam an chay huallallo carvinchup kawsaskantaka runa mikuskantawanpas ima
hayka ruraskantawanpas nawpak capitulopi rimarkanchik I
kananmi rimasun huarocheripi chay chay kitipi ruraskankunakta /
chay simiri kaymi -1 (chap. 6, fol. 70r)
| This—I know—we will put down later on together with (the story of) their
mutual fight.
We spoke already- 1 know it—in the first chapter about that Huallallo
Carvinchu's life and his eating human beings and all his deeds.
Now—I know it- -we will speak about all his deeds in Huarocheri and those areas.
That story is like this -I know it]
Here the narrator uses the verb rima- 'to speak', which refers to the oral performance
of a narrative, whereas the other verb used, chura- 'to put', implies writing and paper
rather than oral transmission. The passage quoted does not seem to reflect a traditional
device of oral art, although one might find similar references in performance cycles
indicating, for example, that "yesterday we all have heard about the deeds of so-and-
so". It is possible that these connective devices were created by the redactor. Here we
have to consider that, when writing down myth cycles, they have to be ordered in a
certain way, depending on the purpose of the work and on the supposed audience. This
process of re-arranging myths in the written medium in a different or new, now
"chronological", order is probably reflected in these texts; however, it cannot be
verified because the original contextual situation is not retrievable. The redactor's
composition and arrangement of the chapters seems to show an effort at chronological
ordering from the most ancient events towards more recent ones, and within this one
can see a "fieldwork" situation in which the narrator(s) first give the myths and
then—maybe as they were asked to do so—give the "ethnographic" follow-ups, e.g.,
the myth about the goddess Chaupinamca, followed by the description of the rituals
carried out in relation to this same goddess. Just as with the conspicuous use of certain
suffixes that reflect the writer's still imperfect familiarity with writing (Adelaar, in
this volume), with respect to narrative structure there is a discernible effort at imitating
European literary patterns on the level of chapter arrangement.
Going back to the usage of -mi, the narrator also uses it when he talks about
geographical features and socio-political divisions of the region that occur in the story.
He changes from -si to -mi, and sometimes to -nchik, and back to -si again:
(4) chaysi kay wakinnin wawkinkunaka I

rispapas
(insertion made by the narrator referring to the geographical position:)
chay tupicochamanta wichaykuspa
nawpa nanta rinchik /
chaymi quisquitambo sutiyuk /
hukmi/ tumnacha sutiyuk /
maymantam limac niktapas rikunchik I
(end of insertion)
chaymantas nan
"tutayquirika tukuyta atiyta na puchukan" niqta uyarispas13 kutimurkanku /
(chap. 12, fol. 107r)
[Then—it is said—his other brothers, despite having gone on (insertion:)
when climbing from that Tupicocha we go along the old path; that one—7
know—is called Quisquitambo; the other one—I know—is called Tumnacha,
from where—I know—we see what is called Limac (end of insertion) -from
that path—it is said—they returned, having heard—it is said—people say:
"Tutayquiri has already finished conquering everything".]
The narrator's familiarity or otherwise with the spatial location of the narrated events
influences the choice of evidential suffix in any given context. Thus space may be an
important dimension of the personal witness category at work in the use of evidential
suffixes (cf. Howard-Malverde 1990: 74-75).
The same device is used when a certain feature in the landscape (e.g., a stone) is
explained mythically, that is, the story that describes the persons involved is given in
the reportative past mode, whereas the stone that resulted from the petrification of a
character in the story is described as part of the here-and-now the narrator is living,
and therefore this account takes on the affirmative -mi and is told in the present tense:
chaysi chay cauellacaka

mana uyantapas payman tikrarichispa
kuchaman
"hinallam chinkasak chika milky runap kachkasapap
churinta wachaskaymanta" nispa
chikacharkan
(insertion:) maypim kananpas chay pachacamac uku kuchapi kananpas sutilla
iskay rumi runa hina tiyakun
(end of insertion)
chayman /
chaysi'
chay kanan tiyaskanpis chayaspalla
rumi tukurkan. / (chap. 2, fol. 65r)
[Then—it is said—that Cauellaca, without even turning her face back to him,
disappeared toward the sea, saying: "Like this I will flee because I have given
birth to the child of such an ugly man, full of pock-marks",
(insertion:) where—7 know—even now in that Pachacamac Sea, even now
visible two stones like people are sitting (end of insertion) in that direc-
tion.
Then—it is said when arriving at what now is said to be her seat, she turned
to stone.]
These insertions of the narrator's experience, his space and time, are also often marked
by the usage of the directional verbal modifier -mu-, which reflects linguistically the
speaker's relationship with the location he refers to in the verb and allows conclusions
as to where the narrator was at the time he was telling the story or account—or at least
where he was not! In terms of Howard-Malverde (1988: 128), here the narrator gives
a personalised presentation of the events. According to Lanser's (1981: 192) sugges-
156 INSIDE THE TEXT
tion of a spatial axis, the narrator is here close to a fixed spatial frame, leaving the
position of the narrator who looks at the narrated events from a quasi-outside,
panoramic overview (cf. Adelaar, in this volume).
Ethnographic descriptions
So far we have seen a relatively clear usage of -si and -mi in what I call mythical texts
with narrator's commentaries. Let us now turn to the type of text I call ethnographic
description or account because it describes certain ceremonies and rituals related to
the supernatural beings who act in the myths, normally narrated in the third person.
Typical examples of this "genre" are chapters 7, 9, and 10, which describe the
ceremony of cleaning an irrigation canal, the Auquisna ceremony for Pariacaca, and
the ceremony held for Chaupinamca.
In chapter 7, the narrator describes the celebration of the female Chuquisuso, who
had managed to get a good irrigation canal for her community by negotiating and
having sexual intercourse with Pariacaca (chap. 6). The ayllu that venerated
Chuquisuso used to go to her seat and take offerings to her. After finishing this five-day
ceremony and cleaning the irrigation canal, the participants come to the village, among
them a chosen woman who represents Chuquisuso, and there is a great feast. While
the narrator uses -mi to describe this, he switches once to -si, when he says that after
finishing they come to the village:
(6) chaymi na chayta muchaspari . . . /

chaypi tiyapayarkan /
chaysi na chayta puchukaspa
rarka pichaynintawanpas tukuyta na puchukaspax
chaymanta
takispa
runakuna hamurkan /
huk warmiktari
"kaymi chuquisuso" nispa
payta hina alli manchaspa
chawpipi pusamurkanku / (chap. 7, fol. 71 v)
[Now then I know -having adored that one . .. they stayed there.
Then—it is said—when they had already finished that, when they had already
finished everything—it is said—also the cleaning of the irrigation canal, after
that the people came singing.
And in the midst they guided here a woman whom they respected much, saying
"This is Chuquisuso".]
There is another -si inserted later on, as if the narrator or scribe did not participate in
this particular event or wanted to avoid being thought to know the narrated event
through his own experience:
chaymantam . . .
aswakta . . . kumuk karkan I . . . I
na kay rarkakta pichayta puchukamuspari
runakunakta anchapunis combidak karkan
Point of View and Evidentially 157
sarakta purutukta ima haykaktapas kuspa / (chap. 7, fol. 71v)

[After that—I know— ... [a woman] used to give chicha.
After having already finished here cleaning this irrigation canal, she used to treat
people certainly to a lot--it is said—giving them maize, beans and other things.]
Yet the use of "pusamurkanku" 'they guided here' in (6) indicates his personal in-
volvement, then he reverts to -mi. The same occurs in (7) "puchukamuspari"' after hav-
ing already finished here'. Once again, when the narrator refers to Chuquisuso inviting
people to eat, he uses -si as if he had never witnessed that. He then mentions as
eyewitness (-mi) that this ceremony is still being performed in his days (unmarked
tense):
(8) chaytam kanan

na rarkakta pichaspaka
musyaska hinatak
ruran muchan / (chap. 7, fol. 72r)
[After having already cleaned the irrigation canal, that now—7 know—they do
it as if confused, they pray.]
His final remark about how people tricked their parish priest in order to be able to go
on with their ancient customs is linguistically not quite consistent with the preceding
structure and diction—he uses Spanish porque:
(9) porque yallinrakmi paywan takin upyan machaskankama /

padrektari
"rarkaktam pichamuni padre * takikusak upyakusak" nispam llullachin /
(chap. 7, fol. 72r)
[Because moreover—7 know—they dance and drink with them until getting
drunk.
/ know that they deceive the Father saying: "I have cleaned the irrigation canal,
Father, I will dance and drink".]
Possibly the narrator/commentator added these remarks afterward or as a response to

a question.
The overall impression of the change from -mi to -si throughout the description is
that the narrator—although in his last remark he wants himself to be seen as a good
Christian—is still bound by ancient Andean beliefs and their practices. This is evident
in that he uses the affirmative -mi, which indicates that he is an eyewitness or has direct
knowledge, and also because, in the same affirmative mode, he describes how
Chuquisuso, who is a supernatural being, stems from a certain still existing ayllu:
kay [cupara] ayllumantam kanan huk yumay chauincho sutiyuk /

kay chauincho ayllum karkan chay chuquisuso niska / (chap. 7, fol. 71 v)
[From this ayllu [Cupara]—I know—there is now a lineage called Chauincho.
[From] this Chauincho ayllu—I know—was the one called Chuquisuso. ]
158 INSIDE THE TEXT
In looking at the other two chapters mentioned, a similar picture emerges. In

chapter 9, the narrator describes how ancient practices in honour of Pariacaca are still
being performed in different villages and provinces. Whereas the first groups are
presented with the affirmative -mi, later ones are described using -si, as if the narrator
had visited and seen the first ones, but not the Checa, Sunicancha, and Chaucaricma,
whose practices he describes as still existing in his time by using the unmarked tense.
These peoples seem to have lived in the narrator's world, and he knew them, either
by having seen them, or by having heard of them in his time. Toward the end of his
account, he moves backward in time, indicated by his employing "nawpa pacha" 'in
former times', combined with the past tense and -si to enumerate the peoples of an
earlier period and their veneration of Pariacaca (see table 9~ 1).
In chapter 10, after introducing Chaupinamca and her kin relationship, the narrator
describes her feast, which he seems to have witnessed because he uses -mi to describe
the different dances. Only once does he switch to -si, when he mentions a dance
performed with no clothing:
huktam kanan casayaco sutiyukta takik karkanku /

kay cassayacokta takiptinsi
chaupinamca ancha kusikuk karkan /
porque
kayta takispaka
llatansi /
wakillan wallparikunanta churaspallas
takik karkan
pinkaynintari huk wara utku pachallawan pakaykuspa / (chap. 10, fol. 77r)
[Now they used to dance another one—I know- called Casayaco.
It is said that when they danced this Cassayaco Chaupinamca used to be very
happy.
Because when they danced this one, it is said that they were naked.
They used to danccputting on only a few of their adornments—it is said—hiding
their shameful parts only with cotton breechclouts. ]
Table 9-1 The use of -mi and -si in the description of different groups who adore
Pariacaca (chap. 9).
Comment and tense Groups
-mi and past tense Huarochiri province
Chaclla province
Mama province
Colli from Checa llakta
-mi and unmarked tense Surco runakuna
Huayllas runakuna
Concha
no comment suffix and unmarked tense Sunicancha
•si and unmarked tense Checa
Chaucaricma; also the ones from Santa Ana and San
Juan
-si and past tense; nawpa pacha a list of different Yunca groups
Might the influence of Christian education, or his knowledge of the church's morals,
have moved the narrator to deny personal knowledge of such a "barbarian" custom
(viz. his use of -si)? The usage of "pinkaynin", 'their shameful parts' also provides a
hint of this. The narrator of the account may be identical with the one who makes this
comment, but it is also possible that the comment was inserted into the text by the
redactor who takes on the role of an "extrafictional voice" (cf. Lanser 1981: 124).
So far we have seen that, on the one hand, the narrator automatically uses the
appropriate evidential suffix to testify his personal knowledge or ignorance of a
phenomenon; on the other hand, he employs this linguistic means in order to deny or
confirm, his knowledge of certain events he relates, thereby conveying some informa-
tion about himself and about how he wants to be seen by the addressee. Here, then,
we notice a combination of the evidential aspect (how the narrator obtained his
knowledge) with the validational aspect (the narrator's commitment with respect to
the truth of the narrated account).
Special idiolectal or dialectal usages

While the examples I have presented here indicate a rather consistent and well-known
usage of the Quechua evidential suffixes, there are also instances of unusual usages
that are to be explained either as idiolectal or as a Huarochiri dialect feature which
differs somewhat from the known application described.
In chapter 16 the narrator describes the expulsion of Huallallo Caruincho by
Pariacaca to the lowlands. While he uses the reportative -si to this effect (and -mi is
used to refer to other chapters of the manuscript), he also employs -mi when enumer-
ating Pariacaca's brothers, although they definitely belong to the mythical sphere
normally presented in the reportative mode:
(12) kaykunap sutinmi nawpakninmanta pariacaca, /

chaymantam churapa, /
chaymantam puncho, /
chaymantam, pariacarco / (chap. 16, fol. 8lr)
[The names of these—I know—were, from their first one, Pariacaca.
And after that—I know—was Churapa.
And after that—I know—was Puncho.
And after that—I know—was Pariacarco.]
Might the narrator actually be referring here to mountains called Churapa, Puncho,
and so on, which would explain the use of the eyewitness evidential -mi?16
An interesting opposition of-si and -mi can be found in chapter 1, where the ancient
time with its huacas is described in the reportative mode; but the changes caused by
Pariacaca, who expelled theyunca (coastal population) and their god to the lowlands,
as well as the more recent god Cuniraya Viracocha, are described using the affirmative
suffix -mi. These gods' closeness to the narrator is demonstrated by the directional
suffix -mu- in connection with Pariacaca:
160 INSIDE THE TEXT
(13) kipanpi na may pacham huk wakatak pariacaca sutiyuk rikurimurkan/ (chap. 1,
fol. 64r)
[/ know that afterwards already at one time another 'huaca' called Pariacaca
appeared here.]
We also find the demonstrative pronoun kay 'this', and the time adverbials kanan
'now', and nahka 'already', in connection with Cuniraya:
(14) chaymantam kanan huk wakatak cuniraya sutiyuk karkan /

(insertion:) kaytam mana allichu yachanchik,
pariacacamantapas ichapas nawpaknin karkan o kipanpas /
ichaka kay cunirayap kaskanrakmi nahka viracochap kaskanman tinkun / (chap.
1, fol. 64r)
[And after that—I. know—now there was also another huaca called Cuniraya—
(insertion:) we do not know this one well- -Iknow—-whether he was earlier
than Pariacaca or after him (end of insertion) but / know that this
Cuniraya's life was already related to Viracocha's life. |
This, and the fact that the narrator quotes prayers to Cuniraya, which have a binary
semantic structure (or "semantic coupling" as in Guaman Poma, Molina, and
Pachacuti Yamqui) typical of the genre,17 indicates, in my opinion, that he is very
familiar with these gods; as "historical" beings, they may have to be presented in the
affirmed knowledge mode, whereas ancient, mythical, supernatural beings, having
already disappeared before the last pre-colonial gods, are connected with the hearsay
evidential suffix -si. Such a hypothesis would also help explain the use of -mi in (12)
when referring to Pariacaca's brothers and in (10) when speaking of Chuquisuso's
descent. We may want to re-interpret the understanding of "eyewitness" to include
religious phenomena that for the speakers seemed to have formed part of experienced
life. Another instance of a special usage of grammatical devices is reflected in what
Adelaar (1994b: 118) has termed the "complement of divine manifestation", which
means that Huarochiri Quechua, or at least the variety used in this manuscript, made
particular use of certain grammatical functions.
So far ! have discussed the different types of text I consider to be distinct
"genres" with respect to their narrator and his sources of information. Myth-like,
coherent stories are usually told in the reportative mode—that is, with the evidential
suffix -si and the simple past -rka-. Ethnographic accounts and descriptions tend
to be related with the evidential (eye-)witness suffix -mi and either the simple past
tense or unmarked tense, sometimes with the use of reportative -si in order to
document the non-witnessing of an event and/or to deny having witnessed it. It
would thus appear that when writing down myth-type narratives, the narrator(s)
kept to the usage of the evidentials as practised in the oral narration of the myth.
By contrast, with regard to ethnographic or testimonial accounts, the new medium
of writing expresses and fixes personal experience versus hearsay perspective in a
seemingly inconsistent way. This, however, may be explained by further studies
of the texts in the light of the hypothesis that at least some of the manuscript
chapters represent several voices, possibly joined together by a redactor. The
Point of View and Evidentially 161
written version of these accounts would then be revealing a context-performance •

determined usage that is not yet adapted to a consistent employment of these
suffixes in writing, an observation similar to that made by Adelaar (in this volume)
with regard to the status of the suffix -mu-.
The expression of conjecture

Between the two poles of experienced affirmation and non-experienced secondary
information, the Quechua language has another evidential area: of doubt, uncertainty,
or conjecture, which in the Quechua II18 dialects is expressed by means of the suffix
-cha and its allomorphic variants. In the Huarochiri manuscript, -cha is found in
conjectures in the reported speech of the characters and in conjectures made by the
narrator. Especially in reported speech, -cha is often found together with the potential
mode or future tense, which implies insecurity about the possible realisation of what
is said. It can also indicate a warning or advice when it occurs inpakta-ch 'careful'!,
'rather .. . '. The narrator uses it when he expresses suppositions about beliefs and
religious ceremonies of his time, about which he is not completely certain, and when
he does not know exactly the date or the duration of a particular ceremony.
A comment made by the narrator with respect to the Christianisation of the ethnic
groups whose ceremonies are described in chapter 9, refers to old beliefs that are in
the process of disappearing:
ichaka kananka na kunkan kay pisi watallarak

kay doctor francisco de auila alli kunakiyuk yachachikiyuk kaspa I /
chaypas manatakcha sunkukamaka ininmanchu /
natak huk padreyuk kaspaka
hinamantakcha kutinman I /
wakin runakunaka
christiano tukuspapas
manchaspallam.
"paktach. padrepas pipas yachawanman mana alli kaskayta" nispallam
christiano: tukun / (chap. 9, fol. 75r-v)
[But now they already forget [the old customs] within just these few years,
because they have this doctor Francisco de Auila, a good person in charge and
teacher.
However, they do not really seem to believe with their hearts.
If they had again a different Father, they might probably return to these ways [of
old customs].
Some people, after becoming Christians, only because they are afraid—I
know—they become Christians, just saying—/ know: "Careful! Maybe the
Father or somebody might know of my bad life".]
Here the narrator/commentator clearly shows a great deal of scepticism with respect
to the situation of Christianisation, and it becomes evident that only fear moves people
to convert to Christianity. Whilst this shows the narrator's/commentator's detachment
from the narrated events, a personal testimony is quoted in chapters 20 and 21. Here
an artful usage of the evidential suffixes -cha and -mi is found in don Cristobal's
162 INSIDE THE TEXT
meeting with the old huaca Llocllayhuancupa from whom he wishes to distance
himself. But as he is not completely sure whether Llocllayhuancupa has actually lost
his power, he formulates his defiance using the suffix of conjecture -cha:
(16) kayta puchukaptinsi

na kasilla irnapas kaptin
don xpistobalka rimayta kallarirkan
nispa I
(don Cristobal speaks:)
"yaw llucllayhuancupa /
kamtam ari 'runakamak pachakuyuchik' nispa
nisunki /
'paytakmi ima hayka rurak' nispam ari
tukuy runakunapas manchasunki /
imapakmi kanan kayachimuwarkanki /
nukaka ninim
'jesu xpisto diospak churin kaytakcha chikan dios * paypak simintatakcha
winaypas yupaychasak' nispam /
nini /
kayri /
pantanichu /
kam kanan willaway
'chayka manam dioschu * nukam ima hayka rurak kani' nispa
chay pacha kamta manchankaypak / "
nispa /
(don Cristobal finishes speaking)
niptinsi
chay supayka upayarkan / (chap. 21, fol. 88r-v)
\It is said that when this had finished and there was already some stillness, don
Cristobal began to speak, saying:
"Listen, Llucllayhuancupa; I know that they call you 'the one who has made man,
the one who makes the earth tremble', / know that saying 'he is certainly the
one who made everything' all the people fear you.
Why have you had me call here—--7 am here ?
/ say—-1know it: 'Jesus Christ, the son of God, this one is probably the true God;
and his word I will probably always respect'. This is it! If 1 am wrong, then
tell me: 'That one is not God-—I know it. I know that I am the one who makes
everything', so that then I may fear you". It is said that when he | don Cristobal]
had said it, that supay was silent.]
The distribution of the evidential suffixes

Authors on Quechua grammar always suppose that the evidentials form one group of
suffixes, which would also imply that they are used in the same morphological and
syntactic structures. By studying the distribution of these suffixes, we learn something
about the syntactic organisation in which these evidentials are embedded, and their
distribution may be a clue as to different meanings or functions (cf. Jones 1992 on a
Papuan language). My brief presentation of the distribution should be seen as an initial

step in this direction.
Despite a long tradition of Quechua grammar writing, detailed studies on a level
beyond morphology are still scarce and mostly refer either to clause or to discourse
analysis,19 so that my observations are an incipient effort not only at understanding
the evidentials in the Huarochiri manuscript but also in the general context of syntactic
studies of the Quechua language.
In the manuscript, -si is used much more frequently than -mi, a situation which is
easily explicable because of the story-character most of the texts have. But as can also
be expected, in proportional terms -si and -mi are distributed equally. For the analysis
of the distribution of the evidential suffixes -mi and -si, I have chosen to divide the
sentence into units which, in my analysis of the discourse suffixes, have proved to be
useful and meaningful sub-divisions:
author-action (the subject itself and the conjugated verb that includes the sub-
ject, also in its nominalised forms)
patient (the direct object and the nominal expressions with the other relational
suffixes that mark what we traditionally call "cases"20—nominal forms as well
as nominalised forms)
"circumstances" (above all time and space indicators)21
negations
In quantitative terras, both suffixes are distributed relatively equally with respect to
the complex actor-action/patient, on the one hand, and circumstances and negations, on the
other. However, there is a clear tendency for -mi to be used above all with nominal
forms (subject and object), whereas -si is mainly combined with nominalised verbs (es-
pecially -spa and -pti-) when these incorporate the subject of the subordinate clause.22
This is only a general impression because my analysis does not take into account
the difference in the usage of the evidentials by the narrator himself, as opposed to the
usage in reported speech made by the characters. Syntactically, both -mi and -si tend
to be used with the initial elements of a clause, a result confirmed by Rick Floyd (1994)
in his study of evidential suffixes in Wanka Quechua.
The suffix -cha, which appears in the manuscript in different allomorphic forms
({-cha}, {-ch}, {-ch.}, {-h}, {-h.}, {-he}), is used much less frequently than the other
evidential suffixes. As has been mentioned here, it is mainly employed to mark a
conjecture that the narrator/speaker expresses about the characters or about the
narrated account, which is why it is generally found in the ethnographic descriptions
and in reported speech. The suffix -cha is above all combined with nominal expres-
sions, especially interrogative pronouns. It often occurs in the same sentence together
with the future tense or potential mode, and sometimes with the verbs of supposition
unancha- 'to consider, believe, understand as' andyuya- 'to think'. When comparing
-cha with the other evidential suffixes, it seems that the usage of -cha is more easily
captured and described.
A further area of study is the position of the evidentials in relation to the topic
marker -ka and the other two sentence suffixes -tak and -ri (Dedenbach-Salazar Saenz
164 INSIDE THE TEXT
1994). For the time being, after a global survey, there does not seem to be a fixed
distribution rule, but hopefully a syntactic analysis will show certain patterns which
may also allow the determination of different "styles" and different narrators.
Conclusion and prospects

On the whole, the usage of the evidential suffixes in the Huarochiri texts is very similar
to their modern usage in Quechua IIB.23 The reportative suffix -si, often together with
the simple past tense, characterises a genre of story-like texts removed from the
narrator in time and frequently situated in a non-defined space; the characters often
belong to the mythical sphere. The suffix -si also serves to mark events that happened
in a not very remote past connected with historical persons, or events not witnessed
by the narrator.
The suffix -mi is typical of a descriptive genre of rituals and ceremonies, as well
as personal accounts, and is most often used to show that the narrator has been
eyewitness to an event or, for example, knows places from his own experience; this
affirmative suffix is often accompanied by the habitual past tense or the unmarked
tense and also by the verb modifying suffix -mu-. Moreover, -mi is used in connection
with certain supernatural beings, probably marking their integration into human lived
experience.
Conjecture or doubt as to the event referred to is marked with the suffix -cha,
which shows that the narrator/speaker is not certain about his information. In
complex narrative situations -si, -mi, and -cha are used in a deliberate manner that
reflects both the actual evidential situation of the narrator and his attitude toward
the narrated account—for example, whether he is or feels involved—and his
attitude toward the addressee, that is, whether he wants to convey his involvement
in or detachment from the narrated events. Thereby these suffixes also take on a
validational aspect.
When reading the Huarochiri texts carefully and with one's mind on the standpoint
adopted by the narrator, there is more than one voice to be heard; frequently the
narrator of a story seems to be different from a commentator who puts the story into
the overall framework of the collection, and may be different again from a redactor
who may have been responsible for putting the texts together and writing them down.
Thus parts of the texts may be understood as testimonies of an involved narrator who
gives an account of his own culture and environment (in relating myths as well as
personal experiences), other parts as ethnographic descriptions by a more detached
narrator (in describing ceremonies and rituals), and still others as comments about the
narrated account by a commentator who may or may not be identical with the detached
narrator. All this is possibly woven together either by one of the narrators or by a
redactor.
The present study has given us some suggestions as to how to differentiate
narrators' attitudes, genres, and situational contexts; at the same time, it calls for new
research, for example, in addressing the question of the voices and styles of different
narrators. It is to be hoped that this kind of research will also be taken up with respect
to modern Quechua tales so as to make more general statements possible, and perhaps
such a comparative view could also shed new light on the Huarochiri texts themselves.
Notes
1. The meaning of the term edit and its derivatives normally refers to the preparation of a
work for publication (e.g., Taylor's or Salomon and Urioste's editions of the Huarochiri
traditions). The processes mentioned here are better described with the term redact and its
derivatives, emphasising the meaning of reduction to literary form, revision, and rearrangement
(cf. Oxford English Dictionary 1989). The manuscript was found among Avila's papers, but it
is doubtful that he ever had the intention of publishing it—unlike his Spanish Tratado, which
presents his own redaction (paraphrase and translation) of these texts, written in a very clear
handwriting and provided with a proper title page, which may well have been intended for
publication.
Trimborn (1967: 10) says that the handwriting of the Quechua manuscript and the Spanish
Tratado is Avila's. However, a graphological analysis of the manuscript, texts in comparison
with texts attributed to Avila is still outstanding. This might help us in finding out how far he
was personally involved in the (re-)production of these traditions.
2. Adelaar has studied several aspects of the Quechua structure of the Huarochiri manu-
script: his analysis of the demonstrative pronouns and the directional suffix -mu- (Adelaar, in
this volume), his study of reported speech (Adelaar 1990), as well as that of abstract and general
concepts (Adelaar 1994a), suggest the complexity of the manuscript's discourse structure. On
the formal linguistic level there is Urioste's morphological analysis (1973), and Salomon (1991)
and I (Dedenbach-Salazar Saenz 1994) touch on further textual aspects of the manuscript.
3. In Andean studies specifically, by comparison, Adorno (1989) has brought point-of-
view analysis to bear upon the chronicle of Guaman Poma de Ayala, in terms of the construction
of his ideological position through both written text and visual images.
4. Salomon (1991: 32-33), in his introductory essay to the new translation of the
Huarochiri manuscript—which was published after I had written this paper—gives a brief
discussion of the evidential suffixes, which shows similarities with my present analysis here.
5. With respect to the dialectal status of the Quechua used in the manuscript, Adelaar
(1994d) and I (Dedenbach 1982) observed independently the similarity with Northern Peruvian
Amazonian Quechua dialects. For example, the orthography of the manuscript suggests the
non-existence of the post-velar/q/; this is reflected in my modernised transcription in which 1
use [k] exclusively for all the stops represented in the document as {qu, co, Cu}. It can thus be
assumed that the language of the Huarochiri manuscript belongs to the Quechua IIB dialect
group (cf. footnote 18); however, this does not exclude the possibility that it was used as a
"general language" incorporating basic and simplified features of Quechua IIB (cf. Adelaar
1994d: 149-150). With reference to the past tense forms, it is also possible that -sqa- as a verbal
form ("tense") only developed toward the end of the seventeenth century (as I have argued in
Dedenbach-Salazar Saenz 1993).
Taylor (1987a: 21) finds Aru (Jaqi) influence at the lexical and morphological levels; here
I should only like to observe that the texts obviously lack this influence at the syntactic level
which I have been analysing. Perhaps this indicates that an Aru substratum had already been
integrated into the Quechua vocabulary of this area without influencing the syntactic structure,
or that the author of the manuscript was a Quechua speaker without local Aru influence in his
syntactic structure. The question of dialectal classification is further complicated by the fact that
we do not know which changes can be attributed to the narrator(s), commentator(s), or
redactor(s).
166 INSIDE THE TEXT
6. I use a modernised transcription (see also note 5) based on a photocopy of the original
manuscript, indicating chapters (capitulos) and folios, in which all names are spelled as in the
manuscript; the different writings of the suffix -cha are kept in their original spelling because
they represent sounds that arc not easily reconstructable (Taylor represents them by c). All
sibilants, mostly written as [s], and some as {ss}, { c } , {x}, and {z} in the manuscript, are
transcribed as {s}.
To make reading easier, the subordinate clauses are indented once when the subject of the
main clause and the subordinate clause is the same; when there are different subjects, the
subordinate clause is indented twice. The oblique stroke marks the end of a sentence; quotation
marks are used to indicate reported speech, and within this, asterisks delimit sentences and
clauses. All other punctuation marks (including the vertical bar) are kept as in the original.
For those who do not know Quechua, I translate the manuscript passages into English. The
reader will notice the frequent usage of 'then' which reflects Quechua chaysi, a connective
whose translation is also found in many modern Andean tales, and even Hillman's (1994)
presentation of a narrative in Spanish shows many occurrences of entonces. My translation
should be understood as a help for the reader, and at the same time tries to capture some of the
diction and structure of the Quechua of Huarochiri, which is why the sentences are based very
closely on the original text rather than stylistically elaborated and adapted to current English.
As this essay examines the role of the evidential suffixes, these have to be translated into English,
which—like other Indo-European languages—uses other devices such as lexical ones in order
to express evidentiality (Anderson 1986). 1 translate the affirmative mode as 'I know', the
reportative as 'it is said', and the conjecture as 'probably', 'maybe'. These expressions of
comment and the word/phrase they are connected with in Quechua appear in italic type.
7. When comparing this construction with that of the preceding sentence, it might also be
interpreted as an enumeration: 'the llama, the mountains', parallel with 'this mortar, the grinding
stones'.
8. This is a possessive construction with -paq.
9. 1 use the term mode to describe not just the function of the evidential suffix itself but
the whole affirmative or reportative context this suffix creates.
10. It is not easy to grasp the meaning of religious concepts which—since the Spanish
conquest—have certainly undergone transformation. "Huaca" denotes an object, feature, or
place that is special and different; untranslatable as it is, it implies the idea of something "sacred".
11. As with "huaca", it is difficult to translate "supay", interpreted by the Spaniards as
'demon' or 'devil'; in Andean belief it seems to have been related to the "world below", and
may have referred to an ambiguous numen. It is interesting to observe that in chap. 20 the word
"huaca" only occurs in the first part where its history is told; in the following part, that is don
Cristobal's report, only "supay" is used. Also in the next chapter, "huaca" is only used twice.
Maybe this shows that it is really don Cristobal's discourse which is represented here, and he
has already adapted the Christian-influenced, now negatively connotated "supay" in order to
describe Andean native gods (although he still shows traces of the original Andean use when
qualifying supay as "mana alli"—that is, there can be 'good', "alli", and 'bad', "mana alli",
supay) (cf. Taylor 1980; Crickmay 1994).
12. See also the discussion of the "ethnographic descriptions" for the interference of the
commentator, and note 15.
13. There is a very thin final { - s } detectable in the manuscript. (The corrections made in
the texts might also shed light on the process of redaction).
14. Perroud and Chouvenc: musya- "saber, conjeturar, adivinar, pensar, ver con-
fusamente" 'to know, conjecture, prophesy, think, see confusedly'; Taylor: "confundidos [por
el demonio]" 'confused [by the demon]'; Trimborn: "mit Vorbedacht" 'intentionally'. Taylor's
translation seems to me the most acceptable one, although because of the broad spectrum of
meanings of musya- it might also imply divination or perhaps a state of trance during the
ceremony.
15. Another possible interpretation of the shift from -mi to -si is the assumption that the
narrator of the beginning and the end of the chapter is not identical with the one who narrates
the ceremony itself. However, as this more complex narrative situation, which might also be
related to the overall structure of the manuscript texts, is only emerging in my current
analyses—after having finished this essay—it will have to be kept in mind, but cannot be dealt
with in detail here.
16. So far, I have not been able to locate these names geographically (e.g., in maps of Lima
and its surroundings from 1750, inMapasy Pianos, Peruy Chile, 33 and 33 bis, Archive General
de Indias, Sevilla; in a map from 1788 which is in the British Library, London; in German
Stiglich, Diccionario Geografico del Peril, Lima 1922; in the 1:100,000 maps of the province
of Huarochiri by the Instituto Geografico Nacional, Lima).
17. For studies of the poetic structure of colonial Quechua, see Husson 1984 and 1985,
and Mannheim 1986a, and for a challenging view, Itier 1993.
18. For an overview of the dialectal subdivision of the Quechua language (according to
Torero and Parker), see Dedenbach-Salazar Saenz et al. (1994: XIII): Quechua I comprises the
Central Peruvian dialects; Quechua II, the Southern Peruvian, Bolivian, Argentinian, and
Chilean, as well as the Northern Peruvian and Ecuadorian dialects, with IIB including the
dialects spoken in the Peruvian departments of Amazonas, San Martin, Loreto, Ucayali, and
Madre de Dios; in the sierra and oriente provinces of Ecuador; in Putumayo and Narino in
Colombia; and in Acre in Brazil. The morphological differences between Quechua I and II are
considerable; on the lexical level, there is a marked distinction between Northern Quechua and
Central Peruvian Quechua on the one hand and the Southern dialects on the other.
19. On clauses, see Carlson de Coombs 1975; Cole et al. 1982; Costa 1972; Weber 1976.
On discourse, see Levinsohn 1976 among others and Wolck 1979.
20. Except for the spatial and temporal expressions constructed with -pi.
21. In quantitative terms, with both suffixes (-mi and -si) the temporal indicators dominate,
the reason for this being that I have included also the ones that mark the course of the narrative,
as, for example, chaysi and chaymantam. In a further analysis it will be necessary to differentiate
within this group of temporal markers the ones that refer to narrated time—thus establishing
temporal relationships within the narrated events—and the ones that connect the sentences of
the narrative itself in order to provide a chronological development of the narrative.
This group of "circumstances" also includes hina-, which when it occurs with evidential
suffixes mainly has the function of connecting sentences or clauses, combined (a) with the
nominalising suffix -spa, (b) with the connective -tak.
Another word included in the category "circumstances" is imana-, which occurs virtually
only with the affirmative suffix -mi and frequently serves to introduce an insertion of personal
knowledge about a phenomenon of the narrator's world; it can be translated as "as" or "like",
and it tends to have an explanatory function. It may be convenient to consider it as a monolexeme
in which -mi has become part of the word itself.
22. -si is also often connected with the deictic pronoun chay forming chaysi, and as such
connects sentences.
23. Cf. notes 5 and 18.
IV
ORALITY AND LITERACIES AS

REFLEXES OF COLONISATION
AND RESISTANCE
10
MARTIN LIENHARD
Writing from Within

Indigenous Epistolary Practices
in the Colonial Period
B Before the European conquest, the autochthonous societies of the continent to the
west of Europe carried out their basic semiotic practices within a communicative
framework which—but only for the sake of contrast with those systems which
privilege (even fetisruse) writing—I shall call "oral". The colonial and modern groups
made up of their descendants (direct or indirect), despite having reproduced a similar
system for their own internal purposes, not only interiorised the writing fetishism (cf.
Platt, in this volume), but succeeded in mastering writing for the purpose of dialogue
with the "other".
Seemingly irreplaceable, the concept of "orality" in fact raises a whole series of
problems. By virtue of its etymology, the term reduces the whole diversity of semiotic
practices in an "oral"'' society to verbal expression, yet we do not know whether the
spoken word fulfils the same function in "orality" as does the written word in the
system based on written communication and its fetishisation. Amerindian "orality"
always was—and still is—a complex semiotic system relying not merely on verbal
("oral") communication, but also on the most diverse media, including plastic, graphic,
choreographic, gestural, musical, and rhythmical (cf. Rappaport, in this volume). Far
from being ignorant of "graphic" communication, several Amerindian "oral" systems
even countenanced the use of some type of "writing" for the notation of words or of
whole speeches; consider, for example, the Mesoamerican glyphs or the Andean quipu
system, tactile and visual at the same time. When I speak of Amerindian "orality",
therefore, I shall be referring to an eminently multimedia system.
171
172 ORALITY AND LITERACIES/COLONISATION AND RESISTANCE
Before the European conquest, the setting for this multimedia Amerindian dis-
course was a part of the world characterised by relative cultural and linguistic
homogeneity. In cultural, though not necessarily political, terms, communication took
place between "equals". The arrival of the Europeans and, more than anything else,
the fact that they took political control of the territory created the need for a new,
intercultural type of communication between the groups who were marginalised by
the conquest and the new masters. In this context, the defeated (basing their actions
principally on the linguistic resources of their powerful interlocutors) produced a
mode of discourse capable of serving them in their dealings, one that I shall call
"indigenous discourse addressed to outsiders" (Lienhard 1992). Although traditional
modes of expression, always bearing the stamp of "equality", continued to apply
within the Amerindian communities on the inside, this new indigenous style came to
govern communication, in both oral and written form, between the "conquered" and
the "conquerors" and their descendants (Lienhard 1990).
This "indigenous discourse addressed to outsiders" manifests itself in two basic
varieties: firstly, the testimonies of natives transcribed—as a general rule—by a
representative of the opposing party, and, secondly, the letters and manifestos which
the conquered themselves and their descendents addressed, from the very beginning
of the colonial period, to their European or criollo interlocutors. In the following pages,
I shall attempt to identify a particular type of discourse—the "indigenous letter
addressed to outsiders"—and the communicative practice corresponding to it.1 The
texts themselves, the instruments of such a practice, will mostly be of interest in that
they constitute the recognisable "traces" of a specific process of communication that
is always highly politicised. Drafted in the context of latent or overt ethnic/social
conflict, these indigenous letters manifest the fact, not only on the thematic level but
also in the way in which they are written. The "indigenous letter addressed to
outsiders" took on a role as part of a communicative system the basis of which was
fixed in the sixteenth century: an asymmetrical system, dominated by the figure of the
Spanish king. As I shall attempt to show, this system—whose "rules" were broken in
all sorts of minor ways from the very start—steadily lost its validity until, during the
indigenous insurrections of the eighteenth and nineteenth centuries, a radically new
approach was formulated.
The basic process of communication

At the start of the colonial period, as borne out by currently accessible sources (see
Lienhard 1992), indigenous epistolary practices took place in the context of a privi-
leged relationship with the king or his representatives. Basing their terms on the feudal
ones of the time, the authors of the letters—indigenous communities or individuals—
regarded themselves as "vassals" of the king, their "lord", with the duties and rights
that this entailed. In terms of the social hierarchy, then, the message was addressed
from "below" to "above". But the native vassals were not only below the king: they
were also, as colonial "vassals", far removed from the centre, at the "periphery" of
the world. This meant that their messages to the king also had to cross the cultural and
political chasm that separated the periphery from the centre. From the viewpoint of
Writing from Within 173
the Indians, it was a case of movement from "within" (the indigenous community) to
"without" (the world of the outsiders).
This asymmetry imposed two rules on the emergent discourse. Firstly, the situation
of inferiority within the hierarchy dictated a rhetoric of respect; secondly, the move-
ment from periphery to centre (or from "inside" to "outside") demanded the use of
modes of expression with which the addressee was familiar. If the first of these rules
was never openly infringed, the degree of enthusiasm brought to bear on the rhetorical
simulation of feelings of respect enables us to gauge, up to a point, how genuine the
natives' sense of affiliation to the colonial order was. As regards the second rule, any
breach of it, whether minor or radical, gives us grounds for supposing a certain spirit
of cultural resistance. Thus, the very form of the letters, sometimes more than their
content, may be an indicator of the tension on the ethnic/social front at a particular
time and place.
Judging simply by the body of evidence referred to, the authors of indigenous
letters in the colonial period were recruited almost exclusively from the indigenous
elite: native nobility, traditional authorities, local councillors. Only the members of
this elite (and not all of them) actually had the opportunity to be introduced to written
culture or, failing that, access to someone who could do them the favor of drawing up
a letter while they dictated it. They were also the only ones with any degree of power
or experience of dealing with it. However—and this is a feature that characterises the
majority of the indigenous letters—the members of this elite were wont to speak, even
when putting forward matters of concern to themselves as individuals or to their
particular social group, in the name of the indigenous community as a whole.
Born out of a context of negotiation, indigenous letter-writing is therefore, in a
broad sense, political. Its raison d'etre is nearly always a desire to protest or to assert
one's rights or claims, and the most diverse forms are deployed to this end. The
integration of other discourses—historiographic, ethnographic, poetic, etc.—is sub-
ordinated to this primary aim. Any expression of feelings, when such appear, has
nothing subjective about it and serves only to reinforce a political argument.
As I said earlier, indigenous letter-writing of the colonial period was characterised
by a privileged relationship with the king (and his representatives). The ever deeper
cracks that appeared in the whole social structure around the end of this period
complicated this relationship. In the eyes of the indians, the legitimacy of the local
representatives of the king in fact continually lost ground, and this fact was manifested,
as we shall see, in a profound modification of the process of communication already
outlined.
Once independence was won (though not by the indians), the epistolary relations
between them and their criollo interlocutors ceased to be carried out within a stable
framework. From the beginning, the rulers of the republics lacked—and not only in
the eyes of the Indians—the "divine" legitimacy of the king. In this qualitatively
different situation, the politics of the indigenous groups lost its relative coherence
(which had never amounted to uniformity) and began, according to the new deals, to
tend toward division, dispersion, and ambiguity when confronted by criollo political
forces. In this context, indian-led letter-writing definitively abandoned its vertical
relationship with the supreme authority and began to move in all sorts of horizontal
directions.
The defence of "caste" interests

From the middle of the sixteenth century, certain groups of the Mesoamerican nobility,
especially in the central area of Mexico (Mexico-Tlatelolco, Tetzcoco and Tlacupan),
found themselves involved in intense letter-writing activity. These Mesoamerican
groups were at that time the only sector of the Amerindian population with relatively
easy access to an education equivalent to that of the literate classes in Rennaissance
Europe, thanks above all to the work of the Franciscan missionaries. For this reason,
their letters (written in Spanish, Nahuatl, or Latin) are models—at least on the sur-
face—as far as respect for the conventions of letter-writing is concerned. An illustra-
tive case in this regard is the letter in Latin addressed by don Pablo Nazareo and his
wife, a niece of Motecuhzoma, to the king of Spain on 17th March 1566 (Paso 1939-
1942, t. X, 89-129, translated into Spanish by Agustin Millares Carli, and reproduced
in abridged form in Lienhard 1992:44-53). In it, the "rhetoric of respect", or more pre-
cisely the praise of the king, is highly obsequious to the point of the obscene:
It being, o most unconquered king, a distinctive property of kings and princes,

by consequence of their divine power, that they enlighten all other mortals, just
as the Sun [the Latin original has Phoebus] gives light to the whole world, whose
splendours in this region of New Spain are spread by the more than sublime
greatness of your majesty, so that if there were any light in us, far from shining
it would be dimmed, and our feeble spirits, dazzled to their uttermost depths by
the majestic brilliance of so much light, would be unable to bear it. (Lienhard
1992:44-53)
What did don Pablo and his wife hope to gain from their letter? On the one hand,
as they proceed to point out in between the quotations from Ovid and the Bible, a
substantial income; on the other, the recognition of their "natural dominions" and
"patrimony"—lands to which they had historical title. Their argument is based in the
first instance on the right of succession, but the signatories also place emphasis on the
assistance that members of their families had given to the Spanish during the conquest,
along with the "disinterested" contribution of don Pablo himself to the cause of
Christianity and of the Spanish. Don Pablo had been, amongst other things, rector of
the College of the Holy Cross in Tlatelolco, a Franciscan institution which with its
humanist education played a central role in the assimilation of the indigenous and
mestizo elite of the central area of Mexico.
The recognition of seigneurial rights and titles, and the concession of a more
substantial income by the Spanish crown, are amongst the claims most often voiced
in the letters drafted in Mesoamerica by the descendents of the pre-Hispanic nobility.
Even if their rhetoric is tailored—not always with the degree of excess of the letter
from don Pablo and his wife—to the expectations of their interlocutor (the king), they
tend also to incorporate certain modes of expression of pre-Hispanic origin. Don Pablo
himself, abruptly interrupting his praises and his Ciceronian circumlocutions, intro-
duces—in Latin—at least two discursive genres of recognisably Aztec origin: the
"historical-genealogical-title" and the "frontier treaty". In recreating a traditional
discourse, these sequences constitute, in the context of the letter, a sort of foreign body.
It is highly likely, in fact, that these insertions are basically the result of the "transcrip-
tion" of some documents in pre-Hispanic style. The resulting cultural hybridisation
that characterises don Pablo's letter is found again, in the most varied forms, in many
indigenous letters,
As well as simply defending the privileges of the nobility, other Mesoamerican
letters, based on a similar historical/genealogical argument, defend the interests of
local indigenous autonomy. In such letters, the list of signatories sometimes includes,
as well as the traditional lord or lords, other "caciques and principals", local indige-
nous government representatives of the "notables" and the "natives", and the scribe.
A sort of leitmotiv punctuates the majority of the letters emanating from the
indigenous noble classes in the sixteenth and seventeenth centuries: the loss of
economic and political authority which the traditional lords had suffered from the
conquest onward. Expressed with varying degrees of pathos, this topos usually
contains an attempt to demonstrate that the Indians are utterly and completely
incapable of paying the tributes or taxes demanded of them. In its letter of October
1554, the indigenous government of Cholula attempts to anticipate, with just such an
argument, the expected introduction of tithes—a tax that "penalises" sowing and the
growing of "things from Castile" (Paso 1939-1942, t. VII, 269-270, reproduced in
Lienhard 1992: 36-38):
They have told us that we are to give tithes, and we all feel very great pain about
this, and the macehuales (peasants) become despondent and say that they shall
flee.... We beseech Your Majesty for the love of God not to impose tithes upon
us, because we are very poor and many ills would befall us. (Lienhard 1992: 38)
The plea is formulated with all the marks of respect, and the vindication is
attenuated by the—albeit rhetorical—expression of Christian "feelings". The only
discordant note, "the macehuales become despondent and say that they shall flee", is
almost drowned out by angelic harmonies. In stirring up the spectre of rebellion and
desertion, this veiled threat constitutes, nonetheless, the most forceful argument
contained in the letter.
Complaints and disputes from

the absolute periphery
If the letters written by the literate Mesomericans of the upper classes bear the stamp
of a shared (and highly cultured) set of writing norms, others, emanating from the
"periphery" of the colonial territory, or drafted by persons lower in the hierarchy or
who had had a less mainstream education, evince a much more spontaneous approach
to writing. Unlike those already considered, they are wont to refer (sometimes in rather
excessive detail) to very concrete occurrences: episodes of ordinary everyday oppres-
sion suffered by the indians. Very remote from the archaising rhetoric of the more
"aristocratic" letters, the style of this type of text comes closer, instead, to that of oral
narration. The fragment that follows is from a letter (Anderson et al. 1976: 226-229,
reproduced in Lienhard 1992: 77-80) written to the king in Nahuatl by someone
referred to as the mayor of Jalostotitlan (Jalisco); the (contemporaneous) translation

into Spanish was made by the Royal Audience of Mexico:
Once again a sacristan lad, eight years old he was, in a big way (the curate)
whipped him, in great measure did he flay his flesh: he fainted and was a week
in bed, and as soon as he rose, he fled, and the priest asked his mother about him
and to her did say: "What is become of your son"? And his mother answered
him: "Great travail you gave him; for that reason he has fled". And at this the
priest became wroth, and then on Sunday, when he began scattering holy water,
"asperges" as he called it, in the midst of the church, as he walked around
scattering this holy water, he hit the mother of the sacristan lad on the head with
the hyssop, and drew from her much blood. (Lienhard 1992: 79)2
The written testimony of the

indigenous intellectuals
Some indigenous letters freed themselves from the conventions of the epistolary
tradition to take the form of an "autonomous" literary genre of unexpected dimensions.
Firstly, I have in mind the lengthy letter-narrative (66 folios) which the Inka Titu Cusi
Yupanqui, supreme authority of the rebel Inka state of Vilcabamba, dictated for his
colleague, King Philip II of Spain, in the year 1570 (Yupangui 1985); secondly, the
vast letter-chronicle (1189 folios), which the supposed Quechua cacique Guaman
Poma de Ayala attempted to send to Philip III around the year 1615 (Guaman Poma
de Ayala 1980).
Despite their proportions, these widely known texts display all the traits of the
shorter indigenous letters. In particular, they share with them (although not in
grammatical terms) a sense of speaking for the whole community. Their explicit
addressee is the king of Spain himself, who is supposedly above all the concerns of
colonial life. Their basic motivation, however, is still clearly the assertion of claims.
Other indigenous chronicles, like that of Santacruz Pachacuti Yamqui (1968 [1613])
in Peru, or the "Historical Compendium of the Kingdom of Texcoco" from
Ixtlilxochitl (1975 [1608]), descendant of the lords of that place, come close, though
not explicitly, to being an epistolary genre in the terms that we have been attempting
to define.
Far from limiting themselves to a specific complaint or vindication of rights, such
texts express with a high degree of literary elaboration the whole historical vision of
their authors and the sections of society that they represent. Here, the indigenous
signatories have ceased to be humble writers of supplicatory letters in the European
style but have become authors entirely in their own right, individuals conscious of a
radically new literary practice. In each case, it is difficult to be sure to what extent
their highly unconventional mode of narrative expression belongs more to the Euro-
pean or to the indigenous tradition. The dominant feature of their composition, without
any doubt, is hybridity: if the choice of form (the letter-chronicle or the narrative
account) gives an outward semblance of European conventions, some of their linguis-
tic/narrative resources and the sources on which they draw relate by contrast to an
autochtonous oral heritage.
Reformist and revolutionary platforms

From the eighteenth century onward, the indigenous communities started from time
to time to abandon the notion that the only person who could be explicitly addressed
in the process of communication was the king of Spain or his representatives. Several
of the epistolary texts of the eighteenth, nineteenth, and twentieth centuries, sometimes
termed memoriales, are "open letters", or manifestos addressed to what would today
be referred to as "public opinion". Their authors were the elected or self-proclaimed
spokespersons of the indigenous communities.
The style of such manifestos, erudite in style and lacking all features of oral
enunciation, ceases to allude to the oral tradition. They are perfectly geared to the
cultural and linguistic norms of the interlocutor, but this does not imply any political
submission to the colonial or criollo authority. It has ceased to be anything more than
a tactical device designed to ensure the best reception of the message. Moulded in
"acceptable" forms, the message which these texts transmit continues to be "indige-
nous" (although without the connotation of "pre-Hispanic"), because it relates to one
of the political options that had been steadily forming within the indigenous world.
Written around 1749, the "True Representation and Humble and Lamentable
Exclamation Made by the Whole Indian Nation to the Lord King of all the Spains'
Majesty . . . " (Bernales Ballesteros 1969, reproduced in abridged form in Lienhard
1992: 241-254), was one of the early indigenous manifestos. Defended and dissem-
inated by Fr. Calixto de San Jose Tupac Inca (descendent, on the maternal side, of the
Inka Tupaq Yupanqui), this text—or one of its previous versions—came to serve in
the 1740s as the political platform of a major neo-Inkan sector of the Peruvian
population (Bernales Ballesteros 1969). Whilst it is, in formal terms, a "letter" or
petition to the king, this manifesto was widely circulated—seemingly also in printed
form—before as well as after its delivery to the official addressee (1750). Tired of
having to petition the king and his council for everything, the "indian nation", sensing
the breezes of reform then blowing, was attempting to reach the ear of all persons of
goodwill.
Inspired, as the author points out at the beginning of the text, by the lamentations
of the biblical prophet Jeremiah, the eminently literary form of this "lament" consti-
tutes a sort of (superficial) concession toward the addressee:
Defecit gaudium cordis nostri. Versus est in Luctum chorus noster. The happi-
ness of our hearts has been lost. Our canticle has become a lament; because the
joy of being Christians and vassals of a Catholic monarch has fallen from our
hearts, on finding ourselves abased by Christians and affronted by Catholics.
(Lienhard 1992: 248)
However, at heart, the argument of the text follows a rigorously autonomous logic.
The endorsement of Christianity and the monarchy is accompanied by the emphatic
condemnation of "real Christianity" and veiled criticism of a king who does not
honour the undertakings contracted by his great predecessors. The whole text breathes
the desire for greater political and religious autonomy for the natives. A precocious
call to an incipient "public opinion", the neo-Inka Representation served at the same
time as a platform for negotiation with the king and—thanks to the publicity
achieved—as a means of exerting pressure upon the addressee and his government.
Voices of insurgency
Whether to defend the remnants of their autonomy or to strengthen their position in
the process of negotiation, almost all Amerindian communities have taken up the
armed struggle at one time or other in their history. War does not mean the interruption
of letter-writing—on the contrary, from the eighteenth century at least, indigenous
insurgency has tended to manifest itself in an intensification of the dialogue in letter
form with the Spanish or criollo adversary.
The state of insurgency has deep repercussions on the communicative process
underlying these indigenous letters. Instead of being addressed from a position of
inferiority to an interlocutor considered superior (not to say "divine"), the subject that
emerges in the writings of the fully mobilised communities is an "I" or "we" who
addresses his adversary as equal to equal, or even inverting the relationship, as
"superior" to "inferior".
The Guarani War (1752-1756) and the Tupacamarist

Rebellion (1780-1781)
According to the terms of the frontier treaty of 1750, Spain was to hand over to
Portugal seven flourishing townships of Christian Guaranis in the area east of the
Uruguay river. The governor of Buenos Aires was the person charged with executing
the necessary expulsion orders in the name of the king. If for the Jesuits the prospect
of the loss of seven of their reducciones was difficult to accept, for the Guaranis it was
a frontal attack on their very existence, physical, economic, and cultural.
In the series of letters in Guarani (Mateos 1949;3 four reproduced in Lienhard
1992: 330-346) which they addressed to the governor in 1753, opposing the handover
of their ancestral territory to Brazil, the elements of an autonomous indigenous mode
of expression emerge amid the protestations of fidelity to the king and the gospel. The
seven indigenous responses to the threats of the governor offer, with formal variations
as well as more fundamental ones, the same type of argument: the Guaranis were never
conquered by the Spanish, rather they "gave themselves" to God and "chose" the "holy
king", God's representative on Earth, as their protector. The king confirmed them their
possessions, in the name of God. Immutable, like God, the king cannot have changed
his mind; it is inconceivable, therefore, that he should wish to expel them from lands
that he himself—or one of the earlier royal "incarnations"—entrusted to the care of
those under his protection.
"Immutable in his will" and not linked in any way to Spanish colonialism, this
king is a mythical figure, a product of the Guaranis' imagination. In recognising solely
the authority of a "mythical" monarch, they divorce themselves not just from the
colonial power of Rio de la Plata but from that of the historical Spanish king—who,
of course, was "mutable in his will".
"Masters" of the king, the Guaranis are equally so of God, of a God imposed on
them, historically, by the Spanish. Secure in this certainty, the caciques of the town
of Santo Angel address the governor:
Even if you bring your cannon, we shall not fear. God Our Lord alone, we being
poor indians, shall help us greatly, and the holy Angel also shall be our aid and
protection. It may be that God Our Lord will place you in our hands.. .. you
trust in your cannon and artillery. May it serve you well, being that in which you
place most trust. All that you say to us is in vain.... It will all avail you nought.
(Lienhard 1992: 342)
The insurgents, the "we" of the letter, now place themselves at the centre of the world,
ready to take any measures whatsoever to defend their autonomy, and arrogate to
themselves alone the interpretation of the values which the agents of the universal
monarchy claimed to defend: Christianity and the idea of royalty. The style of the
Guaranis' dialectic, particularly when we take into account the sense of holy war that
appears in some of their letters, suggests the messianic character of their movement.
Far from being the precursor of criollo emancipation which some historians have
made him out to be, Jose Gabriel Condorcanqui Tupac Amaru, leader of the great
Andean Uprising of 1780-1781, in the letters which he addressed to his adversaries
(in elegant Spanish) used a discursive logic reminiscent of that of the Guaranis. Indeed,
in his plea-letter to the visitador Areche (5th March 1781), a sort of political testament,
Tupac Amaru portrays himself as embodying, with no possibility of contradiction, the
"will of the king" (Durand Florez 1980-1982, t. Ill [1981], 204-222, reproduced in
Lienhard 1992: 258-272). According to his virtually irrefutable argument, not only is
the violence deployed against the local functionaries of the king himself deserving of
no punishment whatever, it is even worthy of a royal reward: "And if killing those
guilty of lese-majeste is rendering a great service to the crown, we, by killing the chief
magistrates and their minions, are rendering great service to Your Majesty and are
worthy of reward and gratitude" (Lienhard 1992: 269). If the Guaranis knew that they
could count on God against: the outsiders, Tupac Amaru is, as can be gathered from
his letter to the visitador Areche, the very instrument of divine justice:
For as in the holy church of Sangarara, being there lodged the Most Holy
Sacrament, they slaughtered women, which drew God's wrath down upon them
straightway: for just as they did not venerate the Holy One, neither did the Holy
One avail them, and as abusers of clergy they perished without assistance of
clergy. Although I sent a pair of lads around the streets proclaiming that they
should return to Christian ways (for my mind was not to kill them but rather to
take them on one side, give them the benefit of my reason and set them on the
road to salvation), Heaven in its lofty judgement closed the opportunity to them,
and with their own hand they gave themselves up to la Parca (Death), being thus
the authors of their unhappy fate. (Tupac Amaru refers doubtless to his victory
at Sangarara, on 18th November 1780; Lienhard 1992: 266-267.)
The self-proclaimed representative of God and king in Peru, Tupac Amaru seems
in fact to subscribe to an Andean code of values and norms of behavior, as embodied
in the header that appears on his letters and manifestos: "Indian of royal Inka blood".
Speaking with the authority conferred on him by, inter alia, his seventy thousand
indigenous soldiers, Tupac Amaru addresses himself to the criollos with the intransi-
gence that the Inkas were wont to display in the face of their adversaries ("Warning
of Tupac Amaru to the Criollos of Cusco", 20th November 1780, in Durand Florez
1980-1982, t. Ill [1981], 97-98; reproduced in Lienhard 1992: 257):
If spurning this my warning you do the contrary, you will experience your
ruin, my gentleness will turn to fury, and, just as I can say it, so have I strength
to do it, as I have at my disposal seventy thousand Indians, and other
provinces which have offered themselves to me and which I have at my
orders; so do not make light of this my warning which is born of my love
and clemency. (Lienhard 1992: 257)
The unqualified respect for the "will of the king", and the free interpretation of
the will of God, is obviously a cover for a different political reality: that of the rebirth,
under the powerful impetus of the "indian of the royal stock of the Inkas", of Andean
autonomy.
The process of recovery of a measure of autonomy on the part of indigenous
societies in the colonial period is, beyond a doubt, the common denominator of the
uprisings led respectively by Jose Gabriel Condorcanqui (Peru) and Nicolas Nenguiru
(Paraguay). In the letters just referred to, not only is the theme one of restoration of
freedom, but the desire for this restoration also brings about a profound modification
of the communicative process which until then had governed indigenous letter-writing
habits. This transformation of the process gives rise to a new phenomenon, the
"manifesto addressed to and against the outsiders", which is not bound in any way by
the system of "feudal" rules that originally applied. In appropriating for themselves
the "divine" authority of the king, the insurgent indians do not address themselves to
a superior, nor do they speak from the periphery to the centre. They are, rather, on the
point of inverting the roles, and making themselves the centre of their own world.
Yucatan Caste War, 1847-7

It is in the letters of the rebel Mayas of Yucatan (mid-nineteenth century) that the
inversion of roles—and the rejection of all the traditional rules and conventions of
communication by letter between the Indians and their adversaries—reaches its
culmination. As a clear example of this we may cite the letter that Juan de la Cruz,
prophet and "messiah" of the "speaking cross" movement, sent on 28th August 1851
to Miguel Barbachano, the governor of Yucatan (Bricker 1981: 208-218; reproduced
in abridged form in Lienhard 1992: 129-135). Carrying the indigenous appropriation
of universal religion and royalty to its logical conclusion, the author presents himself
from the very outset as the Son of God the Father himself, as the creator of the dzules
('outsiders, whites'), the indians, the Negroes, and the mulattos. In his capacity as
divine authority, situated—like the king in the ideology of universal monarchy—
above both opposing groups, Juan de la Cruz orders that an end be put to the "mutual
slaying" between Indians and dzules, that his dzul "children" free the Mayan prisoners,
quit indigenous territory, and celebrate thirty high masses in his honour. Here, we are
very far indeed from a simple assertion of rights. Should his commandments, inspired
only by love of his "children", not be obeyed, the prophet promises a truly "divine"
punishment:
Don Miguel,
All these my commandments
that I am sending to thee,
Those here,
It is necessary
That thou obey them.
Because [if] I see
That thou art not obeying them,
I will place a great punishment
On the city of Merida
And over thy land. (Bricker 1981: 216)4
If the inversion of the roles of "vassal" and "lord" is, in this text, truly spectacular,
the rejection of epistolary conventions is no less obvious. Like the other insurgent
authors who emerged during the course of the "caste war", Juan de la Cruz insists in
his dialogue with the chief of the dzules on using the Mayan language (a language
which Barbachano doubtless understood). In itself, the choice of the indigenous
language did not necessarily imply a rejection of all European rules of discourse. Juan
de la Cruz, however, far from simply conceiving the letter in Spanish and putting it
into Mayan, adopts the prosody—parallel distichs—of Mayan ritual language (Bricker
1989: 341-343) and presents us, in short, with a text totally rooted in Mayan verbal
culture.
The peasants 'struggles against spoliation: Mexico (1856) and

Peru (the 1880s)
Already anachronistic by the second half of the nineteenth century, the values that the
rebel Mayans defended, whilst at the same time subverting them, still referred to the
colonial experience. In the same period, other insurgent native factions claimed, in
opposition to the "national" landowning oligarchies, to represent certain values which
are in themselves properly criollo, such as "patriotism", "democracy", the "nation",
the "people", "civilisation", and "progress". Frequently, rejecting their marginalised
status, they claimed for themselves the role of vanguard of the "people" or of the
"nation" in defense of a republican order which was, in reality, highly unfavourable
to them (Anti-landowner proclamation of the indians of Zacoalco and San Cristobal,
Jalisco, 30th September 1856, in Reina 1980: 148-150; reproduced in Lienhard 1992:
139-143):
The indigenous people of the town of San Cristobal, in union with those of
Zacoalco and the white notables .. ., in common accord recognise no other type
of government but the representative, popular, federal type with no restitution
[restriction?] whatsoever, upholding it with weapons in hand until the last drop
of blood shall have been spilt. (Lienhard 1992: 142)
The peasant and indigenous notion of liberalism was quite distinct, of course, from
that of the liberal classes who wielded power. For the peasants, "emancipation" meant,
above all, the recovery of their lands:
The predecessors of the indigenous peoples took up arms in eighteen ten, serving
under the orders of the illustrious leader of Independence, don Miguel Hidalgo
y Castilla, in the firm belief that with the Spanish government overthrown, they
would recover the vast possessions which the latter had usurped from them with
bloodshed to the detriment of all the other neighbours; an episode which was
witnessed with pleasure by the landed classes, who more than any others are to
blame for the misery and the backwardness in which our unfortunate fatherland
finds itself. (Lienhard 1992: 141)
This (conditional) deference to the republican government, in this case that of the
liberals, does not signify the abandonment of indigenous autonomy, but rather its
suspension. If the government does not comply with what the indigenous communities
expect of it, they will indeed return to defending their rights for themselves: "[We
beseech the higher government to] be so good as to examine the contents of the
enclosed, and to lose no time in giving its higher verdict on the matter, to avoid the
peoples having to reconquer by arms their usurped properties and reclaim them as
property of the Indians" (Lienhard 1992: 143).
The claim to an indigenous "patriotism" also appears in the language of the
Peruvian montoneros who fought against Chilean occupation in the early 1880s. The
native guerrilla fighters of Comas, in the central mountains of Peru, wrote as follows
to a collaborationist landowner (Letter from the guerrilla leaders of Comas, 16th April
1882, in Manrique 1981: 393-394; reproduced in Lienhard 1992: 284-286):
We know that you, along with other traitors of our beloved Fatherland,5 are in
this Province, communicating with, and giving instructions as to how the
Peruvians can be ruined, to those perfidious Chilean bandit invaders, traitors,
like yourself, of their country. . . . Do not include us in the ranks of the barbarians
. . . , as we, with right and justice on our side, unanimously rise up to defend our
Fatherland: we are true lovers of the land of our birth.
The "fatherland" of the Comas guerrilleros has, to be sure, little in common with that
of the criollo oligarchy. Once again, as in the case of the Tupacamarist Uprising, it is
a Utopian "fatherland" constructed not from the perspective of criollo Lima, but from
the standpoint of the ex-Inkan highlands.
Here again, the inversion of roles in the letter-writing process is signalled by the
fact that the sender does not defer to the values of the addressee, but rather subordinates
those values to himself, making of them values of his own. In appropriating the idea
of "fatherland" from their criollo adversaries, the authors of indigenous or peasant

letters such as these constitute themselves—in political and communicative terms—at
the very centre of that "fatherland", relegating their adversaries to an ignominious
"periphery". In the letter from the Comas guerrilleros, native speakers of Quechua,
one can also observe the appropriation (and the equally radical transformation) of
Spanish, the language of their adversaries.
Conclusion
As regards the terminology employed, the "indigenous discourse addressed to the
outsiders" is dressed up, almost inevitably, in the values—or the former values—of
their addressees: Christianity and monarchy, republic and fatherland, people and
nation, progress and civilisation. As evidenced by the texts of the indigenous letters,
the political significance of such terms depends, nonetheless, on the specific place
that the interlocutors occupy in the process of epistolary communication. During
the "consolidated colonial period", with the indigenous sender still in a position
of inferiority and on the periphery with respect to the "centre" (the king), the
official meaning of these terms was not questioned—at least not openly. As we
are dealing here with eminently "diplomatic" texts, it is difficult at times to gauge
to what extent the indigenous people actually subscribed to the cause of the
"outsiders".
At the moments of indigenous counter-offensive, which occurred continually
throughout the eighteenth and nineteenth centuries, on the other hand, the indigenous
authors of letters, manifestos, and proclamations, occupying—for the moment at
least—the "centre" of the process of communication, do not hesitate to invert the
official semantics of these terms to give expression to the persistence of an indigenous,
autonomous, interpretation of society and history. Masters of their destiny, the rebel
indians appropriate to themselves the values—or ex-values—of their adversaries; they
subvert them and fling them back at their former masters, just as they return blow for
blow. The subversion consists, in fact, in interpreting them completely literally and
carrying them through to their logical conclusion. If God is universal, any human
community may appropriate him. If the king is king by divine right, there is no reason
why the colonial power, which is merely temporal, should arrogate to itself the role
of being His exclusive representative; the king belongs to all those who follow God's
law. If the Republic guarantees equal rights for all citizens, the war against discrimi-
nators and despoilers is a just war. The Fatherland does not belong to those who
proclaim themselves its owners, but to those who defend it against its adversaries, and
so on. Stripped of the ideological content that they had acquired through their use by
the Spanish or the criollos, the values introduced and abused by the "outsiders" are
converted, in the indigenous texts, into forceful arguments against those who falsely
claimed to uphold them.
ACKNOWLEDGMENTS Translated from the Spanish by Christy MacHale.

Notes
1. Epistolary communication of a horizontal kind (i.e., between the members of an
indigenous community themselves), a different phenomenon from that dealt with here, is the
subject of the work of Sarah Lund (in this volume).
2. "4° oc cepa ce piltoni sachristan 8 xihuitl quipia cenca miec oquimecahuitequic cenca
oquixipehuac ynacayo ocotlahua ce semana huetztoc omoquetzac niman otzoloc auh yn totatzin
quitlatlania ynantzin quilhuia catia moconeuh quitoa ynantzin oticmacahuitequic otzoloc auh
yea oqualanic totatzin auh yn iquac ypan domingo ya teteochihua yea tlateochihualatl yquac
mitoa asperges oncan nepantla teopan yahui teteochihua yca ysopo oquihuitequic cihuatzintli
ynantzin pilton sacristan quiquatzayanac yxquich esti oquiquixtilic". (Anderson et al. 1976:
168)
3. Mateos (1949) does not provide the Guarani originals, but only a contemporary
translation of them, presumably the work of one or more of the Jesuit missionaries in the area.
4. Dn. Miguel
tulacal lein Val mah than

Cin tuchitic techa
halo
tza
Cah aoocbesic
tumen Cin Vilic
mah tan aoocbesice
Cin oaic humpel unohochil Castigo
tu noh cahil Cansihoo
.yokol alumil. (Bricker 1981: 216)
5. The use of the term fatherland in translation, here and in the ensuing discussion, is
deliberately chosen to reflect the patriarchal connotations of the original patria (translator's
note).
11
SARAH LUND
On the Margin
Letter Exchange among
Andean Non-Literates
I In this essay, I am concerned with examining the meaning of written forms of

communication for a largely non-literate1 Quechua-speaking society. My particular
focus in exploring this question rests on the interrelationship between the spoken word
and the written text in communication between close family members living far apart.
In such exchanges, memory plays an essential role in the oral delivery of the message
content, while the written form has the role of legitimator and potential source of sorcery.
Thus, the personal letter takes on the characteristics of an epigraphic phenomenon, placing
on record the names and identity of persons connected with the physical object.
The preoccupation with the oral/literate dichotomy has a long tradition in the
Western history of ideas. Rousseau, in his essay "On the Origin of Languages",
brought to our attention the inherent conflict between orality and textuality. His views
of literacy emphasised the enslaving impact of that medium on the natural spirit of
man, an idea which was to later influence Levi-Strauss, McLuhan, and finally Derrida
(Havelock 1986: 36), Jack Goody's earlier writing on literacy (1963, 1977) was also
firmly placed in this sequential tradition, the consequences of literacy concerning "The
Domestication of the Savage Mind". According to Goody, literacy made possible the
growth of knowledge and the shift toward the objectification of the individual (cf.
Harbsmeier 1984: 78; Havelock 1986: 113-114). Most of these arguments give
primacy to the irreconcilable contradictions between the oral and the literate; the
contrasts in the communicative media creating in themselves distinct kinds of societies
with different kinds of conscious citizens.
185
In my opinion, the view of orality and literacy as mutually exclusive stems from
a far too rigid typological distinction between oral and literate society. In the former,
communication is seen to be exclusively acoustical. In terms of the latter, literacy is
equivalent with phonetic writing only. Certainly this dichotomy is far too severe, and
we are forced to wonder whether there was ever a purely oral society untouched by
the use of some system of signs to communicate.
In our preoccupation with difference, we have neglected the powerful ways in
which these interrelated forms of communication articulate. Recently this oversight
has been drawn more fully to our attention as we experience our own literate tradition
being powerfully transformed by a form of secondary orality (Havelock 1986: 31).
The written preparation behind all of the oral presentation in contemporary media
confronts us with the dramatic interplay between the oral and the written. It is the
dynamics of this interchange which I wish to explore here, not in terms of confronta-
tion, infiltration, or forced progression along a scale of complexity, but as a kind of
dialogue between interdependent media. This articulation and interchange has no
doubt always been a part of human experience.
This study must also be placed in the particular ethnographic context of the
resettlement of highland villagers (Quechua-speaking runa)2 both to the east and to
the west of the Andes.3 I have been preoccupied with experiential aspects of this
mobility in the hopes of attaining insight into runa views on the process of separation
and the essence of unity. Initially, I focussed on rituals of departure and arrival, treating
them as essential rites of passage that mark the important thresholds of transforming
experiential worlds. One of the obvious themes that arose from this discussion centres
on villagers' confrontation with literacy in three very specific contexts: that of
acquiring identity papers; of signing work contracts; and finally, of writing letters as
part of ritual exchange between families residing in the highlands, the coast, and in
the eastern jungle. In a largely non-literate village setting, these three contextualities
seem to me to provide a particularly latent opportunity in which to explore wider
theoretical issues concerning the interface, as opposed to the contrasts, between orality
and literacy and between text and context. Further, the interplay between personal
consciousness and social domination in the context of emerging literacy can be seen
as particularly enlightened within the context of these relatively contained ethno-
graphic situations.
Letter-writing in particular is one of the situations that Walter Ong (1986: 156)
has characterised as forming a margin between the oral and the textual. As John
Salisbury said: '"Letters speak voicelessly the utterances of the absent'" (quoted in
Clanchy 1979: 204). Thus, I have chosen of the three possibilities for this chapter to
focus particularly on letter-writing between villagers from Matapuquio. I will be
dealing with letters authored and witnessed by wider family constellations, written
down by local scribes, and delivered by travellers who are either departing from or
returning to the village. It will become apparent that this courier takes on the role of
the ultimate interpreter of the message in a situation in which both the author and the
addressee communicate through a medium that they do not master and in a foreign
language only poorly understood. In this communication maze, it will also be appre-
ciated that letters are viewed with great suspicion, as media of sorcery.
On the Margin 187
My choice of letter-writing as a point of departure rests on several important

considerations. First, while identity papers and work contracts are textual compo-
nents in relationships between individual villagers and the wider socio-cultural
universe of Peruvian society, letter-writing is an internal communication between
closely related villagers living at a distance, a kind of communication that struggles
with a general lack of knowledge about respective contexts. Further, much of the
dynamic of this communicative exchange either struggles to impose the ascendency
of a certain set of collective values over individuals removed from the collective
pressures of life in the village, or they form part of the verbal assurances that such
responsibilities are still respected despite lengthy absences. Thus, I find letters and
letter-writing a particularly poignant example of the complex shiftings between
shared meanings and disparate appreciations, between collective memories and
divergent expectations, between the fear of witchcraft and the longing for a genuine
message.
The written oral interface

Before turning to the specific instance of letter-writing, let us take the opportunity to
discuss in a more abstract sense the contrasts and similarities between orality and
literacy. A seeming consensus lies in the importance given to the free-floating nature
of the written word, disassociated as it is from the context of its production. Some
authors have gone so far as to argue that, more than any other invention, writing has
changed the parameters of human thought in striking ways (Goody 1986; Hallpike
1979: vi; Havelock 1986; Ong 1982: 14-15). The role of literacy has been seen to
have powerful liberating effects fundamental to logical analytic modes of thought, a
sceptical and questioning attitude, and the recognition of the important distinctions
between time and space, myth and history. A literate people have the wherewithal for
innovative, democratic, and socially humane thinking, essential to economic devel-
opment, the process of urbanisation, and the rise of complex government (Gee 1988:
196). To be literate is to be civilised (Havelock 1986: 35-36).
This obvious idealisation masks the historical complexity of literacy in society (cf.
Graff 1979). By addressing a particular written practice of a people poised on the
margins of a literate tradition, I hope to address some of the ambiguity inherent in the
role of literacy. I will emphasise how literacy is embedded in certain social settings,
thus implying that the process of interpretation is an integral aspect of the practice of
literacy. Further, the basis for interpretation is essentially bound up with socialisation,
with the social groups and institutions that form the context or the conditions of the
text's production (Gee 1988: 209).
Plato has been given the distinction of being the first writer to record his
scepticism toward writing, and some of the doubts he expresses about the literate
mode certainly hold relevancy for our discussion here. Plato held the written mode
as suspect because he saw it as undermining the role of memory, especially in the
process of acquiring knowledge. "For Plato, one knew only what one could defend
in face-to-face dialogue with someone else" (ibid., 197). The written text was
suspect because of its quality of seeming completeness, giving it an illusive
authority. In a contradictory way, Plato also points out the great weakness of
writing in that it cannot stand up and defend itself. For Plato, true knowledge
comes when one person makes a statement and another asks, "What do you mean"?
Such a request forces the speaker to "re-say", that is, to say in different words,
what he or she means. In the process he sees more deeply what he means, and
responds to the perspective of another voice/viewpoint. (p. 197).
In contrast to the written word, orality presupposes dialogue. It is from the
context of communicative exchange that one can say that authorship as we know
it does not exist in a non-literate society (Foley 1986: 8). The oral narrative is
collectively fashioned, and the context must be appreciated to grasp the meaning.
As opposed to the universalising nature of the written word, such discourse is
particularistic, taking into account the conditions of performance and the various
roles of oral forms in the culture as a whole. From an oral tradition, the immediacy
of the audience is an inherent aspect of the performance, allowing as it does for
an ongoing assessment of what can be taken for granted as shared understanding
(Todorov 1984: 34). This is one of the most difficult transitions to be overcome,
when someone from a largely oral tradition assumes a literate mode of communi-
cation. There is no first-hand experience of interaction by which to judge the posing
of the message. As we shall later come to appreciate more fully in the particular
case of letter-writing, it is with difficulty that the orally oriented writer, who is
basically delivering a monologue in the form of a letter, can imagine the reader
and anticipate her/his reactions to the words. An oral orientation presupposes an
active interlocutor (Sweeney 1987: 178-179).
Within the historical context of colonialism, indigenous peoples of the Americas
frequently hold the belief that their languages cannot be written down (e.g.,
Brotherston 1979; Perrin 1986: 211). For them, learning to read and write involves
assuming a foreign language with unfamiliar vocabulary and grammar, a language
that is strongly associated with their subordination. The power of elites is held to
be inseparably bound up with their power of writing; their exclusive source of a
kind of materialised and infinite knowledge. Here, too, Christian missions have
been active for centuries, evangelising through the medium of the written word in
a holy book. These foreigners are assumed to be in the possession of special powers
of divination by means of which they can communicate with their gods. As opposed
to indigenous means of supernatural communication based on the personal inter-
mediary of the shaman, the external nature of the written word is held to be
particularly potent and of another order, making shamans of all outsiders (cf. Perrin
1986).
Seen in this light, literacy is a part of the yawning gap between an indigenous
world and that of outside domination. Thus, for many indigenous peoples, learning
to read and write implies putting aside an entire way of life, accepting the premises
of an elite society, and ultimately embracing foreign gods. In such a situation,
literacy implies far more than the "technologising of the word" (Ong 1982), the
additional accumulation of technical skills. Far from being part of an inevitable
process, this perspective on literacy rather poses a complexity of problems laden
with normative ambiguity.
On the Margin 189
Matapuquio/Chanchamayo/Lima:
The letter circuit
Keeping these brief comments on a complex field in mind, let us turn our attention to
the specific instance of letter-writing between Matapuqenians. That this simple ex-
change is fraught with tensions and ambiguity can be suspected in terms of what has
been said so far, but first let me develop the background to this situation more fully.
Matapuquio is a runa village in the southern highlands, in one of Peru's poorest
departments, Apurimac. Though holding their own territory, the villagers have been
under the influence of the local hacienda for centuries, by working on the estate to
gain access to vital pasturelands, additional agricultural fields, and more recently to
cash income (cf. Skar, H. 1982). Neither wages nor any form of formal education were
available to the villagers before the mid-1960s. The change occurred when the local
hacendado (land boss) began to offer money in payment for work and built the village
school at his own cost on hacienda property bordering the village. The school
curriculum is limited to the first three years of primary school. The teachers corne from
the surrounding mestizo towns, but they hold their posting with a degree of disdain,
and their attendance and efforts are sporadic. Village children have many family
responsibilities, and schooling in general, coupled with the teachers' attitudes in
particular, are viewed by parents as highly questionable. While all instruction is carried
on in both Quechua and Spanish, the goal of teaching is to learn to read and write in
Spanish. In fact, none of the teachers that I have known in the village were able to
write Quechua, and they were fascinated to see my books in that language. With no
books, and few of the students having paper and pencil with which to work, the school
is woefully hampered in being able to teach reading and writing skills. On completing
three years of "compulsory" education, the vast majority of village children have
learned to speak some Spanish, and they can write and recognise their names.
There are villagers who know how to read and write, however. Young men and
women who have left the village for the coastal city of Lima have often, and at
considerable sacrifice, gone to night school. When these young people return to the
village to marry and settle down, they begin to serve as scribes, writing down the many
letters which are composed in the village and sent with travellers leaving for the coast
or to the eastern jungle.
From a village perspective, the two groups of non-residents in Lima and
Chanchamayo form what they would call colonies (mitmaqkuna). Those living outside
are subordinate to their families and ultimately their village in the highlands ((ayllu).
This hierarchical relationship of mitmaq to ayllu is fundamentally associated with the
continuing rights to village land, irrespective of long absences. One of the main
expressions of this subordination can be found in a fairly complex system of exchange
between village and jungle and between village and city. The letter forms a part of
this exchange. Though this is not the only context in which letters circulate, neverthe-
less, it is one of the main situations in which villagers are compelled to communicate
in writing.
Let me be more concrete. When someone is leaving Matapuquio for the jungle or
for Lima, they take with them agricultural produce to give to their families living
outside. This food is considered an important medium by which physically to bind the
non-resident villagers with the substance of their lands and their kin in the highlands.
It is a kind of pledge that village rights are being maintained for the ones that are away.
In return when someone travels up to the highlands from the coast, dry goods, clothing,
and money are sent along as the appropriate response to the agricultural gifts. People
living in the jungle send fruit, coffee, and money. In this way, produce and objects
associated with one region are exchanged with those from the other, thus becoming
important anchors in reproducing patterns of kin-based socioeconomic relations
between ecological zones.4
The letters accompanying such commissions are mainly a kind of inventory. They
are a recognition of what has been previously received and a declaration of what is
enclosed. They serve to keep a check on material possessions and as such are evidence
of the power relations between villagers, and over worldly possessions, all bounded
within the system of exchange. This exchange has close associations with the Incaic
and colonial mita system, a kind of work tribute which also had as an aim to acquire
resources unavailable in the home environment (Murra 1982; Wightman 1990:
86-88). If we see these exchanges of produce and letters of inventory between mitmaq
and ay llu as tribute, in the sense of a testimonial of respect with an urgency of cohesion,
we can more easily appreciate how closely literacy is associated with the exercise of
power at a distance. The bulk of the letters I have available at this time seek to keep
a check over the movement of material possessions and human beings. They serve as
a control mechanism within the dynamics of a local tribute system in which pre-em-
inence of the highland community is constantly seeking reaffirmation from its mitmaq.
The role of the written document becomes that of authoritative voice checking,
comparing, and demanding accountability.
Letters exchanged between Matapuqenians are concerned with accountability, and
in this role they have a definite connection with contractual agreements. They are
documents with evidential value linking writing with units of wealth, particularly
money. From this perspective there are many similarities between the role of the letter
and that of the contract and the identity papers mentioned in the introduction, as being
the initial encounters with literacy for departing villagers. All three of these kinds of
documents can be interpreted as demonstrating the qualitative shift from status to
contract, the long-term implications of which are the disassociation of law and custom
(Goody 1986: 144).
The three texts all bear the mark of strong legal undercurrents which play off
the tension between an abstracted legal position with the specificity of in-
dividualised agreement. In the example of the personal letter reproduced below,
we see that acceptable standardised form limits the particularised message at the
same time giving expression to the essential particulars of a specific transaction
completed in the past.
Matapuquio, el 21 de Setiembre 1985

Estimados compadrcs,
Reciben de Esteben Quispe y Epifania Cotarma sus compadres, asi como de
Victor su ahijado y de Evardista Cotarma los mas calurosos saludos.
Les agradecemos del dinero que nos ban mandado; Este agradecimiento viene dc
todo corazon.
On the Margin 191
Epifania ha estado mal de salud pero con el favor de Dios ya este bien.
Sin otro particula me despido en el nombre del Senor.
Abrazos
Epifania
Nota: Cuando van a venir para aqui?
[Matapuquio, 21st September 1985

Dear Compadres,
Receive the warmest greetings from your compadres Esteban Quispe and
Epifania Cotarma, as well as from Victor your godchild and from Evardista Cotanna.
We thank you for the money you have sent. This gratitude comes from the bottom
of our heart.
Epifania has been ill but with God's grace now is well.
With no other news, I will close in the name of the Lord.
Abrazos
Epifania
P.S.: When are you going to come here?]
The collective orality of letter exchange

There is another side to letter-writing and exchange between Matapuqenians that
remains to be explored, one that gives us an opportunity to take into account the many
aspects of a letter's production and delivery. I would move away from the generalised
institutional effects of literacy, in this particular case letter-writing, toward an explo-
ration of the indissoluble links between literacy, context, and meaning. Here we are
concerned with a type of text, the letter, and the various ways of both producing and
reading those letters as social and historical creations.
For the majority of Matapuqenians, a kind of collective orality is the single most
important aspect of creating and receiving letters. This is true of the inventory letters
described here, as well as letters with a more intimate content. There are so few who
can take on the role of the literate in the communication so that letter writing and
reading is a group effort. Someone is leaving for Chanchamayo, and a woman, newly
widowed, will try to reach her son with the news of his father's death. She wants her
son to come home to help her. It is arranged that a scribe will come to her house to
write the letter. The courier/traveller is also present, as are several married children,
and one of their spouses. The deceased man's brother is specifically asked to come.
People perch around on the low walls of the patio and the letter is begun, the scribe
labouring over a piece of lined notebook paper with a small pencil stub lost to view
in his large fist.
At the top is entered the place and date of the letter and the complete name of the
addressee. Without further introduction, the text begins with a progress report on the
welfare of those present, and continuing in this formulaic vein inquires into the health
of the long absent son from whom they have not heard for a very long time. There are
no capital letters or punctuation to mark sentences, but if read orally the meaning is
clear enough. The main point of the letter follows: the father is dead, and the son is
needed at home. In very brief terms, the mother insists that the son return and promises
to find him a wife and provide everything he needs to begin his own family. Again in
a very stylised manner, the letter is closed 'for lack of more to say' ("sin mas que decir
te despido"), with greetings also from the father's brother, as well as sisters, brother,
and brother-in-law being specified. Another piece of paper is folded into a kind of
envelope, and the name of the son is written on the outside with only one indication
of address, Chanchamayo. No one really knows where the son is in Chanchamayo,
which is a huge territory, but maybe someone amongst the other villagers residing
there will have had word of him.
This is a summary of the letter's text, but is no indication, whatsoever, of the lively
discussion carried on in conjunction with forming the letter. The father's brother is
the youth's godfather (padrino), and he confers with his sister-in-law about which
family they should approach in asking for a wife. It becomes apparent that the courier
is not leaving for Chanchamayo by chance, but is being sent by the widow to find her
son. The courier is the widow's younger half-brother, and there is much discussion as
to whom he shall seek out to ask for help when he arrives. The siblings and
brother-in-law, much older than the absent youth, obviously feel strongly that their
brother must return to take up his responsibilities as youngest child—that is, the
support of his mother and the cultivation of the remaining fields that have not already
been divided out between them. Their presence, in addition to that of the mother and
father's brother, adds authority to the letter. Long after the scribe is finished with the
brief epistle, and the widow has put her thumbprint at the bottom of the letter over her
name, the courier is being coached as to what must be said to truly convince the son
to return.
Eventually, the traveller departs for the jungle, and finally he does manage to find
his nephew who is working with four other youths from Matapuquio on a coffee
plantation at Santa Inita. The letter is delivered, but none of them is sufficiently literate
to read it through, though they recognise the name written below the fingerprint and
can decipher the village name, the date, and the boy's name. Eventually they ask for
help with the reading, but long before the actual text is read and the content of the
letter delivered, the uncle has been telling about the father's death and how the boy
must come home. He can relate how the elder siblings are committed to his return to
take up his inheritance and that the mother and father's brother are already concerned
with finding a wife for him. Questions and answers, problems and solutions as to how
things can be arranged are being aired long before the formal message is ever received.
In fact, the letter itself is almost forgotten in the oral discussion about its content, or
rather the complex message behind its content.
In writing about rituals of travel, Van Gennep (1960: 37) describes the letter
as a kind of sign that wins recognition for the traveller, automatically incorporating
the courier into the receiving group. Through the letter, the traveller is able to
identify himself with others he meets. As I have learned to appreciate among
Matapuqenians, the letter really serves as an object of trust, authenticating the oral
message which the courier delivers. In many cases, I would go so far as to say
that the letter is the authenticating symbolic object and helps solve the problem
of communication at a distance. For many, the tangible letter conveys the views
and feeling of the writer's wishes, far and beyond what the written content actually
communicates. The letter is a physical reinforcement of what is being said (Clanchy
On the Margin 193
1979: 245-246). In fact, though the youths scrutinised the letter's headings and
signature, they failed to show any interest in critically examining the written
contents of the document. The letter becomes simply a subsidiary to the memorisa-
tion of the message and not its replacement.
Every text is of a type. When considering the interface between literacy and orality,
the letter, the type of text under discussion here, poses requirements in terms of our
expectations of both form and content. "One always and only learns to interpret texts
of a certain type in certain ways through having access to, and ample experience in,
social settings where texts of that type are read in those ways. One is socialised into
a certain social practice" (Gee 1988: 209). Within a highly literate social practice, the
familiar letter as a type is associated with intimate conversation. In the eighteenth and
nineteenth centuries, it was cultivated as a literary art form within the context of an
intimate relationship. The essential texture of this genre gave ascendency to the
conversational paradigm, favouring the speaking tone of voice (Redford 1986: 2).
For the runa, interpretation of a letter involves an entirely different practice. As
depicted in the case above, the letters consist almost entirely of formulae, listing titles
and epithets of sender and addressee. They confirm relationships, but do not express
a one-to-one kind of intimacy between individuals. Sentiments are generalised to the
extent that one can wonder at their impersonal tone. However, it should be recalled
that one of the biggest barriers of literacy for an orally oriented person is to imagine
the audience. The audience postulated in many of the letters written by Matapuqenians
would seem to be pure stereotype.5
The runa letter generates meaning on its terms above and beyond the written
meaning on its page. I would venture to say that the burden of meaning within the
practice of letter exchange rests on the experience of the letters as presence (Foley
1986: 16). One integrative theme in my research (Skar, S. 1994), which has allowed
me to place ethnographic elements in new relationships, discusses the total identifi-
cation runa often assume between the sign and the object represented.6 Many of their
beliefs in magic are founded on assumptions about sympathetic and contagious links
in the sequence of cause and effect. Letters are held to be very forceful objects in this
sense and are almost always received with suspicion and fear, as well as joy and
anticipation. The most disturbing letters are those exceptional ones actually received
through the post. In one instance, the addressee actually kept the letter arid the envelope
of such a posted letter in separate places. The envelope was obviously very disturbing
to him with its stamps and elaborate address. It was the envelope he took out to show
me first, asking in a conspiratorial tone: "How could they possibly know me, know
my identity number"? He did not appreciate how an address worked with its levels of
inclusiveness. Nor did he realise that that which we call a postal service is not a limited
group of people who know you intimately, but rather a relay of de-personalised links
from periphery to centre to periphery again. That the post was able to deliver a letter
to him on the basis of what was written on an envelope obviously gave them some
kind of power over his person. In this medium, total strangers knew him too intimately,
and the thought was disturbing. There was a possibility that the letter, especially the
envelope, could have some malevolent power over him since it came to him via
strangers.
Among Matapuqenians there is a strongly held belief that letters are objects of
sorcery (winapi). Felix Cardenas, a village outsider who keeps a simple shop in his
house, is known to specialise in written sorcery. It was claimed that he once wrote
such a letter for his wife who was unhappy with her daughter-in-law for accusing her,
by letter, of neglecting the cow left in the mother-in-law's care. In the process of
writing the reply, don Felix took three lighted candles and turned them upside-down
saying the full name of the daughter-in-law, then proceeding to write a letter. I was
assured that one could kill an entire family in this way. At any rate, in this case, the
mother-in-law was writing to her son, but the sorcery was meant for the daughter-in-
law, who is named. The bewitched letter caused the son to beat his wife for no apparent
reason. But the daughter-in-law's family knew that it was the misunderstanding over
the cow left in Matapuquio which was ultimately behind the beating. The vindication
of the mother-in-law was made possible through the letter, which was sent by a
different courier from the one who delivered the original accusation.
It is a dangerous thing to receive a letter too open-heartedly because it may not be
what it seems. Loss of fortune, abandonment by spouse, or the fact that one never
returned to the village have all been attributed to bewitched letters. Through the letter
as object, the malevolent attitudes of the sender are placed right at the heart of the
receiver. No matter what the tone of the written words might be, the contagious effect
of the evil intent is what becomes communicated. The letter has a kind of organic
identity with the sender. When received by the addressee, the link is not broken. Once
put into contact with the receiver, powerful objects such as letters of sorcery continue
to exert an influence.
Conclusion
Largely hampered in their movements earlier, it is mostly during the course of the past
twenty-five years that Matapuqenians have moved out of their village to establish
permanent settlements in the jungle and on the coast. This separation has stimulated
the use of literate forms as a means of communication between villagers at a distance.
However, the use of writing in the letters exchanged between villagers takes on
meaning in terms of other forms of communication. Problems of authenticity, of
trusting what is written, continue to rest largely on the persuasive powers of the
courier. Memory, instead of written content, still comprises the main medium of the
message relayed between groups. An intimate one-to-one conversational tone associ-
ated with the familiar letter-writing has no place. The letter's set phraseology and
physical format reflect links with techniques of memory and oral tradition as well.
It seems paradoxical that, within written tradition, the letter is a literary genre/type
which strives for the oral tone, the writing down of intimate conversation, and as such
might be characterised as a literate form at the margins of orality. From the standpoint
of a non-literate tradition such as that of Matapuquio, letter-writing has nothing to do
with conversation. Rather, it is characterised by formulaic ingredients: an elaborate
focus is placed on devices of authentication such as the signature, seal, dates and place,
and identified witnesses. The physical document is scrutinised, while the written
On the Margin 195
contents are given little or no critical examination. Letters for Matapuqenian readers
serve as important signs of recognition, legitimising the accompanying oral message.
Our discussion of the role of letter-writing in a non-literate society has thrown into
relief the ambiguity and complexity in the dynamic relationship between the written
and the spoken. The limited use of literate forms relates at many levels to social
processes and structures, an appreciation of these being necessary to reveal the full
significance of the written. The role of letter exhange is a complex reality with
contradiction and discontinuities arising out of the social setting of which it is a part.
Thus, the letters exchanged by Matapuqenians become a convincing example of the
inextricable link between text, context, and meaning.
ACKNOWLEDGMENTS My research has been funded by the Norwegian Research Council for
the Social Sciences and the Humanities (1976-77), by the Swedish Social Science Research
Council (1984, 1986), and most recently by the Institute for Cultural Comparative Research
(1991-92). I am grateful for their generous support.
Notes
1. I use the term non-literate in the title and the text as opposed to illiterate. The former
emphasises the acoustic character of successful oral communication; the latter is couched in the
negative image of failed communication in a medium not mastered (cf. Havelock 1986: 119).
2. Quechua-speaking villagers in the southern sierra refer to themselves as runa, or 'human
beings'. I have chosen to use their term of self-reference in this chapter, out of a sense of respect
for their community and in recognition of the many difficulties involved in using other terms
of reference such as "indian".
3. Lives Together Worlds Apart. Quechua Colonization in Jungle and City (S. Skar 1994)
explores in a more comprehensive way the uses of literacy in the context of migration.
4. This dynamic economic interdependence between contrasting ecological areas is re-
ferred to as "vertical economy" (Murra 1982) and has been a crucial factor in interregional
relationships for centuries.
5. See Sweeney (1987: 180) for a discussion of Malay letter writing and the difficulties
posed in communicating a piece of news.
6. Another powerful example of this is found in Allen's contribution to this volume. There
the discussion focuses on the use of inqas in the context of pilgrimage, while in this case letters
bear the qualities of their authors. In both cases the aspect of movement in space is important
and seems clearly associated to more generalised ideas about the flow of life energies.
12
TRISTAN PLATT
The Sound of Light

Emergent Communication through Quechua
Shamanic Dialogue
For Norman
It is becoming increasingly difficult to interpret Andean oral testimonies in isolation

from the wide range of visual memory techniques (including forms of "writing")
which they invoke and presuppose (Arnold et al, 1992; Condori and Ticona 1992). In
a previous article (Platt 1992), I argued that even the oral performances of modern
North Potosf shamans are grounded in a deep historical belief in the authority of
graphic inscriptions (qillqa) which can, in principle, be "made to speak" (parlachiy).
The experience of the Spanish invasion led much of this graphic authority to be
resituated in the alphabetic code and representational aesthetic of the invaders—as,
classically, in the work of Waman Puma (Adorno 1986), The authority of the
alphabetic code developed particularly in the context of the catechetical and notarial
discourses reproduced through the administrative and archival practices of the colonial
church and state (Mannheim 1991), while the new representational aesthetic emerged
above all in the wealth of sacred art aimed to skew native perceptions of the divine in
the direction of a Christian body of iconographic conventions (Gisbert 1980; Fraser
1982; MacCormack 1991). Further, our awareness of different forms of literacy
among colonial Andean populations (e.g., Adorno 1982; Itier 1991; Duviols and Itier
1993; Lienhard 1992) has grown to the point where it is becoming necessary to ask to
what extent the existence of mass alphabetic illiteracy during the Republic is not itself
196
The Sound of Light 197
a relatively recent product of a process of modernisation which, at the same time, has
created for purposes of its own reproduction the image of a "purely oral" Andean
culture.
The oral text here presented belongs to an Andean shamanic tradition that has also
been marked by the great imperial bureaucratic revolution introduced into European
statehood by Spain, where the most elaborate documentary and archive-based methods
of government thitherto known in Europe were developed in the sixteenth century
(Gonzalez Echevarria 1990). The result is, in the first instance, a form of textuality
that seeks authorisation on both oral-performative and scriptural grounds. In the
seance I shall present, the mountain spirits, or jurq'u, are summoned by the shaman
to answer questions put to them by a small domestic congregation, but their speech
and identity are marked by the scriptural context within which they act. The political
and religious context is constructed through dialogue by incorporating references to
the literate practices present in the wider society, which both legitimate and are
legitimated by the oral procedures of the seance.
For within this inscriptional context, where Andeans recognise "writing" as a
major instrument of colonial authority, while at the same time seeking to domesticate
it, the shamanic seance manages to keep open an oral and dialogic space within which
contexts, frames of interpretation, and truth conditions are up for negotiation (cf.
Duranti and Goodwin 1992). This has various consequences for Andean approaches
to the construction of knowledge: (1) alphabetic metaphors may be used creatively to
authorise other forms of graphic inscription; (2) the "dictates" transmitted by the
mountain spirits through the shaman to the congregation can be discussed and
reinterpreted according to local perspectives; and hence (3) "the letter of the law" is
seen to be at least minimally flexible and modifiable in relation to local realities,
Finally, the transcript allows us to analyse the way in which communication is con-
structed over an evening made up of a sequence of short conversations between the
shaman, the congregation, and the mountain spirits that serve a powerful local St James
known as "Father Pumpuri" (Tata Pumpuri)—who himself remains off-stage. From
hesitant beginnings, dialogue becomes gradually more fluent, until the antiphonal af-
firmation of speech in the last section allows the seance to be recognised as a success
by all parties. Attention to the wider aural context allows us to see this collective rec-
ognition of communicative competence emerging from a rich tilth of supporting noise:
whistles to summon the jurq'us, the clinks of bottles during libations, the crackle of
coca-leaves being handed round, the sighs and invocations of the shaman in trance, the
semi-audible murmurings of the congregation, and finally the rhythmic beating of
wings that announces the arrival and sustains the presence of the mountain spirit in the
form of a condor-messenger from the lofty frontiers between earth and sky.
Communication with unknown sources of sacred power is an inescapable aspira-
tion for the ethnographer who wishes to write about religion. When Evans-Pritchard
said that only believers could do so, he meant that otherwise the vital connections, per-
spectives, and chiaroscuro of the believers' truth will inevitably escape one. While my
own faith is insecure, volatile, and often vanishes, I have tried to write as a believer to
help represent the "reasons for faith" that underlie these acts of communication with
the personalised forces of the cosmos. In support of this, I can at least claim to have felt
the other side of faith in Father Pumpuri's power, which is fear. However, it will be
useful here also to invoke that suspension of disbelief which is expected of the audi-
ence at any dramatic performance. The relation between the development of Andean
shamanism and that of Andean drama is clearly a fruitful avenue for future research.
Meeting Tata Pumpuri

Geographically, Santiago (St James) de Pumpuri is a tiny hamlet of a few straggled
houses surrounding a small adobe chapel, situated on a high plateau above 4000 metres
above sea level near the northwestern borders of Macha territory. Looming over it, on
the borders with the altiplanic Department of Oruro, rises the mountain range known
as the Cordillera of the Friars. Pumpuri has been one of the rural annexes of the Indian
parish of San Pedro de Macha at least since the seventeenth century. This parish
(doctrina) was founded by Viceroy Francisco de Toledo in the 1570s as an adminis-
trative and evangelical focus for the highland segment of an important dual society,
Macha, which had been the ethnic "capital" (cabecera) of the pre-Hispanic "federa-
tion" (senorio, or nation) of Qaraqara. Today, the hamlet falls on the borders between
two cabildos (rural tribute-paying groups), each belonging to one of the five minor
ayllus into which each of the two moieties of Macha have been divided since
pre-Hispanic times.
The old white-washed and thatched chapel of Pumpuri can be seen from far off,
at the ceremonial entry (ch 'isiraya) to the sacred plain where the cult of St James is
centred. Here pilgrims pause to chew coca and pour libations before approaching the
tiny village. The appearance of the tower of the old chapel made me uneasy when I
first saw it in 1971:' with only one turret surviving the summer rains like a peaked
hat, and two bells that seemed almost like eyes, its whitewashed adobe base was
stained until almost halfway up its height by the blood of sacrificed animals. Inside,
the shadowy reredos (retabld) was full of clustered images, with that of St James
dismounted in the middle, his aquiline face black-bearded and crowned with a hat, the
sacred heart of Jesus suspended over a crescent moon with its horns upturned, and the
Bolivian tricolour over one shoulder.2 Today, the old chapel is flanked by a modern
church roofed with corrugated iron, where the image of Father Pumpuri has been
relocated by priests aiming to "clean up" the Pumpuri cult, although his devotees say
that he still prefers the old adobe chapel (Figure 12-1).
The power of St James of Pumpuri is recognised over much of the southern Andes:
pilgrims arrive for his feast on 25th July from northern Argentina,3 Chile, and southern
Peru, as well as from distant parts of Bolivia—La Paz, Oruro, Potosi, Sucre, Santa
Cruz, the coca-growing Yungas of Chapare, and the eastern lowlands. The strength
of the faith he inspires is visible in the strong emotions aroused in his devotees:
unconditional adoration and obedience, pride, anguish, weeping, fear and trembling.
He is known for his medical powers: he can kill and he can cure, and he is sometimes
invoked as the Angry Doctor (phina miriku). His treatments are perceived either as
disasters (disgrasya), with which he can punish ritual neglect and moral
misdemeanours, or as good fortune (surti), with which he rewards the devout. Health;
fertility of humans, animals, and crops; and economic prosperity are offered as the
reward of religious faith.
Figure 12-1 Tata Santiago de Pumpuri: a photo sold to pilgrims during the fiesta.
As doctor,4 St James is patron of curers and shamans, but he does not tolerate black
magic or devil worship (something of which his church critics are not always aware).
However, neither does he belong exclusively to the realm of the High God (altu dyus),
as does another "miracle", Tata Killakas, who is based in an altiplanic shrine near Lake
Poopo. Tata Pumpuri partakes in the Glory of the Upper World (janajpacha), but he
also has "roots" (saphi) of Glory in the Inner World (ukhu pacha). He is therefore a
clear expression of the need to "offer to both sides" (purajman jaywayku), to the pow-
ers of Without and those of Within, characteristic of Andean Christianity, which rec-
ognises the essentially creative role of the chthonic powers in generating new life in
tense coexistence with the Upper Deities and the High God (cf. Platt 1987).5
The Tata goes to battle raised above merely human foot-soldiers on his horse: his
most terrifying weapons are lightning (glurya sintilla, the Flash of Glory), thunder,
and thunderbolt (rayu), which are imagined as the flash, rumble, and bullets of his
arquebus.6 The thunder also represents the sound of his horse's hooves as he gallops
through the clouds. His metallic bullets (walas, from Spanish bald) streak to earth with
the lightning—k'aj! in Quechua onomatopoeia—and fulminate animals, houses,
church towers, people, and especially those fated to be a shaman, oryachaj.. For these,
the experience is an initiation: they die, but are reassembled at a second flash and
resuscitated at a third. The triple imagery of Easter, as well as of the pre-Hispanic cults
of Charcas, is thus reproduced in their death and resurrection. In their new vocation,
they will communicate with Father Pumpuri and place their indian clients in living
contact with him, either channelling his voice directly or through the medium of the
condor-mountain spirits which are his servants.7
At the spot where the initiation took place, now sacralised as surti parisirun (the
place where the Luck appeared), the future yachaj finds one of Santiago's bullets, still
smoking with sacred energy, and nurtures it with dishes of sacred foods (glurya
jampis) while learning to become a shaman. The best way to neutralise its dangerous
energy is, indeed, to channel it into a shamanic session, or Council (kawiltu).
Spirit possession
The shaman's effort to construct the conditions for communication between congre-
gation and spirit messenger is a difficult and risky undertaking. The bullets themselves
must be taken to hear Mass in a Christian church from time to time, to prevent their
"eating" the initiates by draining them of energy till they die. My friend Santiago, one
of the main actors in the text that follows, had been struck by lightning, but resisted
the call to become a shaman. He suffered from a long-drawn-out illness, one symptom
of which was the arrival of a condor in his bedroom while he was asleep, eager to
occupy his body. The experience left him frightened and shaken, and in 1977 he
eventually died. As so often in South Andean thinking, inherently dangerous sources
of sacred power can only be tamed partially for the benefit of human society if they
are placed under the aegis of the Christian church; in Andean Christianity, this includes
the obligation to accept the role of shaman if one is nominated by the lightning bolt
of St James.
Spirit possession is itself dangerous and exhausting: the spirit enters in the shape of
a bird—generally a condor8—and possesses the shaman, giving him extra intelligence
(aswan intilijinti) and changing his voice to the point where the audience can hear him
and the spirit conversing as two separate people. To reach this pitch of spiritual suscep-
tibility, he must chew quantities of coca leaf, smoke tobacco, drink neat alcohol and
grape spirit, and pour libations for the sacred sources of power while approaching the
state in which he can suddenly fall into a state of trance and spirit possession.
The seance (kawiltu, from Sp. cabildo 'council') I attended and taped in August
1971 consisted of six separate sessions, divided into one group of four and one of two.
Before they began the candle was extinguished, and the spirit was heard to enter and
speak in complete darkness. Within each group, there were pauses between the
sessions when people talked in the darkness, commenting on the answers given by the
spirit and phrasing new questions. These "intervals" are a valuable commentary on
the form and content of the preceding session of shamanic dialogue. The candle was
relighted during the central "intermezzo", which separated the two groups, and then
again at the end. The overall frame for the event was therefore given by this oscillation
between light and darkness, establishing an elementary visual code within which the
aural experience was Inscribed.
Most of the elements that contribute to the global signification of the event are
aural, and have therefore been caught on the tape, though they cannot include
smells, touch, and the intimate feeling of familial warmth shared before the
invisible "action" on the altar table. Gesture is only available in aural form: actions
are etched on the darkness. Intonation, murmurs and rustles, hesitations, the
clinking and blowing of bottles, the "whistle" of the shaman's bullet, the slurps of
the condor drinking alcohol offered by the yachaj, and especially the rhythmic
flapping of his wings as he perches invisible on the altar table—all combine with
distant background noises to make up an overwhelmingly aural field of meaning
that dispenses with all visual cues beyond the basic opposition between the candle
and the dark.
Setting the scene

The yachaj we asked to hold the council was a compadre of my friend Santiago
and his wife Feliciana, the couple I was staying with in Liq'unipampa. His name
was Ignacio Mosques.9 One bright winter's morning in August 1971, we went to
the little mill house on the river nearby where he was grinding wheat flour. The
water glittered beneath the sunlight as it streamed through rocks and pebbles down
from the distant salt mines and licks at Salinas. A turf-lined canal threw a heavy
flow of water on to the wooden water wheel beneath the mill, and the sound of
the stone mill wheel could be heard rumbling inside the house. Upstream, over a
bed of tumbled stones and boulders, lay the hot baths of Phutina, and further
on—out of sight from the mill house—the gold and antimony mines of Churina.
The world smiled.
We entered, and Santiago began to explain things while I watched the flour dusting
the edge of the lower millstone as the upper stone turned ponderously on its vertical
shaft. At first Ignacio was reluctant; he said his bullets had been playing up, making
him ill, and he had had to take them to the Sanctuary of Tata Killakas to hear mass.
But the tape recorder clearly interested him, and he liked the idea of taping a session.
Finally he relented and agreed to come that very night. We promised him 15,000
bolivianos (August 1971 rates).
During the rest of the day all the children in the hamlet were full of excitement.
They played in the sunlit patio at being the jurq'u who would be summoned by
the yachaj, imitating his first greeting—winas nuchis wawas kristyanus!'Good
evening, my Christian children'!—and laughing at their own efforts. But when that
evening, well after dark, Ignacio finished his meal of liver with hot sauce (but no
Christian salt) and entered the little room where we were going to hold the council,
the children fell silent.
From the moment the yachaj entered, the dramatic performance had begun.10
Laughter was wiped from every expression. By candlelight, the altar table (misa) was
prepared behind the door: an old wooden box, on it a woman's weaving (ajsu), folded,
then a smaller rations-cloth (inkhuna), then another folded cloth. The yachaj accepted
quantities of coca leaf which he covered in the folds, along with his two bullets. He
began to take coca leaf continuously, offering it round regularly to the congregation
who received two handfuls at a time, seated, between two cupped hands (yanantin;
see Platt 1986). An empty aspirin bottle was filled with pure alcohol and given to the
yachaj: this was the una, the 'offspring' of the bigger bottle (wutilld) full of diluted
alcohol, which was passed round for all to drink.
We discussed with the yachaj the questions we wanted to ask and who was going
to put them to the mountain spirit. The yachaj continued drinking and taking coca leaf,
combining tranquil courtesy with a certain reserve, as if he were saving his strength,
or was simply allowing himself to enter the state of increasing stupor produced by the
combination of coca leaf, alcohol, and cigarettes. Sometimes he heaved a sigh and
murmured barely audible invocations—to Holiest Father (the Sun), Holiest Mother
(the Moon), Father St James of Pumpuri, etc.—while he drank pure alcohol and libated
with drops of the diluted alcohol. One of his bullets had a little hole in it, which he
filled with neat alcohol as an offering for the jurq'us; then he blew a libation with a
long-drawn-out spluttering whistle that the jurq'us would hear and respond to. As the
evening wore on, the whistle became less diffuse and shriller. Finally, he asked us to
extinguish the candle.
We were now in complete darkness. At first there was barely a sound, except for
the dogs barking outside, or a distant child crying in the night. From time to time we
heard the yachaj blow into the bottle and the una, and in the little orifice of his bullet
(wala), which now emitted an acute whistling call to the distant jurq'us. 11
Then he began to speak in a normal manner, asking us what we wished to know,
laughing at my questions about Inka and Chullpa. Everyone began to converse again,
unseen. We went on taking coca leaf, and I continued to feel a growing stupor, the dark-
ness wrapped me round, the bullet continued whistling and the bottles huffing and
puffing, and I heard dimly the clinking of glass and the rustle of coca leaves from the
direction of the altar table. The murmurs of the yachaj as he poured libations alternated
with conversation; he communicated a sensation of relaxed dignity. He ventured a
question—and at that moment the sound of violent flapping broke out high up against
the wall behind him; we were compelled to silence as it descended toward the floor,
hitting the box on which the altar was laid out, and striking my knee and the micro-
phone. And a loud, authoritative voice greeted us: "Good evening, Christian children"!
THE COUNCIL8
Liq 'unipampa, August 1971

Macha, Moiety Alasaya, Ayllu Alaquyana, Cabildo Phichichua.
Scene: A small square room, thatched and plunged in darkness, in the hamlet of the
Carvajal patrilineage on the low puna region of Macha. Throughout their visit, the
mountain spirits continue beating their wings in a regular flapping as they sit bal-
anced, invisible, on the altar table (misa).
Session 1
Enter jurq 'u with wings flapping
Jurq'u
winas nuchis, wawas kristyanus!
i "chunkaiskayniyuj chakullani" niwaychij, kristyanus;
"jisus maria waqaychiri" niwaychij, kristyanus;
ima ofrisisunkichij, wawas kristyanus?
[Good evening, Christian children!
And call me "Twelveb Chakuiiani"c say to me, Christians;
"Jesus Mary Protectress"d say to me, Christians;
what is your problem, Christian children?]
Santiago
kunan "kay chimpapi kanman", nin, "uj chullpa santus kanman", nin.
kanmanchu tatay? chayta watukuy munayku.
[Now, he says, "over there might there be a Chullpa saint",e he says;
might there be, my father? That's what we want to ask.f]
Jurq'u
asi ti pikaru pis!
awir, kay wirjintachu parlachiy munankichij, awir, kristyanus?
o sinu kay uj qullutachu, wawas kristyanus, awir?
aKawiltu, Sp. cabildo (municipal council), is used to designate the dramatic speech genre j refer to as
'council' or 'shamanic session'. The text was taped in Liq'unipampa, hamlet of the Carvajal patrician, in
August 1971, and transcribed and translated in the days immediately following the session.
For the importance of the number 12 in Northern Potosi, which is libated to as 'Father Twelve' (tote
chunka iskayniyuj), see Tristan Platt (1983). Pre-Hispanic duodecimality has probably converged with such
Christian expressions as the Twelve Apostles. Here the sacred number functions as an honorific.
"The name of a neighbouring hill.
dwaqaycha- 'protect ; + -iri(Aymara agentive nominaliser) 'protector'.
e
The Chullpa are the pre-Inka people who lived before the rising of the sun. The name is also given to
the house-like burial tombs that are often found at archaeological sites. These tombs all face east, according
to modern Macha, because the Chullpa thought the sun was going to rise in the West. When the sun rose
most were burned up in the holocaust, except for a few who dived into the water and became the modern
ethnic group Uru-Chipaya. 'Chullpa' therefore refers to a pre-Inka culture, whose traces are to be seen in
their archaeological remains, which exert dangerous forces on the living and whose pre-solar time persists
in a shadowy form within the earth, manifesting itself above all at the new and full moons. Andean time is
therefore not linear like an arrow but cumulative like the layers of a cake.
This question arose from an experience a few days before. Santiago and I had gone to Milluri to see
whether the legend of a chullpa saint's altar and cross located inside a neighbouring hill was true. We had
found a tunnel entrance, and I had crawled in to find a parting of the ways, one going down but silted up,
the other emerging on another part of the hillside. The question had remained unresolved, so I had asked
for it to be put to the Jurq'u.
f
ln Potosi Qucchua, final syllable stress indicates loss of a suffix, whose identity must be inferred from
the wider sentential and discursive context. Cf. Cerron-Palomino 1987 for the same process in Cochabamba
Quechua. In this text, the missing suffix is the object-marker -to, unless otherwise indicated.
[What a rascal, then!g

Let's see,h do you want to have this virgin speak, then, Christians?
Or else one of these peaks, Christian children, let's see?]
Santiago
"kanmanchu chaypi santus"? nispa kay wiraquchi watukuy munan an'.
[This wiraquchii wants to consult, please, he says "might there be saints
there"?]
Jurq'u
isu pikarus! winu, awir, pusarqamusaj, winu?
[That's it, rascals! Well, let's see, I'll go and bring it here,-) okay?k]
Santiago
awir . . .
[Let's see , . . ]
Jurq'u
a sinur santyaku miriku glurya istrilla nasimyintu!
awir, tata santisimu mustramu!
ima niwasunchus, wawas kristyanus, winu?
[Ah! Lord St Jarnes Doctor Glory Star Birth!1
Let's see, Father Holiest Our Lord!m
I wonder what he'll tell us,n Christian children, okay?]
gIn some forms of this greeting, the derivation is clearly from Spanish sinpecado 'without sin', as an
epithet of the Virgin Mary. For example, in another council I taped with a different yachaj in 1978, we heard
sinpikaru kunsiwira 'sin pecado concebido' (conceived without sin). See Sola 1969 for similar greetings
formulae in use in Chinchero (Cuzco): A. awmarya 'Ave Maria! B. simpikaru 'sin pecado', 'without sin'.
However, 1 am grateful to Rory Hamilton for pointing to another possible derivation from Spanish picaro
'rascal', which would be adapted to the Quechua vocalic, consonantal, and accentual system in exactly the
same way as pecado'. 'pikaru'. Here my translation hazards a derivation from haztepicaro,puesl ! 'be a rascal,
then'! In general, I take pikaru as deriving from picaro. Those who wish can always try substituting 'sin';
it is even possible that a conflation of the two meanings may have occurred: 'sinful rascal'.
"awir", Sp. a ver; translated throughout as 'let's see'.
'I.e., the anthropologist, Tristan. Wiraquchi is the standard term for translating gringo; cf. kawallim,
from Sp. caballero.
^pusarqamusaj: note the combination of suffix -rqu- (-rqa- before -mu-) indicating the jurq'u's sudden
exit, with the cislocative -mu- indicating his expected movement of return to the altar table. The object of
the verb is unclear: it may be simply the answer, or it may be the chosen peak or virgin itself.
Final syllable stress here indicates missing interrogative suffix -chu.
Istrilla nasimyintu was glossed by participants as 'born up in the tempest'; a derivation from the Star
of Bethlehem seems probable.
'"A string of Spanish-derived epithets for St James: "medico, gloria, estrella, nacimiento, santi'simo,
Nuestro Amo"... 'Doctor, Glory, Star, Birth, Holiest, Our Lord'... 'Holiest Father Our Lord' refers to the
Sun and also to the Christian Host; see Platt (1987).
"Note the frame-indicator niwasunchus, made up of ni- (say) + -wa-sun- (directional 3ps 1 ppl inclusive)
+ -chus (dubitative): deictic reference is to speech from outside the immediate spatial context. The jurq'u
Exit jurq'u with wings flapping
Yachaj
bay tata santyaku tata!
a "chay wirjintachu munankichij, qullutachu? mayqinta wajachfy munankichij"?
"chay wirjin wajachfy munayku,
kanchus imachus awir chay a chullpaj santun a"? ninaykichij a.
[There,0 Father St James, father!
So (he's asking you:) "do you want that virgin, or that peak?
"Which? do you want to call on"?
"We want to call on that virgin you must say then,
"is that Chullpa's saint there or what, let's see then"?]
Santiago
a chayta mana asirtaykuchu a.
[Yes, that's what we're not certain of.q]
Yachaj
"kanchus imataj chaypi"? nispa ninkichij a.
[You must say (to Jurq'u), "I wonder what's there" . . . ?]
Santiago
aaaaa.. .
[Ahhhh]
Yachaj
bay tata santisimu mostramu!
mana nuqanchij jinatajchu parlanku, ratu, tatay, paykuna parlanku.
[There, Holiest Father, Our Lord!
(pouring libations and chewing coca)
They don't speak like we do, they speak very fast, my father,]
here groups himself with the congregation over against Tata Pumpuri, who is expected to reveal a mystery
to all (including the jurq'u). The session shows a constant process of frame-shifting, which may include distant
participants outside the immediate speech situation, such as Tata Pumpuri himself. See Figure 122.
0bay Sp. vaya! A common interjection in Northern Potosi Quechua.
p
Here stress probably indicates the absent discourse-marker -taj. Note that, rather than indicating the
relationship between the sentence in which it occurs and the sentences immediately preceding and following it
(as is usual with such discourse markers), the suffix must itself be inferred by the listener—on being alerted to
its absence by stress—in the light of his/her understanding of the wider context. Compare the study of the
discourse-markers used in the Quechua narratives of Huarochiri in Dedenbach-Salazar Saenz 1994.
q'Again Santiago focusses on the content of the question rather than on the way in which it should be put,
as the yachaj invites him to do. The whole of this interlude is taken up with the yachaj 's comments on the nature
of jurq'u speech, the constraints it places on human speech, and the best way to establish dialogue with it. Not
until his own confusion is invoked (mana nuqanchij jinatajchu) does Santiago connect experientially with the
yachaj's drift (ratu a turwachikun), whereupon metalinguistic dialogue is at last under way.
Santiago
ratu a turwachikun chaytaj . ..
[Yes, very fast, that's what confuses one . . . ]
Yachaj
chayta tata Griguryu, allinta tapuriy, tapurinkichij a.
[That question, father Gregorio, ask it right, all of you ask it then.]
Gregorio
awir tatay .. .
[Let's see, my f a t h e r . . . ]
Yachaj
sinu "yasta kawiltu"! nipullawasunchij a payqa, i? ujta apaykamuytawanqa.
[Otherwise , he'll just say to usr
"Cabildo 's over"!s no? as soon as he's brought in another one.t]
Santiago
ari a.
[Yes, indeed.]
Yachaj
kuntistajtin atintiwasun,
mana kuntistajtinqa ratu chay "yasta kawiltu"! nispa phinksupullanqa payqa a.
bay maria kunsiwira bay!
parlakuychij, tatay, parlakuychij, parlakuy tata Griguryu, imallatapis . .,
[If there's an answer he'll attend to us,u
if there's no answer, at once he'll say "Cabildo's over"! and just rush
out.v]
(Whistles with his bullet and libates)
'Note the incorporation of Sp. ya estdl 'okay, it's finished"! as Quechua complement. Alternatively
from Sp. bastal 'enough now'!
''nipullawasunchij: here the frame-indicator -wa-sun-chij groups the yachaj with the rest of the human
congregation and excludes the jurq'u and his speech as external to that group (frame B, Figure 12-2).
'The use of -yku- (-yka- before -mu-), with a paradigmatic meaning of gentle entry into an enclosed
space, complements the jurq'u's expansive use of -rqu- above inpusarqamusaj (cf. note j). The object of
the action is similarly unclear.
uThe subject of kuntistajtin—from Sp. contestar +-jti- (switch-reference) + -n (3ps possessive)—may
be Gregorio: 'if he answers'. In atintiwasun ( Sp. atender) the frame-indicator -wa-sun- again situates the
jurq'u's speech outside the human speech-community (which includes the yachaj) (frame B, Figure 12-1).
v
From Quechuaphinkiy 'leap, jump' (glossed as Sp. brincarse). The Aymara suffix -su-, equivalent to
Quechua -rqu- (see notes j and t), is preceded by vowel-deletion according to the Aymara morphophonemic
rule. Although Aymara influence is strong among these Quechua-speakers (some of whose great-grandpar-
ents spoke Aymara as their first language), the choice here may be motivated in part by the Quechua
predilection for assonance and onomatopoeia (after nispa lit. 'saying', citative framing); cf. Mannheim
1991.
There, Mary of the Conception, there!

(To the others)
Talk, my father, keep talking all of you, talk, father Gregorio, just
anything .. .w
Gregorio
bay tatay ...
[There, my father ... ]
Session 2
Yachaj
misaman, tatay!
[To the altar table, my father!]
Jurq'u
winas nuchis, wawas kristyanus!
[Good evening, Christian children!]
All
winas nuchis, tatay!
[Good evening, my father!]
Jurq'u
a "chunka iskayniyuj chakullani awugaru" niwaychij, wawas kristyanus
[And say to me "Twelve Chakullani Lawyer"," Christian children!]
Santiago
chunka iskayniyuj chakullani awugaru!
[Twelve Chakullani Lawyer!]
Jurq'u
a "jisus maria anjilawarta waqaychiri" niwaychij, kristyanus.
wa! ima okasyun tiyan? wa! awir, dun santyaku, awir pis!
a dyus pagarasunki muchachu!
"This key utterance shows the importance of unsemanticised sound, murmuring, as the aural ground
upon which the jurq'u's intervention can be constructed. It is swamped by the swishing wing beats of the
arriving condor.
^awugaru, from Sp. abogado 'lawyer'. Here the jurq'u adds a further honorific to his title that
demonstrates his participation in the construction of a legal discourse.
[Ah, Say to me "Jesus Mary Guardian Angel Protectress"! Christian

children!
Wa! What's the problem? Wa! Let's see, don Santiago, let's see then!]
(Drinks alcohol. To the Yachaj)
[Ah, thanks, boy!y]
Yachaj
jina, tatay, wirjinta waturikuyta munashan,
chayta kawalta waturipuy, tatay a.
[That's so, my father, he's wanting to consult the virgin,
so ask her correctly for him, my father.z-]
Jurq'u
wa! "jisus maria" niwaychij, wawas kristyanus!
[Wa! Say to me "Jesus Mary", Christian children!]
AH
jisus maria!
[Jesus Mary!]
Jurq'u
winu kunan, qan, imata munanki, awir, dun Griguryu?
awir, imataj asuntusniyki, awir, waway kristyanu?
[Well now, you, so whataa do you want, let's see, don Gregorio?
Let's see, so what's the matter with you, let's see, my Christian
children?]
Santiago
chay . .. chay qullapi kanmanchu, tatay, chay chullpaspata santusnin?
[In that. . . that peak, might there exist, my father, Chullpas' saints?]
Jurq'u
isu ti pikaru! piru, waway, mana kasqachu.
[That's it, rascal! But, my child, it seems there isn't.bb]
y
Nole how the yachaj here channels the speech by which the Jurq'u constructs him (the yachaj) as a
dependent of some kind—possibly the door-boy or clerk who shows the lawyer into the courtroom.
'The yachaj has decided for Santiago which divinity to choose, and asks the Jurq'u to 'consult the virgin
correctly' (kawalta, from Sp. cabal + -to [adverbial marker]). But it turns out that the Jurq'u has already got
the answer. The display of independent actions and intentions help establish the yachaj and the jurq'u as
two entirely different actors in play.
aa Stress here indicates loss of final suffix -taj.
bb
Another "shamanic flight" of virtuosity: after inviting Santiago to put his question again, the jurq'u
reveals that he has already made the consultation during his absence, and produces the answer like a rabbit
out of a hat. Note the use of the reportative suffix -squ, indicating that the jurq'u has received the information
indirectly from another with certain knowledge; for the implications of Quechua evidentials for the
cpistemoiogical status of Andean history, see Howard-Malverde 1990.
Santiago
manachu kasqa?
[There isn't?]
Jurq'u
mana kanchu, waway, chayqa mana kaj chullpa, inkalla.
[There isn't, my child, and it's not Chullpa, just Inka.cc]
Santiago
aaaa. . .
[Ahhhh . .. !]
Jurq'u
ri timpumanta pacha rumi t'ujyarishajta ruwasqanku, nara mas,
qhuya jinallamin kasqa,
mana imapis kanchu chay ukhupiqa.
[They set the stone exploding,dd from the time of the King, that's all,ee
it seems it's just exactly like a mine,
there's nothing inside there.ff]
Santiago
aaaa yasta tatay.
| Ahbh ... That's it, my father!]
Jurq'u
isu pikaru!
[That'sit, rascal!]
Santiago
ujta tapurikuyta rnunallaykutaj kay ripimanta,
imamantataj kay wiraquchista disgrasya qhatiy munan?
kay ripisitunta p'akikukunku ari.
[There's another thing we'd like to ask you, please, about this jeep,
why does misfortune wish to follow these wiraquchis?gg
This jeep of theirs is very badly broken.hh]
cc
ln tills and the following spereh, the Jurq'u invokes the widespread Northern Potosi conflation of the Inka
with the King of Spain, and contrasts the period of colonial mining with the pre-social ('wild', Jurq'u) Chullpa.
dd
rumi t'ujyarishajta ruwasqanku: the impression of explosive immediacy created by the onomato-
poeic t 'ujyay is reinforced by the use of the inceptive -ri-, the suffix of continuing action -ska-, and the
transformative suffix -ya; the explosion -t'uj!- was, so to speak, perpetually in the act of becoming.
ee
From Sp. nada mas, an example of the jurq'u's heavily hispanised Quechua.
ff
Not a chullpa saint, perhaps; but local mines do contain ogres such as Wari, Satands, or Lucifer,
dyawlus ('devils' Sp, diablos), supay, tiyu-tiya, pachatata-pachamama, and other semi-colonised, but still
subversive goblins; cf. Platt (1983).
gg
The plural wiraquchis here includes my assistant, Wagner Oporto, a mestizo from the valley town
of Micani.
hh
This refers to an occasion following a visit to the shrine of Pumpuri, when my jeep had a puncture
on the return journey. The tension between causality and chance presents itself—particularly as, with one
Jurq'u
isu pikaru pis!
"jisus maria anjilawarta waqaychiri" niwaychij, kristyarius!
winu, dun santyaku, glurya miriku istrilla, waturimusaj waway, awir,
imachus jucha kasqan waway, winu?
[That's it then, rascal!
Say to me "Jesus Mary Guardian Angel Protectress" Christian children!
Well, don Santiago, Glory Doctor Star! I'll go and consult, my child,
let's see!
what the fault" would have been, my child, okay?]
Santiago
awir tatay , . .
[Let's see, my father . , , ]
Jurq'u
isu pikaru pis isu! patriin waturimusun waway, winiin?
[That's it then, rascal, that's it! We'll go and ask the boss ii my child,
okay'?]
Santiago
yasta tatay.
[All right, my father.)
Jurq'u
wa! suraaj ura, wina ura kashaykichij, wawas kristyanus, winun?
[Wa! Let it be a fine hour, a good hour for you,kk Christian children,
okay?]
Santiago
yasta tatay,
i All right, my father, j
Jurq'u
wa! tukuy sunqu tukuy alma kashankichij, waway, winu?
[Wa! Be with a whole heart, a whole soul,11 my child, okay?]
spring bound up with cowhide, the vehicle appeared to be almost limping. We wanted to find out why the
puncture had happened then, and to ux.
"The concept of jucha, chosen by early catechisers to translate 'sin', more correctly denotes any moral
misdemeanour or fault of ceremonial omission.
jjI.e., Tata Pumpuri. Final stress on winu indicates missing interrogative suffix -chu.
kk kashaykichij: the 2ppl imperative + -sha- of continuous action: lit., 'be being (all of you) good time
fine time'!
Imperative has become affirmative (-nki replaces -yki): 'you are being whole heart whole soul, no9?
The phrase was contrasted by participants with iskayriyay, lit. 'to become two', i.e., 'to be undecided'.
Exit jurq'u with wings flapping
AH
ya tatay ya . . .
[Yes, my father, yes .. . !]
Santiago
"mana kanchu" ninqa.
["There's nothing", that's what he says.]
Wagner
"Santos no hay" dice, "mana kanchu" nin "chay jusk'upi".
["There are no saints", he says, "there's nothing in that hole", he says.]
Yachaj
"mana kanchu, ri timpupi urqusqanku" ninsina, manachu?
| "There's nothing, they took it out in the time of theKing,"mmI think
he said, no?]
***
From Session 3
Santiago
kunan, chay chullpa santusqa unay timpu kanmanchu? manachu kanman karqa?
[Now, would Chullpa saints have existed long ago? Wouldn't there
have been?]
Jurq'u
a tiyan a, kay santuariusmanta kayqa a, tata ri, tata ri lisyinsyamun, . . .
u- algun riya, algun timpu a chaymanta kashan waway,
jurq'uspi piru jurq'uspis kashanpunitaj kashan.
[There are, from these sanctuaries,nn father King,00 father King gave a
license ,. .
mm The phrase could be taken to refer to the removal of a pre-Christian mining huaca by Christian
authorities.
""The "sanctuaries" are the sites of major regional miracles and are the foci for great pilgrimages, or
romerias. There are said to be twelve miracles, of which one is Tata Pumpuri himself.
ooThe persons of the Inka and the Spanish monarch are conflated in Father King, tata ri. As the source
of moiety organisation and early modern civilisation, he is contrasted—as the Sun with the Moon—with
the earlier fecund time of Chullpa society. Here he is credited with incorporating the Chullpa into his New
Dispensation as licensed sources of ancient sacred power. The image of an earlier time of superabundant
fertility combined with demographic pressure is found in many other parts of the Andes; compare the age
of Wallallu Qarwinchu in the early seventeenth-century tales of Huarochiri (Salomon and Urioste 1991).
Macha and Huarochiri agree, too, that there was a yet older age, that of the lumas ( Sp. lomas 'mountains';
Macha) or of Black and Night Namca (Huarochiri), which was characterised by total darkness before the
rising of either Sun or Moon.
Another, another day, another time, then, that's when they're, from, my
child,
but in the wild places too they're there right enough.i'P]
Santiago
may jurq'upitaj kashan?
[In which wild places are they?]
Jurq'u
santuarius kashan, chullpa q'asapi kashan, waway kristyanus. . . .
[They're (in) the sanctuaries, they're in Chullpa Pass, my Christian
child. . . . 11
Santiago
aaa! chullpa q'asa . nin. , . , !
[Ahhh! (To the others:) He says Chullpa Q'asa . .. !]
Jurq'u
isu pikarun!
[That's it, rascal!]
Santiago
chayilapichu?
[Just there?]
Jurq'u
chullpa q'asapi kashallantaj a, altu qullun,
chunka iskayniyuj tanka achachilapi, chaypi kallantaj, wawas kristyanus.
[They're at Chullpa Q'asa, all right then, its high peak,
and at Twelve Grandfather Tanka,rr they're there, too, Christian
children.]
pp
The Chullpa represent the ancient pre-Christian ancestors who continue to exist in the Inner
Dimension (ukhu pacha), like the other pagan dead (sajras). . Some sanctuaries, such as Killakas, may
become wholly absorbed by the High God (altu dyus); others retain a partial (Pumpuri) or total (Tanka-
Tanka) affiliation with the Inner World.
qq
Thc Road of the Dead (alma nan) to the Western Ocean (la mar, mar qucha) is said to go through
Chullpa Pass, overlooking the salt-lick of Qharata, which leads up from the low puna where 1 was living to
the high pastures of Macha and K'ulta. There is an important unexcavated archaeological site at the pass,
with Chullpa tombs and settlement remains.
"Tanka-Tanka: one of the highest peaks in the Cordillera de los Frailes, from whose "devils"
Santiago came fleeing, according to the origin myth collected by Jose Luis Grosso. The title of
"Grandfather Twelve Tanka" evokes an ancient pagan huaca with a triple personhood which awoke
priestly paranoia in the sixteenth and seventeenth centuries. His legendary treasure was said to lie
beneath the Franciscan monastery known as La Recoleta in Sucre. The trinitarian dimension recurs in
the feet that Tanka-Tanka marks the boundary between three maximal ayllus—Macha, K'ulta, and
Qaqachaka—according to the Composition and Amojonamiento of Macha in 1646, published in Platt
(1982: app. 1). The name persists at several other sites in modern Potosi and Chuquisaca (information
from Veronica Cereceda and Olivia Harris).
***
Session 6

Jurq'u
winus nuchis wawas kristyanus!
[Good evening, Christian children!]

All
winas nuchis, tatay
[Good evening, my father!]

Jurq'u
a "sus maria anjilawarta waqaychiri" niwaychij, wawas kristyanus,
a "chunka iskayniyuj", wawas kristyanus,
a "chunka iskayniyuj chhankaquru" niwaychij, wawas kristyanus!
[Ah, say to me "Jesus Mary Guardian Angel Protectress", Christian

children,
Ah, "Twelve", Christian children,
Ah! say to me "Twelve Chhankaquru", Christian children!]
All
chunka iskayniyuj chhankaquru!
[Twelve Chhankaquru!]
Jurq'u
a "sus maria anjilawarta waqaychiri"
niwaychij, wawas kristyanus,
wa! tukuy sunquchu kankichij, wawas kristyanus?
[Ah, say to me "Jesus Mary Guardian Angel Protectress", Christian

children
Wa! Are you wholehearted, Christian children?]
All
ari, tukuy sunqu, tatay.
[Yes, wholehearted, my father.]

Jurq'u
a sumaj ura wina ura, wawas kristyanus!
[It's a fine time, a good time, then, Christian children!]

All
sumaj ura wina ura!
[A fine time, a good time!]

Jurq'u
a winu, kay kawalliruykijtapis walijllamin wasin kashan, wawas kristyanus.
[Well now, this caballero of your family is perfectly well, Christian
children!]
***
Jurq'u
salur, wawas kristyanus, a surtirisunchij kay chiqan kawiltu, sumaj kristyanus!
[Your health, Christian children, We'll be lucky with this well-made
Council, fine Christians!ss|
All
bay tatay.
[There, my father!]
Jurq'u
"quri chuqi qullqi chuqi kawiltu" niwaychij, wawas kristyanus.
a, ni tiyanchu, waway? salur, bay a muchachu!
[Say to me "Gold gold, silver gold Council" Christian children.
(Drinks noisily)
Isn't, there any more then, my child? (to the yachaj) Your health, there
then, boy! |
Yachaj
misaman, tatay, misaman!
[To the altar table, my father, to the altar table!]
Jurq'u
a parlakuychij, wawas kristyanus!
[Well, speak, Christian children!)
All
parlakushayku, tatay!
[We are speaking, my father!"]
Jurq'u
a sumaj ura wina urachu kankichij, wawas kristyanus?
[Is it a fine moment for you, a good moment, Christian children?]
ssGood Christians of the jurq'u's persuasion can expect to be blessed with good luck.
A collective affirmation ot communicative competence.
All
sumaj ura wina ura kayku, tatay!
[It's a fine moment for us, a good moment, my father!]
Jurq'u
sinu, kunan, kay walijllamin kay kawalliruyki[j]tapis tatan maman.
[But now, this caballero of yours, his father and mother are just fine.]
Wagner
imamanta, tatay, chay chullpa qullus surujchay munawayku?
[Why, my father, do those chullpa peaks want to give us altitude
sickness?]
Jurq'u
imaraykutaj chay q'uwitata asarpawajchij karqa, wawas kristyanus, awir, ichari, dun
santyaku?
[Because you should have roasted a little q 'uwauu Christian children,
let's see, isn't that right, don Santiago?]
Santiago
ari, tatay.
[Yes, my father.]
Wagner
awir, imataj?
[Let's see, what was that?]
Jurq'u
awir, insyinsyitun asarpariyman, q'ushnirichikuyman, wawas kristyanus, wawas
kristyanus,
kay wina ura allin tinipu, awir—
algun riyamanta paykuna kanku, wawas kristyanus, awir, ichari, dun santyaku?
[Let's see, I'd have quickly roasted a little incense, I'd have made some
smoke, Christian children, Christian children,
at a good and fine time like this, let's see—
they are from another age, Christian children, let's see, isn't that so,
don Santiago?]
Santiago
ari, tatay.
[Yes, rny father.]
Jurq'u
a chaypi chullpa wayra, sanju wayra, wak'aj wayran, sullu wayra tiyan,
wawas kristyanus,
UU
A resinous plant burnt as incense for the earth deities.
kapas chaywanraj tinkunkumanraj karqa,

piru ura uralla pasan, wawas kristyanus.
| Up there you find the Chullpa wind, vv the Trench wind, ww the huaca's
wind,xx the Fetus wind, yy Christian children,
They would probably have met with that,
but in a little while it passes, Christian children.]
All
ari a, tatay.
[Yes, my father.]
Jurq'u
ichapis kaymanta jina .. ,
di ripinti, awir, pilutata jina tukuykuchinman, awir,
nuqanchij imanasunman, wawas kristyanus?
[Perhaps then like.. ..
perhaps, let's see, it'd leave us like a football,zz let's see,
and what would we do then, Christian children?]
AH
ari a, tatay.
[Yes, my father.]
Jurq'u
asi pikaru a!
sumaj ura wina ura kachun, wawas kristyanus.
[So rascal, then!
Let it be a fine time, a good time, Christian children.]
vv
'Up' supplied: the chullpa (Late Intermediate) settlements and terraces are generally found on fortified
hilltops. The chullpa wind is closely related to the fierce Altiplanic winds described by Cobo in the seventeenth
century, which attack the traveller in Lipez with such force that they freeze-dry him in one gust. His
mouth-muscles tauten, making him smile, so that later travellers meet a smiling mummy on their road, which
no doubt (although no source says it) demanded its own offerings from the passer-by. It is this power of
assimilating the living to the state of mummies that characterises the winds emitted by the Chullpa, unless they
are shown respect and placated. For a fine contribution to the study of Andean winds, see Cereceda (1993).
wwS
anju: a deep ditch where the Inner Glory from the Inner Dimension (ukhupacha)a) has struck as
lightning; the place may become an object of individual devotion. It, too, can emit a fierce and deadly wind.
xx
ln Macha today, wak 'a are human-shaped rocks that come alive during the period of intense fertility
that characterises the full and new moons.
yy
The wind that blows from the place where an aborted fetus or unbaptised placenta is buried. Without
the Christian salt of baptism, they grow into a 'naked baby' (q 'am wawa or q 'am una) or 'goblin' (twinti,
Sp. duende), which kills its mother by returning to cat blood in the womb and continuing to eat until it
consumes the heart (kurasun Sp. corazon).
"Glossed as 'we'd crouch and clutch our head to our knees [to protect ourselves from the wind], like
a ball of yarn (muruq 'ujina)', The risk is that 'you'd end up like a stone shaped like a ball of yarn (tukuwaj
muruq 'u rumi jina)'.
All
sumaj ura wina ura.
| A fine time a good time.]
Jurq'u
kunari yastanamin kawiltu kashan, ijus.
[Now the Council is really over, sons.]
Santiago
yastachu?
[Is it over?]
Jurq'u
yastamin kawiltu wawas kristyanus, winun?
[The Council is really over, Christian children, okay?]
All
bay tatay.
[All right, my father.]
!
Jurq u
awir kunan rugamyintu supliku kaywan chunka iskayniyuj kuka sink'an,
jinataj winu, singanitun, q'uwita, dilantiruntawan chuqarparichinkichij,
dun santyaku, winu?
[Let's see, now a plea, a supplication, and with it Twelve Perfect Coca
Leaflets,
and likewise you sprinkle right away wine, singani, q'uwa, with its
Lead Llama,aaa
don Santiago, okay?]
Santiago
yasta tatay.
[All right, my father.]
Jurq'u
awir kupiarichinki kay chullpa wayramanpis, waway, winun?
| And do make a copybbb for that Chullpa wind, my child, okay?]
All
bay a tatay.
[There then, my father.]
aaa
The noblest form of animal sacrifice. The delanteros lead the llama troops on their long journeys to
the distant valleys.
'kupiyarichinki, from Sp. copiar 'copy'. The phrase further confirms the presence of a notarial
vocabulary in the jurq'u-lawyer's discourse.
Jurq'u
chaymanta tarillanman, wawas kristyanus.
[Afterward he'd just find it, Christian children.]
All
bay a tatay.
[There then, my father.]
Jurq'u
asi pikaru a!
jinalla, wawas kristyanus,
[ So, rascal, then!
That's it now, Christian children.]
All
bay tatay.
[ All right, my father.]
Jurq'u
ari, yastamin kawiltu kashan, yastamin kashan, wawas kristyanus, ijus,
a ratukamana, wawas kristyanus, iju,
yasta wilajap'ichikuychij , . .
[Yes, the Council's really over now, it's really over, Christian children,
sons,
till soon, Christian children, son,
right now, light the candle . . . |
Exitjurq 'u with "wings flapping
Wagner
ratukama, tatay!
[Till soon, my father!]
Yachaj
wila jap'ichiychij, tatay!
(to Santiago)
[Light the candle, my father! |
Commentary on the Text

The condor's speech can be analysed into types of speech act—greeting, self-in-
troduction, formulaic repetition of sacred names or liturgical phrases, incitation to
consult, terminating remarks, etc.—which are repeated in a regular sequence from
one session to the next. For example, the self-presentation is generally followed
by the formulaic repetition of sacred names, which in turn leads to the incitation
to consult. There is also a significant catechetical element-—invitations to the
audience to repeat his words with the phrase niwaychij, wawas kristyanus 'say to
me, Christian children'!—and the development of the whole seance across six
sessions reflects the audience's increasing confidence and ability to join in the
responses. This gradual emergence of the capacity to make easy verbal exchanges
is part of a process by which the relationship between spirit and congregation is
progressively achieved.
One feature that merits comment is the way in which inter-person directionals
(person subject + object marking) and the Quechua distinction between exclusive and
inclusive first-person plural, are used at different moments to shift between sub-groups
of speakers and addressees. I call these features of deictic reference "participant
frame-indicators" because they specify the bounds of the relevant speech group(s) at
different moments in the conversation.12
For example, participant frame-shifting allows the jurq'u sometimes to group
himself reassuringly with the congregation in opposition to Tata Pumpuri, and at other
times to group himself more loftily with his master Tata Pumpuri, as against the yachaj
and the congregation. Thus, ima niwasunchus, wawas kristyanus, winu? 'what do you
think he'll say to us, Christian children, okay'? is analysable as ni- (root 'say') +
-wa-sun- (directionai 3 p. sing. > 1 p. pi. inclusive) + -chus (dubitative), with the
-wa-sun- ('to us') establishing the jurq'u's membership of the group (congregation
plus yachaj) to be addressed by Tata Pumpuri (frame C, Figure 12-2). On the other
hand, in the phrase imatachus dijtawanqal 'I wonder what he'll dictate to me"? (not
included in the selection presented here), the replacement of ~wa-sun- with -wa-nqa
(directional 3 p. sing > 1 p. sing + 3 p. sing, future) leaves the jurq'u alone as the sole
recipient of Tata Pumpuri's message in a separate, off-stage, speech situation, from
which the yachaj and the congregation are excluded (frame E, Figure 12-2).
Figure 12-2 Participant frames (A-E) in play during the Council.

On other occasions the use of the pragmatic resources of Quechua allows the
yachaj to group himself, now with the jurq'u as opposed to the congregation, and then
again with the congregation as against the jurq'u. Thus, when the yachaj and the jurq'u
exchange messages (e.g., misarnan tatayl 'to the altar table, my father'! or dyus
pagarasunki muchachu! 'ah, thanks, boy'!), they are instantiating a frame that em-
braces them both in a clearly hierarchical relationship, but excludes everyone else
(frame D, Figure 12-2). Again, the phrase kuntistajtin atintiwasun ... 'if there's an
answer, he'll attend to us' can be analysed as kuntista- (root < Sp. 'contestar', 'answer')
+ -jti- (switch-reference nominaliser) + -n (3 p. possessive), followed by atinti- (root
< Sp. 'atender,' 'attend, pay attention to') + -wa-sun (directional 3 p. sing. > 1 p. pi.
inclusive), where the 'to us' (-wa-sun) on this occasion refers to the yachaj and the
congregation as the group to be "attended to" by the jurq'u (frame B, Figure 12-2).
The result of such manoeuvres is that, rather than a static set of participants
established in a fixed set of relationships, the internal structure of the groups in
interaction is constantly being modified according to the communicative requirements
of the moment. The ability to manipulate these features of the Quechua person system
is an important rhetorical resource in the negotiation of the emerging communication
between the parties.
Session 1
The first session shows both the failure to construct a connected exchange of sentences
and initial efforts on both sides to correct the failure. The jurq'u first appears with the
sound of wingbeats descending from the roof and repeatedly striking the improvised
altar table (misa). Blowing and whistling, the spirit greets his "Christian children" and
instructs them to address him as chunka iskayniyuj chakullani 'Twelve Chakullani'.
'Twelve Chakullani Lawyer', to give the spirit his full identification (see session 2),
is therefore giving audience to his Christian children in council (perhaps modelled on
a court of session).
But people are still taken by surprise; they do not respond to the spirit's instructions
or answer his questions directly. He tells them to pray to Jesus Mary Protector, and
repeats his own sacred name, but they still do not answer. Then, to bring them to speak,
he asks them directly: ima ofrisisunkichij, wawas kristyanusl 'What is your problem,
Christian children'?
My friend Santiago now stutters out the question prepared beforehand. A few days
before we had visited a chullpa church (chullpa inglisya), a hole in a hill which was
said to date from the time of the pre-Inka gentiles called Chullpas and to contain an
underground altar with a cross on it belonging to a chullpa saint. We had resolved to
ask the jurq'u whether there might be a chullpa saint inside.
The jurq'u then enunciates a standard exclamation, and asks whether we want to
make one of the Virgins speak the answer, or whether we want to consult a mountain
peak. Without answering him directly, Santiago repeats the question. The jurq'u
thereupon announces that he will go out and bring back something unspecified,
probably the answer.
A torrent of sacred names precedes his parting shot of "What will he tell us,
Christian children, well"? The sacred names make it clear to all what sort of a spirit
they're dealing with: he cites his patron, Lord St James of Pumpuri, the Doctor
associated with the Glory, the Star, and the Birth, and thus with the Holy Sacrament
and the body of Our Lord as incarnated in the Sun and the Host. We know, therefore,
that the speaker is a messenger from the Realm of Glory (glurya parti) and from the
patron of shamans and curers, St James of Pumpuri.
Interval 1
This includes a revealing discussion of how to talk to mountain spirits. The shaman
gives the congregation some quick indications on the correct way to address the spirit,
in preparation for the second session. These comments include a key passage of
"speech about speech", indicating the rules of linguistic behaviour which the congre-
gation must follow for communication to be achieved.
As the wingbeats fade away toward the roof of the house, the shaman goes over
the spirit's questions and tells the audience what they should have said. He instructs
them to be on the ball, to put their questions with promptness and clarity. After all,
"they don't speak like we do, they speak all in a moment". "Yes", answers Santiago,
"so quickly that one gets muddled (turwachikuri)". The shaman insists that the
questions must be well formed (allinta tapuriy 'ask well'), or else the spirit will put
an end to the council right away. He continues: "If there's an answer he'll attend to
us, if there's no answer he'll stop the council". Then, blowing a long-drawn-out
poignant whistle with his bullet, he calls the spirit back, saying first to all present,
"Talk among yourselves, talk among yourselves, anything . . . "
The most important thing of all, then, is to keep talking. While questions must be
clear, it is also essential to keep up a to-and-fro hum of conversational sound if the
spirit is not to abandon the council. And the spirit can only appear if there already
exists an expectant buzz of conversation, a formless murmuring, with no specific
semantic content at all. The condor-lawyer has to be able to interrupt, to compel the
audience to silence by the swishing wingbeats of his poncho.
So the raw material of speech—sound—must be present as the crucial support
upon which the dominant voice of the spirit-lawyer can construct itself in order then
to elicit the form of dialogue. His liturgical repetition of his own and other sacred
names is itself a ploy of the same order: the names are repeated in a continuous stream
to constitute the relevant universe of sacred referents, but also to give the audience a
clear model which they can repeat and thus generate the dialogic text of the council.
Session 2
The second session is more successful. Santiago gives the jurq'u back his name
"Twelve Chakullani Lawyer", the jurq'u sips noisily the little cup of alcohol offered
him by his "boy" (muchachu, i.e., the shaman), and everyone responds finally with
"Jesus Mary" at the spirit's invitation. More expansive, the spirit again asks his
Christian children what their problem is, and Santiago repeats the question.
Now at last comes the answer. There is no chullpa saint inside the hill. The hole
belongs to what we may call (following the mythic conflation of the Inka with the
King of Spain) the "Inka-Hapsburg" period: it is "just a mine from the time of the
King".13
Santiago quickly moves on to his next question, also discussed previously. On
returning shortly before from Pumpuri, my jeep had broken a spring, and Santiago
wanted to know why misfortune (desgrasya) was pursuing us. Again the spirit says
he will leave to find out what fault (jucha) we have committed, this time by asking
his patron directly; and out he flaps.
Interval 2
An interesting feature (of which there are other examples in later sessions) is the way
the condor's answers can be glossed, expanded, and reinterpreted in discussion
between the yachaj and the congregation until a consensus is arrived at. There is room
here for members of the congregation to feed in their own views, and the conclusion
may differ from what the condor originally said. In this case we find a significant
modification: the yachaj suggests that the jurq'u had said that the chullpa saint was
taken out in the time of the King (ri timpupi urqusqanku ninsina 'They took it out in
the time of the King, I think he said'). This implies that it used to be there but was
removed in the Inka-Colonial period. The phrase urqusqanku implies deliberate action
to remove the chullpa sacred materials, which would fit very well with what is known
of priestly efforts to "extirpate idolatries" in the region, where the retrieval of the idol
was necessary before it was formally disposed of.14 In fact, the jurq'u's actual words
were that the stone had been blasted out (rumi t'ujyarishajta ruwasqanku) during
colonial mining explorations (qhuya jinallamin kasqa),15 which refers to a more
profane aspect of the colonial exploitation of Charcas.
In the relation between these two types of speech—shamanic dialogue and
"interval"—we can observe, therefore, a reflection of the way in which state and ethnic
hierarchies were themselves founded on the consensual ethic of the local community.
Vertical enunciation from positions of power (shamanic dialogue) is contrasted with
the communal chewing over of the pronouncements of authority and their reinterpre-
tation through local debate (intervals).
Session 3
In the selection here published, a short section is included from the third session on
the same subject. After hearing that there was no chullpa saint in the hole I'd entered,
I asked for a further question on whether there were such beings as "chullpa saints"
at all. Although I had already been told of certain chullpa saints (e.g., St Gabriel, St
Jerome, and St Philip), I was curious to hear what the jurq'u would say.
He obligingly brought into play another important legal term: lisyinsya (<Sp.
licencia). The King Inka was said to have given a "license" (tata ri lisinsyamuri) for
some chullpa saints to continue receiving cult in the sanctuaries, places where a
miracle has appeared, or in the "wild places" (jurq'us) above the level of cultivation,
where the archaeological remains of chullpa settlements are situated and where the
mountain spirits live. '6 The term suggests the projection of an Hispanic legal discourse
onto the relationship between Inkas and Chullpas.
Session 6
We now jump forward to the last session, which is by far the most successful. This
can be seen by the way everyone joins in sending back the responses to the new spirit,
Chhankaquru: first the greeting, then his name with formal title, and then two
formulaic phrases which recur constantly. In this session, we find fluent and courteous
exchange reaching a metalinguistic level. The jurq'u repeats the phrase used at the end
of the first interval by the yachaj: 'speak'! parlakuychij\ But now it does not elicit a
formless muttering in preparation for the spirit's arrival: rather, the reply is nothing less
than an affirmation of communicative competence, 'we are speaking'! parlakushaykul—
and with this the success of the council is formally affirmed (cf. note tt).
From then on, the rest is plain sailing. A few additional questions and answers are
taken as the pretext for some ritual advice on how to deal with the effects of the chullpa
peaks today (surujchi, wind, etc.), who had provided the point of departure for the
whole council. The jurq'u announces that the council is now really over, and this is
accepted without demur. Finally, the spirit gives some general ritual advice to
Santiago, and suggests he "copies" it (kupiarichinki) for the chullpa wind as well. The
similarity with the traslado of a legal document again supports the hypothesis that
legal language and ideas provide a literate framework within which the oral procedures
of the council are inscribed. It also suggests that when we analyse the structure of ritual
offerings as a text, there are good Andean reasons for doing so, rather than simply
those pertaining to a particular academic fashion.
Conclusion
We have observed various techniques by which the participants in the seance negotiate
their relationships and collaborate to create the preconditions for successful commu-
nication. The notes to the transcription merely indicate the wealth of context evoked
and negotiated by native speakers in the course of constructing dialogue. Throughout
the seance, it is the task of the yachaj to enable the congregation to communicate with
the jurq'u, while that of the jurq'u is to provide a bridge between the congregation and
Tata Pumpuri. The first is the pre-condition for the second. Emergent communication
with the jurq'u is therefore a way of smoothing people's relationship with the most
powerful local representative of church and state. While Tata Pumpuri is the divine
doctor who carries life and death in his hands, the jurq'u combines the roles of notary,
parish priest, and family practitioner: he elicits speech and relays it to the great regional
specialist in Pumpuri, returning with words of knowledge, comfort, and practical
advice.
We can see the council, then, as being modelled to an important degree on the
consultation by indian clients with a local mestizo lawyer (abogado), who is in touch
with the higher levels of the judiciary (Pumpuri). These higher levels dictate their
messages to the condor-lawyer as their intermediary.17 And the whole structure is

embraced by the form of the state: hence, when the spirit has to go to England to find
out if my family are well, he says he will be going to talk with uj istaru, 'another state',
who as representative of all those residing in its jurisdiction will give audience and a
reply to the condor's question.
The underpinning of the Bolivian church and state with the medical authority of
Tata Pumpuri and his messengers is coherent with the theocratic nature of the state in
which Andean runa conceive themselves to be living. It also legitimises the existence
of this state in a way that provides a touchstone with which to judge the performance
of any real, existing government. For the unifying concept that relates the different
aspects is that of Law—natural, human, and divine. Creating a chain of communica-
tion with these ideal forms of church and state can compensate in part for the
deficiencies of those concrete beings who incarnate local government and religion at
any given time.
Our analysis has led us, therefore, to appreciate the powerful forms of legitimation
that could be constructed for the colonial and Creole States: the colonial order could
even penetrate and occupy Andean creative and imaginative spaces that might be
thought to be immune to such domination. But, at the same time, domination is
legitimised through negotiation during sessions and reformulation during intervals.
This phenomenon confirms—at the levels of oral performance and the interactive
construction of communication—the existence of an ideal "pact" between communi-
ties and state, which established the normative context for coexistence and mediation
within a multi-ethnic political order.
In fact, of course, these mechanisms of legitimation have broken down repeatedly
during the long and agitated history of Northern Potosi. To take just one example, we
have evidence that in 1927 the mountain spirits supported the indians in their rebellion
against hacienda penetration, and one mountain peak called Condornasa was even fed
the remnants of a particularly abusive hacendado who had been killed by the indians
(Platt 1982; Rivera Cusicanqui and THOA 1992). Further study of Macha shamanism
could therefore include an oral consultation with Condornasa, among other important
peaks. This would lead us to understand better the points of rupture and inversion in
these oral and literate forms of social legitimation, which emerge at key moments of
economic crisis and political unrest.
ACKNOWLEDGMENTS Parts of this text were presented at the 1991 conference "Textuality in
Amerindian Cultures: Production, Reception, Strategies" held at the Institute of Latin American
Studies, University of London; at the 2nd International Conference on Ethnohistory (Coroico
1991); at the conference on "Andean Music and Cosmology" (Berlin 1992); and at the St
Andrews conference on "Context" (1994). My thanks to the other participants in these meetings
for their comments. The text discussed was transcribed and translated with the help of Santiago
Carvajal and Wagner Oporto. My thanks to Primitivo Nina, Quechua language assistant at the
University of St Andrews, and to Spanish honours students taking Amerindian Studies during
Martinmas Term 1991 for much stimulating discussion of both text and translation.
Notes
1. This first visit was part of a more extended series of visits to other parts of Macha,
beginning with a year's residence (June 1970-June 1971) in the valley parish of San Marcos de
Miraflores and three months on the Puna near Castilloma in June-August 1971. Later brief visits
to Pumpuri were made in 1982 and 1985.
2. Studies of the Pumpuri cult are currently in preparation by Jose Luis Grosso, Uni-
versidad del Valle (Cali), and Virginie Royer de Vericourt, Institut des Hautes Etudes de
1'Amerique Latine (Paris). The loss of St James's horse is a feature in the origin myth of the
cult: St James appears in Pumpuri fleeing from the "devils" of the great mountain of Tankatanka;
see Grosso (1995).
3. In Jujuy, Bolivian migrants have formed confraternities dedicated to the local cult of
Tata Pumpuri.
4. Grosso (1995) has shown convincingly that Tata Pumpuri's composite formation
includes San Cosme, patron saint of doctors, as well as San Bartolome and Santiago.
5. As well as "devils" of various classes (dyawlus, supay, sajras, tiyu-tiya, sanjux, lugaris,
q 'ara wawas, etc.), the Inner World contains Chullpas and Inkas; the Roots of Tata Pumpuri as
of other Mountains and Towers; and the Holy Spirit, which is about to emerge—some say in
the year 2000—to initiate a new age after the overturning of the present world. In some ways,
Tata Pumpuri seems to be holdi ng the fort for believers duri ng the present age of the neo-liberal
Anti-Christ (Platt 1988). Grosso (1995) reports that the relation between Inner and Outer is also
present in Tata Pumpuri's image, which is said to consist of an original miracle in stone later
covered in stucco and wood "by a painter" so that St James should not leave Pumpuri. The
power of the miracle is thus concealed within the visible image, and Grosso rightly compares
this practice with the concealment ofhuacas within Christian images denounced by the Second
Council of Lima in 1567, as well as by the Synod of Charcas in 1773.
6. The Jesuit lexicographer of Aymara, Ludovico Bertonio, translates 'arquebus' as illapa,
elsewhere the name of the pre-Hispanic lightning god (Bertonio 1956 [1612], 1:66). In
pre-Hispanic times, the lords of Qaraqara went to war on litters (lampa or andd) carried on the
shoulders of their vassals, and bearing a sling with which they dispensed bolts on the model of
Illapa. See Tristan Platt, et al., in press; especially the presentation to part 1.
7. For a description of the formation and linguistic practices of an Aymara shaman, see
Huanca(1989).
8. The flights of Andean shamans, in the forms of condors, falcons, swifts, and other birds,
are known from pre-Hispanic times; see Salomon and Urioste (1991: 88-89).
9. Ignacio Mosques, now dead, was from moiety Majasaya, minor ayllu Wakhuata,
minimal ayllu or cabildo Milluri, whose territory begins on the other side of the river and is
visible from Liq'unipampa.
10. This passage, from the beginning of the paragraph to the entry of the mountain spirit,
is based on a Spanish version (Arica 1975) of my notes written down in English the day after
the council (August 1971).
11. In another council I attended in 1978, a different shaman summoned the jurq'us by
ringing a hand bell like an altar boy at mass.
12. For discussion of the various types of frame and sub-frame invoked by speakers in the
process of negotiating different interactive relationships, see Kendon (1992). In the Macha
council, Quechua linguistic resources that can be heard in the darkness stand in for the visible
and gestural manoeuvres referred to by Kendon.
13. For Andean approaches to the periodisation of historical time, see Platt 1988 and 1992
and, most recently, Harris 1995.
14. See, for example, the activities of padre Hernan Gonzalez de la Casa, priest in Macha,
Toropalca, and Quilaquila in the 1570s; in Platt et al., in press (primera parte, documento 1).
15. Both yachaj and jurq'u use the reportative past -sqa, indicating the indirect nature of
the evidence for their different opinions, although once proposed the opinion may quickly
become treated as established fact, and the more positive form of the verb adopted (e.g., the
jurq'u's mana imapis kanchu).
16. lisinsya 'permission' may also be used to address the altar table when asking permis-
sion to make libations—e.g., misa, lisinsyaykiwan, ch'allarisun . . . 'altar table, with your
permission, we will pour libations' . . .
17. In another council (1978), not only does the condor at one point say, imatachus
dijtawanqa '1 wonder what he will dictate to me'!, he even refers to St James as dijtatur' dictator',
a technical term from sixteenth-century notarial vocabulary (cf. Gonzalez Echevarria 1990).
Other suggestive terms in modern Macha Quechua include convenio or acuerdo to describe the
"legal" relationship between Pumpuri and his devotees.
GLOSSARY
Symbols and abbreviations

Qu. Quechua
Ay. Aymara
Sp. Spanish
var. variant form
pi. plural
< derived from
apu (pi. apukuna) Qu Mountain deity (cf.jurq 'u; awki).

artefacto Sp. Handmade miniature goods used in ritual.
awaq mamacha Qu. Female saint patroness of weavers.
awayu Ay. Woman's mantle.
awki Qu. Hill spirit (cf.jurq 'u; apu).
ayllu Qu., Ay. Unit of social organisation (cf. sayd).
ayni Qu. Principle of reciprocity in labour and ritual obligations.
bayeta (wayita) Qu., Ay. (< Sp.) Homespun plainwoven woollen cloth.
cabildo (kawiltu) Sp., Qu. Municipal council; rural tribute-paying group; seance
during which the yachaj consults the mountain spirits.
cacica Sp. Female political leader (cf. cacique).
cacique Sp. Male political leader (cf. cacica).
ch 'isiraya Qu. Space where rituals are conducted, located at entry points to shrines
and settlements.
227
228 GLOSSARY
chicha Sp, Maize beer.

chinu Ay. Warp threads on the loom; knot-string record (cf. kipu).
chullpa Ay., Qu. Pre-Hispanic ancestors; burial places of pre-Hispanic ancestors (cf.
gentil; machu).
compadrazgo Sp. Ritual kinship.
criollo Sp. Inhabitant of Spanish America of European descent (cf. mestizo;
runa).
despacho Qu. (<Sp.) Burnt offering to the gods (cf. misa).
gentil Sp. Spirit of ancestors (cf. machu; chullpa)
huaca (wak'a) Qu. Sacred object or site.
inqa (var. inqaychu) Qu. Small stone with powerful supernatural properties.
istrilla Qu. (< Sp.) Star that guides the individual's destiny.
janaj pacha (hanaq pacha) Qu. Upper world (cf. ukhu pacha; kay pacha),
jurq 'u Qu. Mountain spirit (cf. apu; avoid).
juturi Ay. Hollows in the land thought to lead from the inner world (cf. ukhu
pacha).
kancha Qu., Ay. Animal corral.
kay pacha Qu. This world, this place; the domain of the living in the here-and-now
(cf. ukhu pacha).
kipu (quipu) Qu. Inka knot-string record (cf. chinu}.
laceadores Sp. Horsemen practised in the use of the lasso.
lista Ay. Stripe in woven poncho (cf. soya; pampa; salta).
machu Qu. Spirit of ancestors (cf. gentil; chullpa).
mallku Ay. Male.
mamacha (var.mmamita) Qu. Female deity; female saint (cf. awaq mamacha).
memorista Sp. Community historian.
mestizo Sp. Person of perceived mixed European and Amerindian descent (cf.
criollo; runa),
misa Qu. (< Sp.) Ritual table; ritual offering (cf. despacho).
misal Qu. (< Sp.) Quartz crystal used in ritual.
mita (mit 'a) Qu. Labour tribute under the Inka regime.
mitmaq Qu. Colonist from the Andean region.
nacion Sp. Community delegation to the pilgrimages in Southern Peru; confedera-
tion of indigenous peoples.
pachakuti Qu. World turning; cataclysmic moment when a former order ends and a
new one begins.
pachatnama Qu. Female earth deity (cf. wirjin tayka).
pampa Qu., Ay. Flat area of land; unpatterned area of woven cloth (cf. salta; saya;
lista),
GLOSSARY 229
paqo (paqu) Qu. Ritual specialist (cf. yachaj).

pukllay Qu. To play or act; also used to refer to bullfights and ritual play.
puna Qu., High moorland typically 3,600 metres above sea level in the Southern
Andes.
reduction Sp. Colonial population resettlement.
resguardo Sp. Indigenous land legitimised by colonial-era titles.
runa Qu. Quechua-speaking person of indigenous cultural origin (cf. mestizo; crio-
llo).
salta Ay. Patterned area of woven cloth (cf. pampa; saya; lista).
soya Ay. Single-coloured area of woven poncho (cf. pampa; salta; lista); unit of
territorial or social organisation (cf. ayllu); lit. 'standing up'.
t 'alla Ay. Female.
tata Qu. Term of address for adult male; male saint (cf. taytacha).
taytacha Qu. Icon representing Christ (cf. tata).
tierrakuna (var. tirakuna) Qu. (< Sp. + Qu. plural suffix) Powerful, animate places
in the landscape.
tocapu (tocapu) Qu. Abstract design in pre-Columbian textiles.
tullqa Ay. Brother-in-law; a ritual role in fiestas in Northern Potosf, Bolivia.
ukhu pacha (uku pacha) Qu. Inner realm; the underworld; domain of spirit beings
(cf. juturi; kay pacha; janaj pacha).
ukuku Qu. Bear figure in Southern Peruvian fiestas.
uywiri Ay. Herder; provider.
visitador Sp. Colonial inspector of population and taxes.
wayno (var. waynu) Qu. Type of song with characteristic rhythm (cf. wayfiu).
waynu Ay., Qu. Term designating an animal marking ceremony; type of song with
characteristic rhythm (cf. wayno).
wirjin tayka Ay. Female earth deity (cf. pachamama).
yachaj (yachaq) Qu. Shaman; ritual specialist (ci.paqo).
yumbo Sp. Lowland Indian (idiomatic).
BIBLIOGRAPHY
Archives
Avila, Francisco de. 1608. Tratado y relacion de los errores, falsos Dioses, y otras supersticiones,
y ritos diabolicos en que viuian antiguamente los yndios de las Prouincias de Huarocheri,
Mama, y Chaclla...MS. 3169. BibliotecaNacional, Madrid.
Notaria Primera de Ipiales (NP/I).
Notaria Primera de Pasto (NP/P).
Tradiciones de Huarochirf. c. 1608. Runa yndio niscap Machoncuna naupapacha...MS 3169.
Biblioteca Nacional, Madrid. [See also Trimborn 1967; Taylor 1987; Adelaar 1988;
Salomon and Urioste, 1991].
Primary and Secondary Sources

Abercrombie, Thomas. 1986. The Politics of Sacrifice; an Andean Cosmology in Action. Ph.D.
thesis. Universily of Chicago.
Acosta, Antonio. 1987. Francisco de Avila Cusco 1573-Lima 1647. In Taylor, Gerald. 1987a,
pp. 552-616.
Adelaar, Willem F. H. 1982. Caracteristicas del quechua de Pacaraos. In Aula Quechua, ed.
Rodolfo Cerron-Palomino, pp. 19-33. Lima: Ediciones Signo Universitario.
. 1988. Het boek van Huarochirf, Mythen en riten van het oude Peru. Amsterdam:
Meulenhoff.
. 1990. The role of quotations in andean discourse. In Unity in Diversity: Papers
Presented to Simon C. Dik on His 50th Birthday, ed. Harm Pinkster and Inge Genee,
pp. 1-12. Dordrecht: Foris. Also in (1991) Past, Present and Future: Selected Papers on
Latin American Indian Literatures including the VIII International Symposium, ed.
Mary H. Preuss, pp. 161-66. Culver City, Calif.: Labyrinthos.
. 1993. La categorfa del plural inclusive en las lenguas americanas: interpretation y
realidad. In Espana: ^Ruptura 1492? ed. Maximiliaan Kerkhof, Hugo de Schepper, and
Otto Zwartjes, pp. 207-216. Amsterdam: Rodopi.
231
232 BIBLIOGRAPHY
. 1994a. Laexpresion deconceptos abstractos y generates en quechua: vision diacronica.

In Andean Oral Traditions: Discourse and Literature/Tradiciones orales andinas: Dis-
curso y literatura, ed. Margot Beyersdorff and Sabine Dedenbach-Salazar Saenz, pp.
1-20 (Bonner Araerikanistische Studien, 24). Bonn: Holos.
1994b. A grammatical category for manifestations of the supernatural in early colonial
Quechua. In Language in the Andes, ed. Peter Cole, Gabriella Hermon, and Mario Daniel
Martin, pp. 116-25. (Occasional Monographs in Latin American Studies, No. 4).
Newark: University of Delaware, Latin American Studies Program.
1994c. ^Hasta donde llega la inflexion? Criterios para una clasificacion interna de los
sufijos verbales en quechua de Tarma/norte de Junfn. In Estudios de lengua y cultura
amerindias I, Actas de las II Jornadas internacionales de lengua y cultura amerindias,
ed. Julio Calvo Perez, pp. 65-83. Universidad de Valencia: Departamento de Teoria de
los Lenguajes.
. 1994d. La procedencia dialectal del manuscrito de Huarochiri en base a sus
caracterfsticas linguisticas. Revista Andina 12(1): 137-54.
Adorno, Rolena. 1986. Guaman Poma: Writing and Resistance in Colonial Peru. (Latin
American Monographs, No. 68). Austin: University of Texas Press.
1989. Cronista y pnncipe. La obra de don Felipe Guaman Poma de Ayala. Lima:
Pontificia Universidad Catolica del Peril
, ed. 1982. From Oral to Written Expression: Native Andean Chronicles of the Early
Colonial Period. Syracuse, N.Y.: Syracuse University, Maxwell School of Citizenship
and Public Affairs.
Aijmcr, Goran. 1987. The cultural nature of ritual and myth. In Symbolic Textures: Studies in
Cultural Meaning, ed. Goran Aijmer, pp. 1-22. Gothenburg Studies in Social Anthro-
pology, 10. Gothenburg: University of Gothenburg.
Alberti, Giorgio, and Enrique Mayer. 1974. Reciprocidad e intercambio en los Andes peruanos.
Lima: Institute de Estudios Peruanos.
Allen, Catherine J. 1982. Body and soul in Quechua thought. Journal of Latin American Lore
8(2): 179-96.
. 1983. Of bear-men and he-men: Bear metaphors and male self-perception in a Peruvian
community. Latin American Indian Literatures Journal 7(1): 38-51.
. 1984. Patterned time: The mythic history of a Peruvian community. Journal of Latin
American Lore 10(2): 151-74.
1988. The Hold Life Has: Coca and Cultural Identity in an Andean Community.
Washington, D.C.: Smithsonian Institution Press.
Anderson, Arthur J.O., Frances Berdan, and James Lockhart, eds. 1976. Beyond the Codices: A
Nahua View of the Colonial World. Berkeley- Los Angeles: University of California Press.
Anderson, Benedict. 1983. Imagined Communities: Reflections on the Origin and Spread of
Nationalism. London: Verso.
Anderson, Lloyd B. 1986 [1982], Evidentials, paths of change and mental maps: Typologically
irregular asymmetries. In Evidentiality: The Linguistic Coding of Epistemology, ed.
Wallace Chafe and Joanna Nichols, pp. 273-312 (Advances in Discourse Processes,
vol. 20). Norwood, N.J.: Ablex Publishing.
Aranguren Paz, Angelica. 1975. Las creencias y ritos magicos-religiosos de los pastores
punenos. Allpanchis 8: 103-132.
Arguedas, Jose Maria. 1941. Yawar Fiesta. Lima: Editorial Horizonte.
. 1955. Taki parwa y la poesfa quechua de la republica. Letras Peruanas 12(8): 73-75.
. 1978 [1956]. Deep Rivers, tr. Frances Horning Barraclough. Austin: University of Texas
Press. Originally published as Los Rfos Profundos. Buenos Aires: Editorial Losada.
Arnold, Denise Y. 1988. Matrilineal practice in a patrilineal setting: Ritual and metaphors of
kinship in an Andean ayllu. Ph.D. thesis. University of London.
BIBLIOGRAPHY 233
. 1992a. La casa de adobes y piedras del Inka: genero, memoria y cosmos en Qaqachaka. In
Hacia un orden andino de las cosas: Tres vistas de los Andes meridionales, ed. Denise Arnold,
Domingo Jimfnez, and Juan de Dios Yapita, pp. 31-108. La Paz: HISBOL/ILCA.
. 1992b. At the heart of the woven dance-floor: The waynu in Qaqachaka. Iberoamericana
16(3/4): 21-66. SjDecial Issue: Literaturas autoctonas en America Latina, ed. M. Lienhard.
. 1993. Adam and Eve and the red-trousered ant: History in the southern Andes. Travesia:
Journal of Latin American Cultural Studies 2(1): 49-83.
. 1994. The "flower in the corral": An Andean textile guide for interweaving worlds. In
Fertility: Images and Icons, Rites and Rituals. Catalogue for the exhibition at the
University Gallery, University of Essex, May 16-June 3, 1994, pp. 20-25.
, and Juan de Dios Yapita. 1992. Fox talk: Addressing the wild beasts in the southern
Andes. Latin American Indian Literatures Journal 8(1): 9-37.
, and Juan de Dios Yapita. 1994. Traditional Maternity in Qaqachaka, Oruro, Bolivia.
Preliminary Report to the European Community. La Paz: ILCA
, and Juan de Dios Yapita with Cipriana Apaza M. In press. Mama trama y sus crias:
Analogies de la produccion de la papa en los textiles de Chukinapi, Bolivia. In Mama
Melliza y sus crias: Una antologfa de la papa, ed. Denise Y. Arnold and Juan de Dios
Yapita. La Paz: HISBOL/ILCA.
, Domingo Jimenez A., and Juan de Dios Yapita. 1991. Scattering the seeds: Shared
thoughts on some songs to the food crops. Amerindia 16: 105-78.
, Domingo Jimenez A., and Juan de Dios Yapita. 1992. Hacia un orden andino de las
cosas: Tres vistas de los Andes meridionales. La Paz: HISBOL/ILCA.
Ascher, Marcia, and Robert Ascher. 1981. The Code of the Quipu. Ann Arbor: University of
Michigan Press.
Aveni, Anthony F., ed. 1988. New Directions in American Archaeoastronomy: Proceedings of
the 46th International Congress of Americanists (Amsterdam, Netherlands 1988), (BAR
International Series 454). Oxford: BAR.
Barriga Barahona, Oscar. 1981. Coleccion de ponchos bolivianos. Exhibition leaflet. La Paz:
MUSEF.
Barstow, Jean R. 1981. Marriage between human beings and animals: A structuralist discussion
of two Aymara myths. Anthropology 5(1): 71-88.
Barthes, Roland. 1977. Rhetoric of the image. In Image, Music, Text, pp. 32-51. New York:
Hill and Wang.
Bastien, Joseph W. 1987. Healers of the Andes: Kaliawaya Herbalists and Their Medicinal
Plants. Salt Lake City: University of Utah Press.
Benveniste, Emile. 1966. Les relations de temps dans le verbe francais. Problemes de
linguistique generale, I, pp. 237-50. Paris: Gallimard.
Bergren, Anne. 1983. Language and the female in early Greek thought. Arethusa 16(1/2): 69-95.
Bernal Villa, Segundo. 1953. Aspectos de la cultura paez: Mitologia y cuentos de la parcialidad
de Calderas, Tierradentro. Revista Colombiana de Antropologia 1(1): 279-309.
Bernales Ballesteros, Jorge, ed. 1969. Fray Calixto de San Jose Tupac Inca, procurador de indios
y la 'Exclamacion' reivindicacionista de 1750. Historia y Cultura 3: 5-35.
Bertonio, Ludovico. 1956 [1612]. Vocabulario de la lengua aymara. Facsimile edition. La Paz:
Universidad Mayor de San Andres.
. 1984 [1612]. Vocabulario de la lengua aymara. Facsimile edition. Cochabamba,
Bolivia: CERES/IFEA/MUSEF.
Beyersdorff, Margot. 1988. La adoration de los reyes magos: Vigencia del teatro religiose
espanol en el Pert andino. Cusco: Centre Bartolome de las Casas.
, and Sabine Dedenbach-Salazar Saenz, eds. 1994. Andean Oral Traditions: Discourse
and Literature/Tradiciones orales andinas: Discurso y literatura (Bonner Amer-
ikanistische Studien, 24). Bonn: Holos.
234 BIBLIOGRAPHY
Bills, Garland D. 1972. The Quechua directional verbal suffix. Papers in Andean Linguistics
1(1): 1-15.
Bloch, Maurice. 1991. Language, anthropology and cognitive science. Man 26: 183-98.
Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University
Press.
Bouysse-Cassagne, Therese. 1987. La identidad aymara: Aproximacion historica (Siglo XV,
Siglo XVI). La Paz: HISBOL/IFEA.
, and Olivia Harris. 1987. Pacha. En torno al pensamiento aymara. In Tres rcflexiones
sobre el pensamiento andino, ed. Therese Bouysse-Cassagne and Olivia Harris, Tristan
Platt, and Veronica Cereceda, pp. 11-59. La Paz: HISBOL.
, and Olivia Harris, Tristan Platt, and Veronica Cereceda. 1987. Tres reflexiones sobre
el pensamiento andino. La Paz: HISBOL.
Bricker, Victoria Reifler. 1981. The Indian Christ, the Indian King. Austin: University of Texas Press.
. 1989. El cristo indigena, el rey nativo. El substrata historico de la mitologfa ritual de
los mayas. Mexico: Fonclo de Cultura Economica.
Briggs, Lucy. 1994. El k"arik"ari en dos textos de lengua aymara: Analisis morfosintactico y
del discurso. In Andean Oral Traditions: Discourse and Literature/Tradiciones orales
andinas: Discurso y literatura, ed. Margot Beyersdorff and Sabine Dedenbach-Salazar
Saenz, pp. 161-97 (Bonner Amerikanistische Studien, 24). Bonn: Holos.
Brotherston, Gordon. 1979. Image of the New World: The American Continent Portrayed in
Native Texts. London: Thames and Hudson.
Brown, Michael. 1986. Tsewa's Gift. Magic and Meaning in an Amazonian Society. Washing-
ton D. C.: Smithsonian Institution Press.
Burke, Kenneth. 1966. What are the signs of what? A theory of entitlement. In Language as
Symbolic Action, pp. 359-79. Berkeley: University of California Press.
Calvo Lopez, Julio-Alexis. 1994. Los evidenciales en las lenguas andinas y amazonicas. In
Estudios de lengua y cultura amerindias I. ed. Julio Calvo Perez, pp. 85-94. Valencia:
Universidad de Valencia.
Cameron, Deborah, Elizabeth Frazer, Penelope Harvey, M.B.H. Rampton, and Kay Richardson.
1992. Researching Language: Issues of Power and Method. London: Routledge.
Carlson de Coombs, Heidi. 1975. La clausula relativa en quechua de Cajamarca. Datos
Etnolinguisticos no. 5. Lima: Institute Lingiiistico de Verano.
Castillo-Cardenas, Gonzalo. 1987. Liberation Theology from Below: The Life and Thought of
Manuel Quintin Lame. Maryknoll, N.Y.: Orbis.
Cereceda, Veronica. 1978. Semiologie des tissus andins. Annales. E.S.C. 33(5-6): 1017-35.
. 1986. The semiology of Andean textiles: The talegas of Isluga. In Anthropological
History of Andean Polities, ed. John V. Murra, Nathan Wachtel, and Jacques Revel, pp.
149-73. Cambridge: Cambridge University Press.
. 1987. Aproximaciones a una estetica andina: De la belleza al tinku. In Tres reflexiones
sobre el pensamiento andino, ed. Therese Bouysse-Cassagne et al., pp. 133-231. La Paz:
HISBOL.
_ 1993. Cette etendue entre 1'AItiplano et la mer...un mythe chipaya hors texte. In
Memoire de la tradition, ed. Aurore Becquelin, Antoinette Molinie, and Danielle
Dehouve, pp. 227-84. Nanterre: Societe d'Ethnologie.
Cerron-Palomino, Rodolfo. 1987. Linguistica quechua. Cusco: Centro Bartolome de las Casas.
Chafe, Wallace, and Joanna Nichols, eds. 1986. Evidentiality: The Linguistic Coding of Epistemol-
ogy (Advances in Discourse Processes, Vol. 20) Norwood, N.J.: Ablex Publishing.
Clanchy, M. T. 1979. From Memory to Written Record, England 1066-1307. London: Edward
Arnold.
Clifford, James, and George Marcus, eds. 1986. Writing Culture. The Poetics and Politics of
Ethnography. Berkeley: University of California Press,
BIBLIOGRAPHY 235
Cohen, Anthony P. 1985. The Symbolic Construction of Community. London: Tavistock.

Cole, Peter, Wayne Herbert, and Gabriella Hermon. 1982. Headless relative clauses in Quechua.
International Journal of American Linguistics 48(2): 113-24.
Colombia, Republicade (Ministerio de Gobierno). 1983. Fuero indigena: Disposiciones legales
del orden nacional, departamental y comisarial—-jurisprudencia y conceptos. Bogota:
Presencia.
Condori Chura, Leandro, and Esteban Ticona Alejo. 1992. El escribano de los caciques
apoderados/kasikinakan purirarunakan qillqiripa. La Paz: HISBOL/THOA.
Connerton, Paul. 1989. How Societies Remember. Cambridge: Cambridge University Press.
Costa, Rachel. 1972. A study of -sqa, -na, -y, and -q nominalizing suffixes in Quechua. Papers
in Andean Linguistics 1(1): 29-77.
Crickmay, Lindsey. 1994. Ancestral spirit and Andean devil: Divergence of a single concept.
Paper presented at the Institute of Amerindian Studies, University of St Andrews.
Unpublished ms.
Cummins, Thomas B, F. 1988. Abstraction to Narration: Kero Imagery of Peru and the
Colonial Alteration of Native Identity. Ph.D. dissertation. University of Califor-
nia—Los Angeles.
Cummins, Tom. In press. "Let rne see"! Reading is forthem: colonial Andean images and objects
'como es costumbre tener los caciques senores'. In Native Traditions in the Post-Con-
quest World, ed. Elizaeth H. Boone and Tom Cummins. Washington D.C.: Dumbarton
Oaks Research Library and Collection.
De Certeau, Michel. 1984. The Practice of Everyday Life. Berkeley: University of California Press.
. 1986. Heterologies: Discourse on the Other. Minneapolis: University of Minnesota
Press.
Dedenbach, Sabine. 1982. Versuche zur Rekonstruktion der in den Quechua-Quellen ver-
wendeten Dialekte. Appendix to: Zur Rekonstruktion des Ackerbauwortschatzes der
Inkazeit. pp. 164—78. Unpublished German version of MA thesis. University of Bonn.
Dedenbach-Salazar Saenz, Sabine. 1993. El desarrollo del sufijo -sqa a la luz de las fuentes
quechuas del siglo XVIII. Paper presented at the "V Coloquio International: El Siglo
XVIII en los Andes", Paris, April 1993. Unpublished ms.
. 1994. El arte verbal de los textos quechuas de Huarochirf (Peru, siglo XVII) reflejado
en la organization del discurso y en los medios estilisticos. In Andean Oral Traditions:
Discourse and Literature/Tradiciones orales andinas: Discurso y literatura, ed. Margot
Beyersdorff and Sabine Dedenbach-Salazar Saenz, pp. 21-49 (Bonner Amerikanistische
Studien, 24). Bonn: Holos.
, Utta von Gleich, Roswith Hartmann, Peter Masson, and Clodoaldo Soto Ruiz, tr. and
ed. 1994. (3rd ed.) "Rimaykullayki". Unterrichtsmaterialien zum Quechua Ayacuchano.
Based on Quechua—Manual de Ensenanza by Clodoaldo Soto Ruiz (Lima 1979).
University of Bonn. Berlin: Reimer Verlag.
Deere, Carmen Diana. 1977. Changing social relations of production and Peruvian peasant
women's work. Latin American Perspectives 4(12): 48-69.
Desrosiers, Sophie. 1988. Les techniques de tissage ont-elles un sens? Un mode de lecture des
tissus andins. Techniques et Culture 12: 21-56.
Dorris, Michael. 1987. Indians on the shelf. In The American Indian and the Problem of History,
ed. Calvin Martin, pp. 98-105. Oxford: Oxford University Press.
Douglas, Carrie. 1984. Toro muerto, vaca es: An interpretation of the Spanish bullfight.
American Ethnologist 11: 242-58.
Dransart, Penny. 1988. Continuidad y cambio en la production textil aymara traditional.
Hombre y Desierto: Una Perspectiva Cultural 2: 41-57.
. 1991 Fibre to Fabric: The Role of Fibre in Camelid Economies in Pre-Hispanic and
Contemporary Chile. D. Phil, thesis. University of Oxford.
236 BIBLIOGRAPHY
Durand-Florez, Luis, ed. 1980-82. Coleccion documental de bicentenario de la revolution

emancipadora de Tupac Amaru: Vol. 1, Documentos varios del archive general de
Indias; vol. 2, Descargos del obispo del Cuzco; vols. 3-5, Los procesos a Tupac Amaru
y sus companeros. Lima: Comision National del Bicentenario de la Rcbelion
Emancipadora de Tupac Amaru.
Duranti, Alessandro, and Charles Goodwin. 1992. Rethinking Context: Language as an Inter-
active Phenomenon. Cambridge: Cambridge University Press.
Duviols, Pierre. 1979. Un symbolisme de 1'occupation, de l'amenagement et de Fexploitation
de 1'espace: Le monolithe "huanca" et sa fonction dans les Andes prehispaniques.
L'Homme 19(2): 7-31.
Duviols, Pierre, and Cesar Itier, eds. 1993. Relacion de antiguedades deste reyno del Piru.
Edition facsimilar y transcription paleografica del codice de Madrid, pp. 127-78
(Travaux de l'IFEA/74; Archives de Historia Andina/17). Lima and Cusco: Institut
Frangais d'Etudes Andines/Centro Bartolome de las Casas.
Earls, John, and Irene Silverblatt. 1976. La realidad fisica en la cosmologia andina. Paris:
Proceedings of the 42nd Congress of Americanists 4: 299-325.
Fabian, Johannes. 1983. Time and the Other: How Anthropology Makes its Object. New York:
Columbia University Press.
Fairclough, Norman. 1989. Language and Power. London: Longmans.
. 1992. Discourse and Social Change. Cambridge: Polity Press.
Fernandez, James. 1985. Exploded worlds—-Text as a metaphor for ethnography (and vice
versa). Dialectical Anthropology 10: 15-26.
Ferrell R., Marco A. 1994. La escritura incaica en Waman Puma. In II Congreso Nacional de
Investigaciones Linguistico-filologicas. Lima: Peru.
Flores Ochoa, Jorge. 1977a. Pastores de la puna: uywamichiq punarunakuna. Lima: Institute
de Estudios Peruanos.
. 1977b. Enqa, enqaychu, illa y khuya rumi. In Pastores de la puna: uywamichiq
punarunakuna, ed. Jorge Flores Ochoa, pp. 211-37. Lima: Institute de Estudios Peruanos.
. 1988. Llamichos y paqocheros: pastorcs de llamas y alpacas. Cusco: Centre Bartolome
de las Casas.
Floyd, Rick. 1994. The Wanka reportative as a radial category: A study in prototypes. In
Language in the Andes, ed. Peter Cole, Gabriella Hermon, and Mario Daniel Martin,
pp. 151-89. (Occasional Monographs in Latin American Studies, No. 4). Newark:
University of Delaware, Latin American Studies Program.
Foley, John Miles. 1986. Introduction. In Oral Tradition in Literature: Interpretation in Context,
ed. John Miles Foley, pp. 1-18. Columbia: University of Missouri Press.
Foucault, Michel. 1972. The Archaeology of Knowledge. London: Tavistock.
Fox, Richard, ed. 1990. Nationalist Ideologies and the Production of National Cultures.
(American Ethnological Society Monograph Series, No. 2).
Franquemont, Edward M., Christine Franquemont, and Billie Jean Isbell. n.d. The unfold-
ing of symmetries: The structuring of Andean cloth. Cornell University. Unpub-
lished ms.
Fraser, Valerie. 1982. Hierarchies and rules of materials in building and representation. In The
Other America: Native Artifacts from the New World, ed. Valerie Fraser and Gordon
Brotherston, pp. 41-55. Colchester: University of Essex/Museum of Mankind.
Friedman, Jonathan. 1987. Beyond otherness or the spectacularization of anthropology. Telos
71: 161-70.
Garcilaso de la Vega, El Inca. 1963 [1609]. Los comentarios reales de los Incas. Biblioteca de
Autores Espanoles, vols. 133-35. Madrid: Atlas.
Gavilan V. Vivian, and Liliana Ulloa T. 1992. Proposiciones metodologicas para el estudio de
los tejidos aridities. Revista Andina 10(1): 107-34
BIBLIOGRAPHY 237
Gee, James Paul. 1988. The legacies of literacy: From Plato to Freire through Harvey Graff.
Harvard Educational Review 38(2): 195-234.
Geertz, Clifford. 1973. Thick description: Toward an interpretive theory of culture. In The
Interpretation of Cultures: Selected Essays by Clifford Geertz. New York: Basic
Books.
Getzels, Peter, and Harriet Gordon. 1985. In the Footsteps of Taytacha. Film. Watertown, Mass.:
Documentary Educational Resources.
Getzels, Peter, Harriet Gordon, and Penelope Harvey. 1990. The Condor and the Bull: A
Documentary Film. London: National Film and Television School.
Gillis, John, ed. 1994. Commemorations: The Politics of National Identity. Princeton, N.J.:
Princeton University Press.
Gisbert, Teresa, Silvia Arze, and Martha Cajias. 1987. Arte textil y mundo andino. La Paz:
Gisbert.
Gisbert, Teresa. 1980. Iconografi'a y mitos en el arte. La Paz: Gisbert.
Goncalez Holgum, Diego de. 1952 [1608]. Vocabulario de la lengua general de todo el Peru
Hamada Qquichua o del Inca. 2nd ed., ed. Raul Porras Barrenechea. Lima: Universidad
Nacional Mayor de San Marcos.
Gonzalez Echevarria, Roberto, 1990. Myth and Archive: A Theory of Latin American Narrative.
Cambridge: Cambridge University Press.
Goody, Jack. 1977. The Domestication of the Savage Mind. Cambridge: Cambridge University
Press.
. 1986. The Logic of Writing and the Organization of Society. Cambridge: Cambridge
University Press.
, and Watt, Ian P. 1963. The consequences of literacy. Comparative Studies in Society
and History 5: 304-45.
Gose, Peter. 1986. Sacrifice and the commodity form in the Andes. Man 21: 296-310.
Gow, David. 1974. Taytacha Qoyllur Rit'i. Allpanchis 7: 47-100.
. 1976. The Gods and Social Change in the High Andes. Ph.D. thesis. University of
Wisconsin, Madison. Ann Arbor, Mich.: University Microfilms.
, and Rosalind Gow. 1975. La alpaca en el mito y ritual. Allpanchis 8: 141-64.
Gow, David D. 1980, The roles of Christ and Inkarri in Andean religion. Journal of Latin
Gow, Rosalind, and Bernabe Condori. 1982. Kay Pacha. Cusco: Centro Bartolome de las
Casas.
Graff, Harvey. 1979. The Literacy Myth: Literacy and Social Structure in the Nineteenth-Cen-
tury City. New York: Academic Press.
Griffiths, Gareth. 1994. The myth of authenticity: Representation, discourse and social practice.
In De-scribing Empire: Post-Colonialism and Textuality, ed. Chris Tiffin and Alan
Lawson, pp. 70-85. London: Routledge.
Grosso, Jose Luis. 1995. La suerte de lo andino, sus saberes y poderes. Adivinacion y mestizaje
en el Norte de Potosi. Master's thesis. Universidad del Valle, Cali.
Gruzinski, Serge. 1985. La memoria mutilada: Construction del pasado y mecanismos de la
memoria en un grupo otomi de la mitad del siglo XVII. In La memoria y el olvido.
Segundo Simposio de Historia de las Mentalidades, pp. 33-46. Mexico: Instituto
Nacional de Antropologia e Historia.
Guaman Poma de Ayala, Felipe. 1980 [1615]. El primer nueva coronica y buen gobierno, ed.
John V. Murra and Rolena Adorno. Mexico: Siglo Veintiuno.
Gutmann, Margit. 1994. Vom Fuchs und vom Rebhuhn. Einige Merkmale des Erzahlens
in der Q u e c h u a - K u l t u r Perus. In Soziolinguistik und Sprachgeschichte:
Querverbindungen, ed, Gabriele Berkenbusch and Christine Bierbach, pp. 147-74.
Tubingen: Gunter Narr.
238 BIBLIOGRAPHY
Hall, Stuart. 1993. Cultural identity and diaspora. In Colonial Discourse and Post-Colonial
Theory: A Reader, ed. Patrick Williams and Laura Chrisman, pp. 392-403. New York:
Harvester.
Hallpike, Christopher. 1979. The Foundations of Primitive Thought. Oxford: Clarendon.
Hanks, William. 1989. Text and textuality. Annual Review of Anthropology 18: 95-127.
Hanks, William F. 1993. Language and discourse in colonial Yucatan. Unpublished ms.
Harbsmeier, Michael. 1984. Early travels to Europe: Some remarks on the magic of writing. In Europe
and Its Others, ed. Francis Baiter et al., vol. 2, pp. 72-88. Colchester: University of Essex.
Hardman, Martha James. 1986. Data-source marking in the Jaqi languages. In Evidentiality:
The Linguistic Coding of Epistemology, ed. Wallace Chafe and Joanna Nichols, pp.
113-36 (Advances in Discourse Processes, vol. 20). Norwood, N.J.: Ablex Publishing.
Harris, Olivia. 1982. The dead and the devils among the Bolivian Laymi. In Death and the
Regeneration of Life, ed. Maurice Bloch and Jonathan P. Parry, pp. 45-73. Cambridge:
Cambridge University Press.
. 1986. From asymmetry to triangle: Symbolic transformations in Northern Potosi. In
Anthropological History of Andean Polities, ed. John V. Murra, Nathan Wachtel, and
Jacques Revel, pp. 260-79. Cambridge: Cambridge University Press.
. 1994. Condor and bull: the ambiguities of masculinity in Northern Potosi. In Sex and
Violence. Issues in Representation and Experience, ed. Penelope Harvey and Peter Gow,
pp. 40-65. London: Routledge.
1995 'The coming of the white people': Reflections on the mythologisation of history
in Latin America. Bulletin of Latin American Research 14(1): 9-24.
Harrison, Regina. 1989. Signs, Songs and Memory in the Andes: Translating Quechua Lan-
guage and Culture. Austin: University of Texas Press.
Hartmann, Roswith, and Udo Oberem. 1981. Quito: Un centre de education de indigenas en el
siglo XVI. In Contribucoes a antropologia em homenagem ao Profesor Egon Schaden,
ed. Thekla Hartmann and VeraPenteado Coelho, pp. 105-27. Sao Paulo: Colecao Museo
Paulista.
Harvey, Penelope. 1987. Language and the Power of History: The Discourse of Bilinguals in
Ocongate (southern Peru). Ph.D. thesis, London School of Economics.
. 1991. Drunken speech and the construction of meaning: Bilingual competence in the
southern Peruvian Andes. Language in Society 20: 1-36.
1992. Jugando por la identidad y la tradition. Las corridas de toros en el sur andino.
In Tradition y modernidad en los Andes, ed. Henrique Urbano and Mirko Lauer, pp.
221-42. Cusco: Centro Bartolome de las Casas.
. 1994. Gender, community and confrontation: Power relations in drunkenness in
Ocongate (southern Peru). In Gender, Drink and Drugs, ed. Maryon McDonald, ed. pp.
209-34. Oxford: Berg.
Hastrup, Kirsten, ed. 1992. Other Histories. London: Routledge.
Havelock, Eric. 1986. The Muse Learns to Write: Reflections on Orality and Literacy from
Antiquity to the Present. New Haven, Conn.: Yale University Press.
Hillman, Grady. 1994. Aldeas sumergidas: Lot's wife and water rights. In Andean Oral
Traditions: Discourse and Literature/Tradiciones orales andinas: Discurso y literatura,
ed. Margot Beyersdorff and Sabine Dedenbach-Salazar Saenz, pp. 83-116 (Bonner
Amerikanistische Studicn, 24). Bonn: Holos.
Howard-Malverde, Rosaleen. 1988. Talking about the past: Tense and testimonials in Quechua
narrative discourse. Amerindia 13: 125—55.
. 1989. Storytelling strategies in Quechua narrative performance. Journal of Latin
. 1990. The Speaking of History: 'Willapaakushayki' or Quechua Ways of Telling the Past
(Institute of Latin American Studies Research Papers, 21). London: University of London.
BIBLIOGRAPHY 239
. 1994. "La gente mas bien hace guerra con los cuentos": Estrategias narrativas en una
comunidad quechua del Peru central. In Andean Oral Traditions: Discourse and Litera-
ture/Tradiciones orales andinas: Discurso y literatura, ed. Margot Beyersdorff and
Sabine Dedenbach-Salazar Saenz, (Bonner Amerikanistische Studien, 24), pp. 117-35.
Bonn: Holos.
Huanca, Tomas. 1989. El yatiri en la comunidad aymara. La Paz: Centre Andino de Desarrollo
Agropecuario.
Husson, Jean-Philippe. 1984. L'art poetique quechua dans la chronique de Felipe Waman Puma
de Ayala. Amerindia 9: 79-110.
. 1985. La poesie quechua dans la chronique de Felipe Waman Puma de Ayala. De 1'art
lyrique de cour aux chants et danses populaires. Paris: L'Harmattan.
Ingold, Tim, ed. 1994. Debate on the Motion "The Past is a Foreign Country". Manchester:
Group for Debates in Anthropological Theory.
Isbell, Billie Jean. 1978. To Defend Ourselves: Ecology and Ritual in an Andean Village. Austin:
University of Texas Press.
. 1982. Nature meets culture in the dialectical world of the tropics. Annals of the New
York Academy of Sciences 1982: 353-63.
, and Fredy Roncalla. 1977. The ontogenesis of metaphor: Riddle games among Quechua
speakers seen as cognitive discovery procedures. Journal of Latin American Lore 3(1):
19-49.
Itier, Cesar. 1991. Lengua general y comunicacion escrita: Cinco cartas en quechua de Cotahuasi
(1616). Revista Andina 17: 65-107.
. 1993. Estudio linguistico. In Relation de antiguedades deste reyno del Piru, ed. Pierre
Duviols and Cesar Itier, Edition facsimilar y transcription paleografica del codice de
Madrid, pp. 127-78. (Travaux de 1'IFEA/74; Archivos de Historia Andina/17). Lima
and Cusco: Institut Francais d'Etudes Andines/Centro Bartolome de las Casas.
Ixtlilxochitl, Fernando de Alva. 1975 [1608]. Compendio historico del reino de Texcoco. In
Obras historicas. Vol. I, ed. Edmundo O'Gorman, pp. 415-521. Mexico City: UNAM.
Jackson, Jean. 1991. Being and becoming an indian in the Vaupes. In Nation States and Indians
in Latin America, ed. Greg Urban and Joel Sherzer, pp. 131-55. Austin: University of
Texas Press.
Jones, Linda K. 1992, In pursuit of discourse particles. In Language in Context: Essays for
Robert E. Longacre, ed. Shin Ja J. Hwang and William R. Merrifield, pp. 127-36.
(Publications in Linguistics, no. 107). Arlington: Summer Institute of Linguistics and
The University of Texas.
Keesing, Roger. 1987. Anthropology as interpretive quest. Current Anthropology 28(2): 161-69.
Kemper Columbus, Claudette. 1990. Immortal eggs: A Peruvian geocracy; Pariaqaqa of
Huarochiri. Journal of Latin American Lore 16(2): 175-98.
. 1994. Curious confessions: Cristobal Choquecaxa, informant and actor of Huarochirui.
In Andean Oral Traditions: Discourse and Literature/Tradiciones orales andinas: Dis-
curso y literatura, ed. Margot Beyersdorff and Sabine Dedenbach-Salazar Saenz, pp.
51-82. (Bonner Amerikanistische Studien, 24). Bonn: Holos.
Kendon, Adam. 1992. The negotiation of context in face-to-face interaction. In Rethinking
Context: Language as an Interactive Phenomenon, ed. Alessandro Duranti and Charles
Goodwin, pp. 323-34. Cambridge: Cambridge University Press.
Kristeva, Julia. 1980. Desire in Language: A Semiotic Approach to Literature and Art, ed. Leon
Samuel Roudiez, tr. T. Gora, A. Jardine, and L. S. Roudiez. New York: Columbia
University Press.
Lame, Manuel Quintin. 1971 [1939]. En defensa de mi raza, ed. and with an introduction by
Gonzalo Castillo-Cardenas. Bogota: Comite de Defensa del Indio. (English translation
in Castillo-Cardenas 1987).
240 BIBLIOGRAPHY
Lanser, Susan Sniader. 1981. The Narrative Act: Point of View in Prose and Fiction. Princeton,
N.J.: Princeton University Press.
Lapidus de Sager, Nejama. 1968. El significado de algunos ideogramas andinos. Contribution
al estudio de la comunicacion prehispanica. In Idiomas, cosmovisiones y cultura
(Publication of Institute de Antropologia). Rosario, Argentina: Universidad Nacional
del Litoral.
Laqueur, Thomas. 1994. Memory and naming in the Great War. In Commemorations: The
Politics of National Identity, ed. John Gillis, pp. 150-67. Princeton, N.J.: Princeton
University Press.
Leach, Edmund R. 1976. Culture and Communication. The Logic by which Symbols Are
Connected. Cambridge: Cambridge University Press.
Levinsohn, Stephen H. 1976. Progression and digression in Inca (Quechuan) discourse. Forum
Linguisticum 1(2): 122-47.
Lcvi-Strauss, Claude. 1966. The Savage Mind. Chicago: University of Chicago Press.
Lienhard, Martin. 1990. La voz y su huella. Escritura y conflicto etnico-social en America Latina,
1492-1988. Havana: Casa de las Americas; Hanover, N.H. (USA): Ediciones del Norte 1991,
revised edition; Lima: Horizonte 1991, 3rd revised and expanded edition.
. 1992. La voz y su huella. Lima: Editorial Horizonte.
Lofgren, Orvar. 1993. Materializing the nation in Sweden and America. Ethnos 3-4: 161-96.
Lowcnthal, David. 1985. The Past is a Foreign Country. Cambridge: Cambridge University
Press.
MacCormack, Sabine. 1991. Religion in the Andes: Vision and Imagination in Early Colonial
Peru. Princeton, N.J.: Princeton University Press.
Mamani Mamani, Manuel. 1990. Myth and music in the livestock marking ritual of the Chilean
Andes. Latinamericanist 25(2): 1-7.
Mannheim, Bruce. 1986a. Poetic form in Guaman Poma's Wariqsa Arawi. Amerindia 11:
41-57.
. 1986b. The language of reciprocity in southern Peruvian Quechua. Anthropological
Linguistics 28: 267-73.
1991. The Language of the Inka since the European Invasion. Austin: Texas University
Press.
Manrique, Nelson. 1981. Las guerrillas indigenas en la guerra con Chile [prologue by Pablo
Macera]. Lima: Ital Peru/CIC.
March, Kathryn S. 1983. Weaving, writing, and gender. In Man (N.S.) 18: 729-44.
Martinez, Gabriel. 1976. El sistema de los Uywiris en Isluga. Anales de la Universidad del Norte
(Homenaje al Dr Gustavo Le Paige S.J.) 10: 255-327.
Mateos, Francisco, ed. 1949. Cartas de indios cristianos del Paraguay. Missionalia Hispanica
6(18): 547-72.
Melo, Jorge Orlando. 1989. Etnia, region y nacion: El fluctuante discurso de la identidad (notas
para un debate). In Identidad: Memorias del simposio "Identidad etnica, identidad
regional, identidad nacional", ed. Myriam Jimeno, Gloria Isabel Ocampo, and
Miguelangel Roldan, pp. 27-48. Bogota: ICFES.
Meneses, Teodoro L. 1983. Teatro Quechua colonial: Antologia. Lima: Ediciones Edubanco.
Messick, Brinkley. 1987. Subordinate discourse: Women, weaving, and gender relations in
North Africa. American Ethnologist 14(2): 210-25.
Mignolo, Walter D. 1994. Afterword: Writing and recording knowledge in colonial and
postcolonial situations. In Writing without Words. Alternative literacies in Mesoamerica
and the Andes, ed. Elizabeth Hill Boone and Walter D. Mignolo, pp. 292-313. Durham,
N.C.: Duke University Press.
Millones, Luis. 1985. La familia endemoniada de Chupamarca. Temas sobre psiquiatria y
psicologia 1: 51-8.
BIBLIOGRAPHY 241
. 1989. En busca de Santa Rosa: Reflexiones en torno a una biografia duradera. Bulletin
of the National Museum of Ethnography of Osaka 14(4): 891-905.
. 1991. Los anos oscuros de Santa Rosa y los suenos de Santa Rosa de Lima. In El umbral
de los dioses, ed. Moises Lemlij and Luis Millones, pp. 121-81. Lima: Sociedad Peruana
de Psicoanalisis y Seminario Interdisciplinario de Estudios Andinos.
. 1992. Actores de altura. Ensayos sobre el teatro popular andino. Lima: Editorial
Horizonte.
Monloya, Rodrigo. 1993. Quechua theater: History, violence and hope. Indiana Journal of
Hispanic Literatures 1(2): 53-77.
Moore, Henrietta. 1989. Paul Ricoeur: Action, meaning and text. In Reading Material Culture:
Structuralism, Hermeneutics and Post-Structuralism, ed. Christopher Tilley, pp. 85-120.
Oxford: Blackwell.
Moreno Proano, A. 1979. Cronologia de la vida de Fray Jodoco Ricke, fundador del convento
de San Francisco de Quito (1498-1578). Boletin de la Academia Nacional de Historia
131-32: 173-90.
Morote Best, Efrain. 1957. El oso raptor. Archivos Venezolanos de Folklore 5(5): 157-78.
Munn, Nancy. 1990. Constructing regional worlds in experience: Kula exchange, witchcraft;
and Gawan local events. Man 25: 1-17.
Murra, John V. 1972. El "control vertical" de un maximo de pisos ecologicos en la economia
de las sociedades andinas. In Visita de la provincia de Leon de Huanuco en 1552. Inigo
Ortiz de Zuniga, visitador, ed. John V. Murra, pp. 429-76. Huanuco: Universidad
Hermilio Valdizan.
. 1975a [1973]. Las etno-categorias de un khipu estatal. In Formaciones economicas y
politicas del mundo andino, pp. 243-54. Lima: Institute de Estudios Peruanos.
. 1975b [1958]. La funcion del tejido en varios contextos sociales y politicos. In
Formaciones economicas y politicas del mundo andino, pp. 145-70. Lima: Instituto de
Estudios Peruanos.
. 1982. The mit'a obligations of ethnic groups to the Inka state. In The Inca and Aztec
States 1400-1800: Anthropology and History, ed. George Collier and Renato Rosaldo,
pp. 237-62. London. Academic Press.
. 1989 [1962]. Cloth and its function in the Inka state. In Cloth and Human Experience,
ed. Annette B. Weiner and Jane Schneider, pp. 275-302. Washington D. C.: Smithsonian
Institution Press.
, Nathan Wachtel, and Jacques Revel, eds. 1986. Anthropological History of Andean
Polities. Cambridge: Cambridge University Press.
Nash, June. 1979. We Eat the Mines and the Mines Eat Us. New York: Columbia University
Press.
Niessen, Sandra. 1985. Motifs of life in Toba Batak texts and textiles. (Verhandelingen, van het
koninklijk instituut voortaal-, land-en volkenkunde, 110). Dordrecht: Holland.
Nora, Pierre. 1984. Entre memoire et histoire: La problematique des lieux. In Les lieux de
memoire, ed. Pierre Nora, vol. 1, pp. xv-xlii. Paris: Gallimard.
. 1989. Between memory and history: Les lieux de memoire. Representations 26: 7-25.
Nunez del Prado Bejar, Daisy. 1972. La reciprocidad como ethos de la cultura quechua.
Allpanchis4: 135-65.
Nunez del Prado Bejar, Juan. 1970. El mundo sobrenatural de los quechuas del sur del Peru a
traves de la comunidad de Qotabamba. Allpanchis 2: 57-119.
Nunez del Prado, Oscar. 1973. Kuyo Chico: Applied Anthropology in an Indian Community,
tr. Lucy Whyte Russo and Richard Russo. Chicago: University of Chicago Press.
Ong, Walter. 1982. Orality and Literacy: The Technologizing of the Word. New York: Methuen.
. 1986. Text as interpretation, Mark and after. In Oral Tradition in Literature, ed. John
Miles Foley, pp. 147-69. Columbia: University of Missouri Press.
242 BIBLIOGRAPHY
Ortiz Rescaniere, Alejandro. 1973. De Adaneva a Inkarri: Una vision indigena del Peru. Lima:
Retablo de Papel.
Ossio, Juan, ed. 1973. Ideologia mesianica del mundo andino. Lima: Ignacio Prado Pastor.
Ossio, Juan M. 1985. Los retratos de los Incas en la cronica de Fray Martin de Morua. Lima:
COFIDE.
Oviedo y Herrera, Luis Antonio de. 1867. Santa Rosa de Lima, poema heroico. Lima:
Administration del Peru Catolico.
Oxford English Dictionary. 1971. The Compact Edition of the Oxford English Dictionary. New
York: Oxford University Press.
. 1989. 2nd edition. Oxford: Clarendon.
Pachacuti Yamqui Salcamaygua, Joan de Santacruz. 1968 [1613]. Relacion de antiguedades
deste rcyno del Peru. In Cronicas peruanas de interes indigena, ed, Francisco Esteve
Barba (BAE no. 209), pp. 279-319. Madrid: Atlas.
1993 [1613]. Relacion de antiguedades deste reyno del Peru. Estudio etnohistorico y
linguistico de Pierre Duviols y Cesar Itier. Cusco:CentroBartolome de las Casas/ln-
stituto Frances de Estudios Andinos.
Palma, Ricardo. 1983. Tradiciones peruanas. Barcelona: E. Cappelletti.
Parssinen, Martti. 1992. Tawantinsuyu: The Inca State and its Political Organization. Helsinki:
SHS.
Paso y Troncoso, Francisco de, ed. 1939-42. Epistolario de Nueva Espana. Antigua Libreria
Robredo de Jose Porrua c hijos, 16 vols. Mexico: Biblioteca Historica Mexicana de
Obras Ineditas.
Perrin, Michel. 1986. 'Savage' points of view on writing. In Myth and the Imaginary in the New
World, ed. Edmundo Magana and Peter Mason, pp. 211-31. Amsterdam: CEDLA.
Perroud, Pedro C., and Juan M. Chouvenc. 1969? Diccionario Castellano Kechwa, Kechwa
Castellano: Dialecto de Ayacucho. Santa Clara, Peru: Seminario.. .Padres Redentoristas.
Platt, Tristan. 1978. Symmetries en miroir: Le concept de yanantin chez les Macha de Bolivie.
In Annales E.S.C. 33(5-6): 1081-1107.
1982. Estado boliviano y ayllu andino: Tierra y tributo en el Norte de Potosi. Lima:
Institute de Estudios Peruanos.
. 1983. Conciencia andina y conciencia proletaria: qhuya runa y ayllu en el Norte de
Potosi. Lima: HISLA 2: 47-73.
. 1986. Mirrors and maize: The concept of yanantin among the Macha of Bolivia. In
Anthropological History of Andean Polities, ed. John V. Murra, Nathan Wachtel, and
Jacques Revel, pp. 229-59. Cambridge: Cambridge University Press.
. 1987. The Andean Soldiers of Christ: Confraternity organization, the Mass of the Sun
and regenerative warfare (XVIII-XX centuries). Journal de la Societe des Americanistes
83: 139-92
. 1988. Cultos milagrosos y chamanismo en el cristianismo surandino. Iglesia, religion
y sociedad en la historia latinoamericana. Actas del VIII Congreso de la Asociacion
de Historiadores Latinoamericanistas en Europa, pp. 119-37. Szeged: University of
Szeged.
. 1992. Writing, shamanism and identity, or voices from Abya-Yala. In History Work-
shop Journal 34: 132-47.
, Therese Bouysse-Cassagne, Olivia Harris, and Thierry Saignes, eds. In press.
Qaraqara/Charka: Fuentes para el estudio de una confederation Aymara (siglos XV-
XVII). La Paz: CID/IFEA/HISBOL.
Poma de Ayala, Felipe Guaman. 1980 [c. 1615]. El primer nueva coronica y buen gobierno. 3
vols., ed. John V. Murra and Rolena Adorno. Mexico City: Siglo XXI.
Poole, Deborah. 1982. Temporal classification and transformation in Andean devotional dance.
Manchester: XLIV International Congress of Americanists. Unpublished ms.
BIBLIOGRAPHY 243
. 1984. The choreography of history in Andean dance. Denver, Colo.: 83rd Annual
Meeting of the American Anthropological Association, Unpublished ms.
Poole, Deborah A. 1990. Accommodation and resistance in Andean ritual dance. Drama Review
34(2): 98-126.
Proust, Marcel. 1987. Remembrance of Things Past. Harmondsworth: Penguin.
Purin, Sergio, ed. 1990. Inca—Peru 3000 ans d'histoire. Brussels: Musees Royaux d'Art et
d'Histoire.
Ramirez, Juan Andres. 1969. Lanovenaal Senor de Qoyllur Rit'i. Allpanchis 1: 61-88.
Randall, Robert. 1982. Qoyllur Rit'i, an Inca fiesta of the pleiades. Bulletin del' Institut Francais
d'Etudes Andines 11(1-2): 37-81.
. 1990. The mythstory of Kuri Qoyllur: Sex, seqes, and sacrifice in Inka agricultural
festivals. Journal of Latin American Lore 16(1): 3-46.
Rappaport, Joanne. 1980-81. El mesianismo y las transformaciones de simbolos mesianicos en
Tierradentro. Revista Colombiana de Antropologa 23: 365-413.
. 1985. History, myth and the dynamics of territorial maintenance in Tierradentro,
Colombia. American Ethnologist 12(1): 27-45.
. 1988. La organizacion socio-territorial de los Pastos: Una hipotesis de trabajo. Revista
de Antropologia 4(2): 71-103.
. 1990. The Politics of Memory: Native Historical Interpretation in the Colombian
Andes. Cambridge: Cambridge University Press.
. 1994. Cumbe Reborn: An Andean Ethnography of History. Chicago: University of
Chicago Press.
Redford, Bruce. 1986. The Converse of the Pen: Acts of Intimacy in the Eighteenth Century
Familiar Letter. Chicago: Chicago University Press.
Reina, Leticia. 1980. Las rebeliones campesinas en Mexico, 1819-1906. Mexico: Siglo
XXI.
Ricoeur, Paul. 1971. The model of the text: Meaningful action considered as text. Social
Research 11: 529-62.
. 1981. Hermeneutics and the Human Sciences, ed. John B. Thompson. Cambridge:
. 1991. From Text to Action. Essays in Hermeneutics, II, tr. Kathleen Blarney and John
B. Thompson. London: Athlone.
Rivera Cusicanqui, Silvia, and Equipo THOA. 1992. Ayllus y proyectos de desarrollo en el
Norte de Potosi. La Paz: Ediciones Aruwiyiri.
Rodriguez, Rosario. 1994. Desde las practicas textuales aymaras. Talk given at the presentation
of the book, Jichha na parlt'a: Ahora les voy a narrar. E. Espejo Ayka, ed. Denise Y.
Arnold and Juan de Dios Yapita. La Paz: UNICEF/Casa de las Americas.
Rosaldo, Renato. 1989. Imperialist nostalgia. Representations 26: 107-22.
Rostworowski de Diez Canseco, Maria. 1986. Estructuras andinas del poder, 2nd ed. Lima:
Institute de Estudios Peruanos.
. 1988. Conflicts over coca fields in XVIth century Peru. (Memoirs of the Museum of
Anthropology, 21). Ann Arbor: University of Michigan.
Sallnow, Michael. 1974. La peregrinacion andina. Allpanchis 7: 101-42.
. 1987. Pilgrims of the Andes: Regional Cults in Cusco. Washington D.C.: Smithsonian
Institution Press.
Salomon, Frank. 1981. Killing the yumbo: A ritual drama of northern Quito. In Cultural
Transformations and Ethnicity in Modern Ecuador, ed. Norman Whitteti Jr., pp. 162-
208. Urbana: University of Illinois Press.
. 1991. Introductory essay: The Huarochiri manuscript. In The Huarochiri Manuscript:
A Testament of Ancient and Colonial Andean Religion, tr. and ed. Frank Salomon and
George L. Urioste, pp. 1-38. Austin: University of Texas Press.
244 BIBLIOGRAPHY
, and George Urioste, tr. and ed. 1991. The Huarochiri Manuscript: A Testament of
Ancient and Colonial Andean Religion. Austin: University of Texas Press.
Sanchez, Luis Alberto. 1950-51. La literatura peruana. Buenos Aires: Editorial Guarania.
Santacruz Pachacuti Salcamaygua, Juan de. 1879 [1613]. Relation de antiguedades de este
reyno del Peru. In Tres relaciones de antiguedades peruanos, ed. Marcos Jimenez de la
Espada, pp. 229-328. Madrid: Atlas.
Schiffrin, Deborah. 1994. Approaches to Discourse. Oxford: Blackwell.
Segal, Daniel A. 1988. Nationalism, comparatively speaking. Journal of Historical Sociology
1:301-21.
Segal, Daniel A., and Richard Handler. 1992. How European is Nationalism? Social Analysis
32: 1-15.
Sharon, Douglas. 1980. El Chaman de los Cuatro Vientos. Mexico City: Siglo Veintiuno
Editores.
Sherbondy, Jeannette E. 1988. Mallki: Ancestros y cultivo de arboles en los Andes. In Sociedad
andina pasado y presente. Contribuciones en homenaje a la memoria de Cesar Fonseca
Martel, ed. Ramiro Matos Mendieta, pp. 101-35. Lima: Fomciencias.
Shopes, Linda. 1986. Oral history and community involvement: The Baltimore Neighborhood
Heritage Project. In Presenting the Past: Essays on History and the Public, ed. Susan
Porter Benson, Stephen Brier, and Roy Rosenzweig, pp. 249-63. Philadelphia: Temple
University Press.
Silverman -Proust, Gail P. 1988. Significado simbolico de las franjas multicolores tejidas en los
wayakos de los Q'ero. Boletin de Lima 57(10): 37-44.
Simpson, D. P. 1973. Cassell's Latin-English Dictionary. 5th ed. London: Cassell.
Skar, Harald 0.1982. The Warm Valley People: Duality and Land Reform Among the Quechua
Indians of Highland Peru. Oslo: Oslo University Press.
Skar, Sarah Lund. 1994. Lives Together Worlds Apart. Quechua Colonization in Jungle and
City. Oslo: Scandinavian University Press.
Sola, Donald. 1969. Cuzco Quechua Language Materials. Ithaca, N. Y.: Cornell University Latin
American Studies Program.
Sontag, Susan. 1977. On Photography. New York: Delta.
Spivak, Gayatri Chakravorty. 1988. In Other Worlds: Essays in Cultural Politics. London:
Routledge.
1993. Can the subaltern speak? In Colonial Discourse and Post-Colonial Theory:
A Reader, ed. Patrick Williams and Laura Chrisman, pp. 66-111. New York:
Harvester.
Stobart, Henry. 1987. Primeros datos sobre la rnusica del norte de Potosi, Serie Anales de !a
Reunion de Etnologia no. 1, tomo HI: 81-96. La Paz: MUSEF.
Strathern, Marilyn. 1987. Comment on Keesing, Roger 1987. Current Anthropology 28(2):
173-74.
1992. After Nature: English Kinship in the Late Twentieth Century. Cambridge:
Street, Brian V. 1993. Introduction: The new literacy studies. In Cross-Cultural Approaches to
Literacy, ed. Brian V. Street, pp. 1-21. Cambridge: Cambridge University Press.
Sullivan, Lawrence. 1988. Icanchu's Drum: An Orientation to Meaning in South American
Indian Religions. New York: Macmillan.
Sweeney, Amin. 1987 A Full Hearing: Orality and Literacy in the Malay World. Berkeley:
University of California Press.
Szeminski, Jan. 1987. Un kuraka, un dios y una historia. (Serie: Antropologfa Social e Historia
No. 2). San Salvador de Jujuy: Talleres Graficos Gutenburg.
Taussig, Michael. 1990. Shamanism, Colonialism and the Wildman: A Study in Terror and
Healing. Chicago: University of Chicago Press.
BIBLIOGRAPHY 245
1992. The Nervous System. London: Routledge.

Taylor, Gerald. 1980. Supay. Amerindia 5: 47-63.
. 1987a. Ritos y tradiciones de Huarochiri. Manuscrito quechua de comienzos del siglo
XVII. Version paleografica, interpretation fonologica y traduccion al castellano. Lima:
Institute de Estudios Peruanos/Institut Francais d'Etudes Andines.
. 1987b. Introduction. In Ritos y Tradiciones de Huarochiri, pp. 15-37. Lima: Institute
de Estudios Peruanos/Institut Francais d'Etudes Andines.
. 1990. Le dialecte quechua de Laraos, Yauyos: Etude morphologique. Bulletin de
1'Institut Fran9ais d'Etudes Andines 19(2): 293-325.
. 1994. Estudios de dialectologia quechua (chachapoyas, ferrenafe, yauyos). Lima:
Universidad National de Education.
Tedlock, Barbara, and Dennis Tedlock. 1985. Text and textile: Language and technology in the
arts of the Quiche Maya. Journal of Anthropological Research 41(2): 121-46.
Terdiman, Richard. 1989. The mnemonics of Musset's confession. Representations 26:
26-48.
Theweleit, Klaus. 1989. Male Fantasies. Minneapolis: University of Minnesota Press.
Thompson, John B. 1990. Ideology and Modern Culture. Cambridge: Polity Press.
Tiffin, Chris and Alan Lawson, eds. 1994. De-scribing Empire: Post-Colonialism and Textual-
ity. London: Routledge.
Todorov, Tzvetan. 1984. The Conquest of America: The Question of the Other. New York:
Harper and Row.
Tomoeda, Hiroyasu. 1988. "La llama es mi chacra": El mundo metaforico del pastor andino. In
Llamichos y paqocheros: Pastores de llamas y alpacas, ed. Jorge Flores Ochoa, pp.
225-36. Cusco:CentroBartolome de las Casas.
Tonkin, Elizabeth, Maryon McDonald and M. Chapman, eds. 1989. History and Ethnicity.
London: Routledge.
Torero, Alfredo. 1964. Los dialectos quechuas. Anales Cientificos de la Universidad Agraria
(Lima) 2: 446-78.
Torrico, Cassandra. 1989. Living weavings: The symbolism of Bolivian herders' sacks. Sucre:
HISBOL/Archivo Nacional de Bolivia. Mimeograph.
Trimborn, Hermann. 1967. Einleitung. In Francisco de Avila, ed. Hermann Trimborn and Antje
Kelm, pp. 1-15. (Quellenwerke zur alten Geschichte Amerikas aufgezeichnet in den
Sprachen der Eingeborenen, 8). Berlin: Gebruder Mann Verlag.
Urbano, Henrique O. 1980. Dios Yaya, Dios Churi, Dios Espiritu. Journal of Latin American
Lore 6: 111-27.
Urioste, George L. 1973. Chay Simire Caymi: The Language of the Manuscript of Huarochiri
(Latin American Studies Series, no. 79). Ithaca, N.Y.: Cornell University.
Urton, Gary. 1981. At the Crossroads of the Earth and the Sky: An Andean Cosmology. Austin:
University of Texas Press.
Valderrama, Ricardo, and Carmen Escalante. 1988. Del Tata Mallku a la Mama Pacha: Riego,
sociedad y ritos en los Andes peruanos. Lima: DESCO, Centro de Estudios y Promotion
del Desarrollo.
Van de Kerke, Simon, and Pieter Muysken. 1990. Quechua mu and the perspective of the
speaker. In Unity in Diversity: Papers Presented to Simon C. Dik: on His 50th Birthday,
ed. Harm Pinkster and Inge Genee, pp. 151-63. Dordrecht: Foris.
Van Gennep, Arnold. ] 960. The Riles of Passage. London: Routledge and Kegan Paul.
Wachtel, Nathan. 1977. The Vision of the Vanquished: The Spanish Conquest of Peru through
Indian Eyes, 1530-1570. Hassocks: Harvester Press.
Wagner, Roy. 1981. The Invention of Culture. Chicago: University of Chicago Press.
Watson-Franke, M-B, 1974. A woman's profession in Guajiro culture: Weaving. An-
thropologica 37: 25-40.
246 BIBLIOGRAPHY
Weber, David J. 1976. Bases para el estudio de la relativizacion con referenda al quechua.
(Documento de Trabajo 10), Yarinacocha, Pucallpa: Institute Linguistico de
Vcrano/Ministerio de Education.
. 1986. Information perspective, profile, and patterns in Quechua. In Evidentiality: The
Linguistic Coding of Epistemology, ed. Wallace Chafe and Joanna Nichols, pp. 137-55.
(Advances in Discourse Processes, vol. 20). Norwood, N.J.: Ablex Publishing.
. 1989. A Grammar of Huallaga (Huanuco) Quechua. (University of California Publi-
cations in Linguistics, 112). Berkeley: University of California Press.
Wightman, Ann. 1990. Indigenous Migration and Social Change: The Forasteros of Cuzco,
1520-1720. Durham, N.C.: Duke University Press.
Williams, Brett. 1988. Upscaling Downtown: Stalled Gentrification in Washington, D.C.Ithaca,
N.Y.: Cornell University Press.
Wolck, Wolfgang. 1979. The structure of discourse as an integral part of Quechua grammar.
Paper presented at the International Congress of Americanists, Vancouver, 1979.
Unpublished ms.
Yupangui (Yupanqui), Titu Cussi (Cusi). 1985 [1570]. Ynstrujion del Ynga don Diego de
Castro Titu Cussi Yupangui, intro. Luis Millones. Lima: El Virrey.
Zorn, Elayne. 1986. Textiles in herders' ritual bundles of Macusani, Peru. In The Junius B. Bird
Second Conference on Andean textiles, ed. Ann P. Rowe, pp. 289-307. Washington D.
C.: Textile Museum.
Zuidema, R. Tom. 1980. El ushnu. Revista de la Universidad Complutense 2(117): 317-62.
. 1986. La civilization inca au Cuzco. Paris: Presses Universitaires de France.
1988. The two pillars of Cuzco: Which two dates of sunset did they define? In New
Directions in American Archacoastronomy: Proceedings of the 46th International
Congress of Americanists (Amsterdam, Netherlands 1988), ed. Anthony F. Aveni, pp.
143-89. (BAR International Series 454). Oxford: BAR.
1989. Burocracia y conocimiento sistematico en la sociedad andina. In Reyes y
guerreros: Ensayos de cultura andina, pp. 488-535. Peru: Fomciencias.
Index
Adelaar, Willem, 5-6,15,160-1,165 n.2,165 Inkas as, 82, 92

n.5 as machu, 53
administrative structure. See reduccion, Anderson, Benedict, 67
resguardo Anderson, Lloyd B., 150
agency, 12, 22-3, 29-32, 35, 39-40, 82 animals, 76, 78-9, 80, 82, 108, 110, 116-17.
of non-discursive memory, 23 See also ritual
Aijmer, Goran, 86 anthropomorphisation of, 109
ajsu. See aqsu ceremonies for, 79, 84 n.17, 85, 89, 91-2
alcohol, 35, 78-9, 89, 96, 200, 202 gendering of, 109
drunk by mountain spirit, 201, 208, 221 ritual dressing of 88-9
Allen, Catherine, 5-6, 12, 15, 107, 195 n.6 ritual imitation of, 90
alpacas, 74, 78, 86-93. See also camelids, ritual marking of, 87, 90, 97 n.l, 117, 127
llamas mythic origins of, 87
as "flowers", 93 anthropology
as humanised, 109 dialogical, 9, 14, 17
alphabet. See also alphabetic script, literacy, post-structuralist, 25
writing symbolic, 9, 12
as metaphor, 197 visual, 97
alphabetic script, 3, 14, 17, 75, 86, 196 apu, 34, 36, 80. See also Ausangate
ancestor(s), 24, 65, 107. See also chullpa, aqsu (ajsu), 202
condemned souls, death, t"aqsu Aru. See Jaqi
as awki, 34 Arguedas, Jose Maria, 43 n.21, 78
condor as, 16-17 Arnold, Denise, 7, 14
as gentiles, 53, 220 Atahuallpa, Inka, 15, 51
247
248 INDEX
August, month of, 51-3, 77-80 Andeanisation of, 69 n.6

Ausangate, Mt, 34, 37, 80 death of, 152
authenticity, 15, 18 n.13 Christianity, 41 n.7, 177, 179. See also Cath-
Avila, Francisco de, 136, 149, 165 n.l olic Church
awayu, 100, 102-5, 110-13, 121-26, 128 Andean forms of, 53, 69 n.6, 200
ayllu, 81, 83 n.l2, 99, 110, 127-29, 189, 190 Christian God, 34. See also altu dyus, dei-
Aymara ties
identity, 96 indigenous appropriation of, 183, 200
language, 94, 96, 206 n.v theatrical representation of, 56, 60
people, 10, 14, 85-6 Chullpa, 202, 225 n.5
ayni, 76, 83 n.7. See also reciprocity as ancestors, 203 n.e, 209 n.cc, 211 n.oo,
212n.pp, 220
Bakhtin, Mikhail, 101 as archaeological remains, 106, 216 n.vv,
Barthes, Roland, 85-6, 96 223
bayeta. See textiles chullpa church, 220
Benveniste, Emile, 17n.2 Chullpa Pass, 212
Bertonio, Ludovico, 126, 225 n.6 chullpa peaks, 215, 223
blood chullpa saints, 203 ri.e, 205, 211, 220, 222
blood-red colour, 109-10, 117, 131 n.3 chullpa wind 216 n.vv, 223
and female creativity, 129 as mummy bundles, 106
menstrual, symbolism of, 117 clothing, handmade. See also awayu, punchu,
in ritual 35, 89, 90-1, 118 textiles, weaving
Bolivian Revolution, 127 almilla, 103
Bombori. See Tata Pumpuri ch'ulu, 103
Bourdieu, Pierre, 17 n.7 as encoding identity, 115
bullfight. See also bulls, condor, death as gendered, 100, 104-5, 118, 122-25,
filmsof,41,43n.21 129-30
interpretation of, 43 coca, 33,35,77-9,89,107,197-98,202,205,
symbolism of, 35-40, 41, 43 n.21 217
as Yawar Fiesta, 43 n.21 pre-Hispanic cultivation of, 53
bulls, 15, 35 Colombia, 55-6
colour, 86, 88-9, 98 n.4, 109-10, 114, 116-
cabildo. See also shamanic session 20, 123. See also blood, dyeing
as administrative council, 42 n. 16, 68 Columbus, Christopher, 65
as tribute-paying group, 198, 202, 225 n.9 commemorative ceremonies, 10-11, 13, 15,
camelids, 89, 93. See also alpacas, llamas 21-4, 26-32, 35, 37, 40. See also Day
Campos Chavez, Pio, 47, 50, 54 of the Indian
Carhuamayo, 45, 47, 51-2 as revising history, 59
Carnival, 56, 62,79, 88, 107 communication. See also letter-writing
Castell-Dos-Ruis, Marquess of, 47 Andean media of, 75, 80-1
Catholic Church, 13, 45-6, 49 as accommodation, 14
as domesticator of sacred power, 224 across ethnic boundaries, 14, 177, 183
Catholicism, 51, 53, 96. See also Christianity as asymmetric, 172-74
Ccreceda, Veronica, 100, 113 as non-verbal, 86
Checas, San Damian de, 138 with spirit world, 34, 82. See also moun-
chicha, 88-9, 142, 157 tain gods, interaction with
Chile, 85-6 compadrazgo, 33
Chinchaycocha, 51-2 Dia Comadre, 88
ch'isiraya, 198. See also kancha, pampa Dia Compadre, 88
Choquecaxa, Cristobal, 136, 147 condemned souls, 76-7, 83 n.10.
Christ, 60, 73-4, 77-8 condor, 15-16. See also bullfight
INDEX 249
and bull, 33-6, 43 n.19, 43 n.20, 43 n.22 altudyus, 199

as lawyer figure, 221-4 awaq mamacha, 74, 77
as mountain spirit that speaks through sha- moon 202
man, 200 taytacha, 74, 77
as ritual brother-in-law, 16 uywiri, 87-8, 92, 94-5
in shamanic session, 200-01 deixis, 5
Connerton, Paul, 59 despacho (burnt offering), 80
context, 3,6, 149-151, 161, 166n.9, 172-3, devils. See supay
186-9,191, 193,195-7, 204 n.n, 224 Dia del Indio. See Day of the Indian
creation of, 6-7, 17 Diaz, Benigno, 41
Copacabana. See pilgrimage discourse, 3-5, 25. See also discourse mark-
Corpus Christi, 73 ers, speech acts, weaving
corral. See kancha approaches to, 4, 17 n.l
cosmological beliefs. See cosmology, An- as hybrid, 15
dean indigenous, addressed to outsiders, 172
cosmology, Andean, 12, 15-17, 24, 34, 84 Mesoamerican, 15
n.26. See also hanaq pacha, kay pacha, presence of the speaker in, 136-38, 143,
pachakuti, ukhu pacha 146-47, 151, 155-57, 164
criollos, 172-73, 177-83 as socially grounded, 4
cultural expression
structure, 5, 136
as empowerment, 11
as text, 5
as enactment, 11
uses of the term, 4, 17 n.2
as evocation, 9
discourse markers, 137,150,205 a.p. See also
as hybridisation, 14
evidential suffixes
as performance, 10. See also representa-
Dominican order, 46-7
tion
cultural identity, 10. See also hybridity, na- drama. See theatre.
tional identity Dransart, Penny, 10, 12, 108
expressed in song, 52-3 Duranti, Alessandro, 5, 6
invention of, 67 dye, dyeing, 89, 109, 120, 122
culture-as-text. See text
Cumbal, 55-61,64-5, 68 ecological zones, 195 n.4
Cusco, 53, 79, 92, 180 and social networks, 190, 192
Ecuador, 56, 59
dance, 27, 35, 51, 58-9, 76-7, 80, 83 n.12, 83empowerment. See also cultural expression
n.13, 101, 158 indigenous, 14
darkness, 11, 106, 152, 200-2, 211 n.oo, 225enactment. See cultural expression
n.12 ethnic identity, 115. See also cultural iden-
Day of the Indian, 55, 57, 60 tity
death ethnic typologies, 45, 58-9. See also criollos,
of Atahuallpa, 51 indigenous nobility, mestizos, runa
in bullfight, 35-7 ethnicity, 15, 45
and fertility, 107 as re-invented, 15
in hands of Tata Pumpuri, 223 in theatrical performance, 55
ideas about, 51, 76 Europe, poetic representation of, 49
of Saint Rose, 46-7, 50 Evans-Pritchard, E. E., 197
of the sun, 152 evidential suffixes, 6,13,24,150-51,156-64,
De Certeau, Michel, 64 208 n . b b . See also evideritiality,
Dedenbach, Sabine, 5-6, 13, 15 Quechua grammar
deities, 75. See also Christ, jurq'u, evidentiality, 17n.lO, 150-51, 160, \62.See
Pachamama, saints, sun also evidential suffixes
250 INDEX
festivals, 47, 50-1, 53, 56, 79, 198. See also hanaq pacha (janaj pacha) (Upper World), 15,
Carnival, Corpus Christi, Day of the 199
Indian, New Year's Day, Qoyllur Rit'i, Hanks, William, 5-6, 8
Tata Pumpuri Harris, Olivia, 17
Day of the Dead, 107 Harvey, Penelope, 10-11, 13, 15
Inti Raymi, 80 hermeneutics, 8
Saint John's Day, 80 history. See also commemorative ceremonies,
Qapaq Raymi, 90 memoristas, memory, oral history
fiestas. See festivals Andean periodisation of, 24, 41 n.7
Figueroa, Luis, 43 n.21 as authoritative narrative, 24, 30
flamingos, 89 as contested, 31-3, 37, 40
fleece, 89, 108-9, 127 correcting of, 60
Flores de Oliva, Isabel. See Saint Rose dominant versions of, 30-2, 40
Flores Ochoa, Jorge, 78 indigenous versions of, 13, 39, 59-61, 66-7
flowering. See also memory and lived experience, 22
and memory, 94 and memory, 22-3
metaphor of, 93-4, 107-8, 127 as non-discursive narrative, 31-2
Foucault, Michel, 4-5 ritual enactment of, 15, 22, 26-32
frame theatrical representation of, 51, 55-8
of interpretation, 197,204 n.n, 206 n.s, 206 and visual images, 55
n.u, 219, 225n.l2 Howard-Malverde, Rosaleen, 4, 8, 83 n.2, 83
participant indicators of, 204 n.n, 206 n.s, n.3, 151, 155
206 n.u, 219-20 huaca (wak'a), 80, 84 n.21, 166 n.10, 166
n. 11, 225n.5
Fujimori, Alberto, 31
emits deadly wind, 216 n.yy
Huancavelica, 45
Garcia, Alan, 31
Huarochiri manuscript, 3, 7, 13, 15, 136-38,
Garcilaso de la Vega, Inca, 80
349, 165 n.l, 165 n.2, 211 n.oo
Geertz, Clifford, 7-8
hybridisation. See hybridity
gender, 102. See also clothing, textile designs,
hybridity, 14-15, 17, 175-76
weaving, writing
chacha kasta (male seedlines), 124
iconicity, 81-2
in division of labour, 99-102, 105, 108
ilia, 78-9
in men's weavings, 100, 104-5 illiteracy, 32, 86, 195 n.l
and relations of production, 128-30 as colonial construct, 196
and textile language, 104 Independence Day
wila kasta (female bloodlines), 124 in Apurimac, 43 n. 18
in women's weavings, 100, 104-5 in Peru, 10-11, 21-3, 26-32, 35, 38-9
Getzels, Peter, 41 indigenous nobility, 174-75
godparenthood. See compadrazgo Inka(s), 24, 53, 82, 225 n.5. See also
Goodwin, Charles, 5-6 Atahuallpa, King of Spain, sun
Goody, Jack, 185 initiation rites, 90
Gose, Peter, 43n.l8 labour tribute, 103
Griffiths, Gareth, 18 n.13 origin myth of, 94
Gruzinski, Serge, 58 panaka (lineages), 92, 94, 98 n.5
Guaman Poma de Ayala, Felipe, 14, 94-95, and red llama (puka llama), 97 n. 1
97 n. 1,176 religion of 84 n.l4, 84 n.l8
Guarani people, 178-79 Titu Cusi Yupanqui, 176
inkhuna. See textiles
haciendas, 56 Inner World. See juturi, t"aqsu, ukhu pacha
Hall, Stuart, 10 inqa, inqaychu, 78-80, 195. See also ilia
INDEX 251
interface symbolic use of, 64

as interpretive concept, 14—15, 17 n.12 Lanser, Susan Sniader, 150, 155
between orality and literacy, 186-7, 193 Leach, Edmund, 86
interpretation, 8, 11, 26. See also Ricoeur learning. See also knowledge
as appropriation, 8 by being "santuyuq" (endowed by a saint),
of culture as text. See text 76
as hermeneutic process, 8,11 knitting, 102
through reception, 8-9 musical skills, 102
intertextuality, 101, 105, 130 skills for life, 76
between poncho and kipu, 128 to spin, 76, 102
Isbell, BillieJean, 77 to weave, 77,101-2, 107, 115
Isluga, 86-88, 90, 92, 95-7, 97 n.3 to write, 86, 101
letter-writing, 12, 172, 186-94. See also liter-
Jackson, Jean, 18 n.13 acy, writing
jambo. See yumbo. indigenous, 174-80
janaj pacha. See hanaq pacha as manifesto writing, 177
Jaqi(Aru), 151,165n.5 as medium for indigenous resistance, 173,
jurq'u (mountain spirit), 197, 201-220, 222- 175, 182
3. See also condor Levi-Strauss, Claude, 81
speech of, 205 n.q Lienhard, Martin, 5, 15
juturi (hollows leading to Inner World), 87,89 light, 78-80, 84 n.l9, 106, 109, 118, 174,201
lightning, 84 n.14. See also Tata Pumpuri,
kancha (corral), 10 shaman(s)
as ritual space, 85-93, 95, 97 n.l. See also as bestower of miniatures, 76
ch'isiraya, pampa as deity, 199-200, 225 n.6
kawiltu. See shamanic session effect of strike by, 78, 80, 200, 216 n.ww
kay pacha (This World), 15 quartz crystals as residue of, 77
Keesing, Roger, 9 Lima, Second Council of, 45-6, 48-50
King of Spain, 56, 172-74, 176-79 lineages, 129. See also gender, Inka(s), wifia
conflated with Inka, 209 n.cc, 211 n.oo, of animals, 87, 93, 108
222 of humans, 87, 108, 125
kinship, 16, 33, 35. See also compadrazgo, linguistic theory, Western versus Amerindian,
condor, lineage, textiles 12-13
kipu (quipu), 75, 99, 127-29, 171 lista, 118-27
kipukamayuq, 129 literacy,5,10,14-15,99,101,127,185-7.566
knowledge. See also learning also alphabet, orality, letter-writing,
Andean construction of, 13 writing
as Inka wisdom, 82 adult classes in, 29
through seeing, 62 incipient in Quechua, 136
Kristeva, Julia, 96, 101 and power, 188
kukuchi. See condemned souls as social practice, 187, 193
Kusch, Rodolfo, 94 llamas, 86-92, 109. See also kancha
delantero (lead llama), 217
Lame, Manuel Quintin, 64-7, 69 n.7, 69 n.9 as "flowers", 93
land claims, indigenous, 60 luck, 35, 200
landscape, 12, 15, 22, 52 Lund, Sarah, 12, 15
as animate, 34, 85-7
as sacred, 47 Macha, San Pedro de, 17, 126-7, 198, 202
as source of power, 34—7 Macias, Juan, 45-6
land tenure, 56, 62 maize. See ritual
language. See also Quechua language maize beer. See chicha
252 INDEX
Mama Anahuarque, 94-5 grammatical features of, 152-3, 160, 164,

Mama Huaco, 94 166 n.6, 167 n.21
Manco Capac (Manku Qhapaq), 92 historical, 22-3, 55. See also history
Mannheim, Bruce, 76 as multimedia, 56
Martinez, Gabriel, 94 Quechua mythical, 24, 41 n.7, 152
Matapuquio, 186, 189, 191-2, 194 structure of 150-1, 154
Maya versus texture, 61-2
language 181 national identity, 28, 31, 39, 50, 56. See also
Quiche, 101 cultural identity, nationalism
of Yucatan, 180 nationalism, 13, 22, 26-30
media, mixed, 5, 15 indigenous, 55, 177
memoristas (community historians), 57,61-2, New Year's Day, 88
67-8, 68 n.4 Nora, Pierre, 23
memory, 22, 94-6, 100-1, 107-8, 116, 129. Northern Potosi, 16, 83 n.3,126,196,203 n.b,
See also flowering, history, nostalgia 209n.cc,212n.rr,224
and commemorative ceremony, 22, 41 n.4 nostalgia, 38. See also memory
and counter-memory, 40 number symbolism. See twelve
of history enacted, 22-3, 60 Nunez del Prado, Oscar, 76
and representational practice, 26, 31
and subjective experience, 22-3,30-1,37, Ocongate, 10, 13, 17, 26-9, 32-3, 41 n.2, 78
39-40 Ong, Walter, 186
Mesoamerica, 174-5 oral history, 4, 13. See also oral tradition
mestizos, 45, 55-6 oral tradition, 8, 41 n.7, 100. See also oral
history, orality, writing
Mexico, 174, 176
orality, 99, 136, 171. See also interface, liter-
migration, 189
acy, oral tradition, writing
Millones, Luis, 13, 15
and writing, 5, 185-8
mines, 222
Oviedo y Herrera, Luis Antonio, 47-9
miniatures, 11, 74-6, 79-82, 83 n.2, 84 n.17.
See also lightning
pacha (time-space), 80. See also hanaq pacha,
mining, 45
kay pacha, ukhu pacha
Mogrovejo, Toribio de, 45-6, 53 nawpa pacha (former times), 152, 158
Montoya, Rodrigo, 17 n.8, 17 n.9 pachakuti, 41 n.7
moon. See deities Pachamama, 34, 52, 92, 94, 209 n.ff. See also
Moore, Henrietta, 25-6 deities, Pachamamaachi
mountain god(s), 10, 107. See also condor, as Mother Earth, 51
deities, jurq'u as tirakuna, 34, 81
interaction with, 34-6, 201-223 as wirjin tayka, 87-8, 92, 94-6
talks through shaman, 201 Pachamamaachi, 92
Muellames, 56-62, 68 pampa
Munn, Nancy, 22, 39 as open ground, 77, 83 n. 11, 111
music, 35, 55-7, 61, 88, 90, 97 n.3, 105, 102, as ritual space, 80. See also ch'isiraya,
136, 140, 143, 147 n.1, 152, 171 kancha
as unpatterned area in textiles, 106, 109,
Nahuatl language, 174, 175 111-3, 115-9, 122-4
Narino, 55-6, 59, 62, 64, 69 n.6, 69 n.7 panaka. See lnka(s)
narrative, 3, 13, 21, 149, 176, 188 Paqaritampu, 92-4
as ambiguous, 41 n.l, 41 n.7 paqo, 77-8, 80. See also shaman(s)
authority of, 24, 30 Pasto, 57, 69
Aymara mythical, 100, 109 patriotism, indigenous, 182. See also nation-
as confrontational, 39--40 alism
INDEX 253
Pelequen, 46 first person plural marking, 147 n.4, 151,

performance, 10-12, 25, 43 n.18, 51, 57, 86, 153,219
92, 96, 100-2, 154, 188, 196, 201, 224 verbal suffix -mu-, 140-6, 147 n . l l , 147
shamanic, 202-218 n.12, 148n.l4, 148n.l6, 155, 159,
Peru, 13, 15,21-3,27-30 160-1, 204 n.j
government policies in, 31-2 Quechua language, 10-11, 24, 136, 151,183,
history of, 30-1 189. See also Quechua grammar
national anthem of, 28,42 n.14 Aymara influence on, 206 n.v
pilgrimage. See also Qoyllur Rit'i dialect classification of, 147 n.8, 165 n.5,
in the Andes, 73-5, 77-81, 83 n.l 167n.l8
to Copacabana, 83 n.2 as hybridised, 17
Platt, Tristan, 10, 11,16 Potosi variety, 203
Plato, 187-8 semantics of, 12
play. See also pukllay symbolic use of, 35, 64
between condor and bull, 33-4, 36-7, 43 as written discourse, 5
n.19, 43 n.22 Quechua people. See runa
poetry, 47-8, 50-2 quipu. See kipu
point of view of speaker, 13, 143-4, 150, 153, Quives, 46, 53
161, 164, 165 n.3. See also discourse, coca fields of, 53
presence of speaker in curacas of, 53
polysemy, 40. See also polyvocality
polyvocality, 25-6. See also polysemy Rappaport, Joanne, 11, 13, 15
post-structuralisrn. See anthropology rebellions. See also uprisings
Potosi, 45. See also Northern Potosi indian, 178-81
power. See also landscape, state-community reciprocity, 35, 81-2, 83 n.7. See also ayni
relations reducciones, 45, 178
local relations of, 33 religion. See also cosmology, deities, reli-
relations, 5, 9-10, 32-4 gious belief
relations in colonial settings, 7-8, 10, 14,popular, 53
16 religious belief, 46, 197
production, 76 religious language, 12
Protestantism, 96 representation, 9-10
Proust, Marcel, 23 resguardos, 56-7, 59, 68
psychoanalysis, 23 Ricaldi, Herminio, 47
pukllay (play, act out), 11. See also play Ricoeur, Paul, 5-6, 8, 11, 17 n.2, 25-6
as ritual action, 73-7 ritual(s). See also blood, miniatures, ritual
Pumpuri. See, Tata Pumpuri drinking, ritual space, wifia
puna, 33, 52, 78, 202 agricultural, 53
punchu (poncho), 34, 100, 102, 118-29, 131 for animal fertility. See wayfiu
n.2 cleaning of irrigation canals, 156
forced feeding in, 83 n.9, 84 n.25
Qaqachaka, 99, 101, 106, 108-9, 112, 115, interpretation of, 10
118,124-30 maize flour used in, 88
Q'ero, 127 of travel, 186, 192
Qoyllur Rit'i, 11, 73-82, 84 n.25 visual imagery in, 97, 97 n.3
Quechua grammar. See also discourse mark- ritual drinking, 38, 88
ers, evidential suffixes, frame by bullfighters, 34
attributive suffix -yuq, 76, 83 n.8 by condor, 34
citative ni-, 138, 204 n.n ritual space. See ch'isiraya, kancha, pampa
conjectural suffix -cha, 161-64, 166 n.6 Rosaldo, Renato, 38
demonstrative pronouns, 137—40 Rostworowski de Diezcanseco, Maria, 98
254 INDEX
Rousseau, Jean Jacques, 18S speech acts, 218

Rothko, Mark, 97 spinning, 76, 102-3, 108
runa, 139, 142-6, 152, 154-6, 158, 161-2, spirit possession, 200
186, 189, 193, 197 n.2 state-community relations, 21-2, 27-32, 34-
5, 37-8, 40, 66, 224. See also power
saints, 45-6, 53. See also Christ, Chullpa, stone(s), 12, 15, 66, 74,78-80, 82, 83 n.3, 87,
deities 89-91, 109, 152, 155, 225 n.5
Our Lady of the Rosary, 47 Strathern, Marilyn, 9, 40
Saint James (Santiago), 200, 212 ri.rr. See sun, 24, 41 n.7, 53, 65, 66, 110, 174
also Tata Pumpuri death of, 152
Saint John, 79 as deity, 202, 204 n.m, 211 n.oo
Saint Martin, 39 aslnkagod, 106, 117-8, 129
Saint Rose of Lima, 13-5, 45-7, 49, 50-3 as ritual orientation, 73, 79-80, 90
Virgin of Fatima, 74, 79 supay, 166 n.l 1,225 n.5
Sallnow, Miehael J., 73-4 syncchdoche, 81-2, 84 n.22
salta, 106, 109-18, 120, 122-5, 128-9
Santacruz Pachakuti Yamqui, Juan de, 92-3, t"aqsu (location of ancestors), 87
176 Tata Pumpuri (Santiago), 197-200, 201-2,
Santiago. See Tata Pumpuri 205 n.n, 209 n.hh, 210 n.jj, 211 n.nn,
saya, 109, 118-9, 122-4, 131 n.7 212n.pp, 219, 221-4,225 n.l, 225 n.2,
school, 24, 31, 35, 86, 99, 101, 114, 127, 189 225 n.3, 225 n.4, 225 n.5, 226 n.17
schoolchildren, 27, 2,9-30 as Anti-Christ, 225 n.5
semiotics, 7-9, 25 named "Doctor" (miriku), 199. See also
Saussurean, 12, 101 death
shaman(s) (yachaj, yatiri), 83 n.2. See also as lightning, 199-200
paqo Taussig, Michael, 40, 44 n.23
as condor-mountain spirit, 200-1 Tcdlock, Barbara, 101
in indigenous theatre, 58 Tedlock, Dennis, 101
initiation of, 200 teirorism, 13, 31, 41
language of, 10, 12 text, 3-11, 2.5. See also discourse, ritual
chosen by lightning, 78, 200 culture-as-text metaphor, 3-4, 7-10, 23-6,
shamanic session, 200-218,203 n.a, 220-223 40
shamanism in the Andes, 197, 224 as distantiating, 25
sheep, 74, 87-8 methods of analysis of, 8
Shining Path, 22, 39-40 as opposed to non-text, 6
silence, 40 in relation to context, 6, 10-11
ritual use of, 37, 91,202, 221 as social process, 5-6, 8
Silverman-Proust, Gail, 127-8 uses of the term, 4
social organisation. See ayllu Western notions of, 4
socialisation text linguistics, 5
of animals, 90 textile designs, 101,104,107-8,111-23,128.
of humans, 90 See also lista, pampa, salta, saya
Sonqo, 74, 76-81, 107 and gender relations, 117-125, 128-130
sound, 11, 56, 86, 88, 97, 97 n. 1, 97 n.3, 200, and the natural world, 111 -113, 115-117
202, 220-1 textiles. See also clothing, textile designs,
unsemanticised, 207 n.w weaving
Spanish conquest, 41 n.l, 65, 172 bay eta, 103
speech. See also mountain gods(s), interaction ch'uspa (coca bag), 78
with inkhuna (inkuna) (coca cloth), 78, 202
speeches, 28-9, 42 n. 15, 42 n. 16, 43 n. 17, kustala (food sack), 103
44 n.24 as revealing kinship, 124-5
INDEX 255
as text, 3, 7, 100,104, 110, 118 wayno (song), 52

as woven territory, 110 wayfiu (animal fertility ritual), 85-93, 95-7,
textual theory, 86-7, 96 97 n.3, 98 n.4
textualisation, 7-8 weaving, 15, 102
textuality, 3, 5, 86, 197 as age-related activity, 113-7
Andean forms of, 75, 81-2 by analogy with agriculture, 105-6, 116,
texture, 3,11,56,88. See also narrative, verbal 126
imagery, visual imagery discourse of, 105-6, 130
conveyed through language, 64 as gendered activity, 100-5,118,121,125,
as non-discursive expression, 61 129-30
as visual, 62-3, 86 mythic origins of, 106
theatre, 11, 15,51,55-6,61 saints of, 74, 77, 106
as channel for political expression, 62,64,67 techniques of, 102-3, 110
as conveying history, 58, 6:3 as "writing", 100, 106, 114
as conveying moral messages, 58, 68 n.3 Weber, David, 151
Los Cumbes drama group, 57 wind, types of, 216, 216 n.yy, 216 n.zz
This World. See kay pacha wina
Thompson, J. B., 8 etymology of, 92
Titicaca, Lake, 92 as sacrificial hollow in ground, 89, 91
tobacco, 33, 200 wiraquchi
Toledo, Viceroy Francisco de, 198 as reference to non-indigenous outsider,
toqapu (tocapu), 75 204, 206
Torrico, Cassandra, 126-7 wool, 103. See also fleece, dye
Tupac Amaru II, 179-80 as gendered, 108
twelve as sacred number, 203, 207, 217 writing, 15, 86, 187, 196. See also alphabet,
letter-writing, literacy
ukhu pacha (uku pacha) (Inner World), 15,53, as brought by Spanish, 82
87,212 n.pp, 216 n.ww. See also hanaq compared with weaving, 100, 106, 114
pacha, kay pacha contrasted with orality, 4, 154. See also
ukuku, 77, 80, 84n.l3 orality
Upper World. See hanaq pacha as gendered activity, 100, 129-30
uywiri. See deities of history, 100
as instrument of authority, 197
Van Gennep, Arnold, 192 symbolism of, 10
Vargas Llosa, Mario, 31 written language, 135
verbal imagery, 66
Viceroyalty of Peru, 46. See also Toledo yachaj. See shaman(s)
Vilcabamba, 176 yatiri. See sharnan(s)
visual imagery, 11, 14, 61-2, 86. See also Yule, Paul, 43 n.21
colour, ritual, sound, texture yumbo (jambo), 59
as cultural text, 95-7
in theatrical costume, 63 Zorn, Elayne, 113, 115
Zuidema, Tom, 98
wak'a. Seehuaca.

(Rosaleen Howard-Malverde) Creating Context in and

Uploaded by

Copyright:

Available Formats

(Rosaleen Howard-Malverde) Creating Context in and

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Rosaleen Howard-Malverde) Creating Context in and

Uploaded by

Copyright:

Available Formats

Creating Context in

2. Sounds Like Life: Sound-symbolic Grammar, Performance, and Cognition

4. American Indian Languages: The Historical Linguistics of North America

11. Referring to Space: Studies in Austronesian and Papuan Languages

New York Oxford

Copyright © 1997 by Oxford University Press, Inc.

Printed in the United States of America

participation in the roundtable on Sunday morning were invaluable to the debate:

I Textualising Histories and Identities in Cultural Performances

2 Peruvian Independence Day: Ritual, Memory,

3 Saint Rose through the Eyes of Her Devotees:

4 The Art of Ethnic Militancy: Theatre and Indigenous

II Multiple Media: Ritual, Weaving, and the Making of Meanings

5 When Pebbles Move Mountains:

7 Making Men in Her Own Image:

III Inside the Text: Processes of Textualisation in Written Quechua Discourse

8 Spatial Reference and Speaker Orientation in

9 Point of View and Evidentiality in the Huarochiri

IV Orality and Literacies as Reflexes of Colonisation and Resistance

10 Writing from Within: Indigenous Epistolary Practices

11 On the Margin: Letter Exchange among

12 The Sound of Light: Emergent Communication

Willem F. H. Adelaar Rosaleen Howard-Malverde

A tAt its inception, the 1991 conference on "Textuality in Amerindian Cultures"

With regard to the concept of "text", in recognition of the different disciplinary

Discourse and text as social process

We must be ready to receive every moment of discourse in its sudden irruption;

Foucault's observations pertain specifically to written discourses, such as the

and orientational suffixes as analysed by Adelaar, and the patterning of evidential

Text in context, text as context

Text as a model for the interpretation

From culture-as-text to the evocation

From representation to performance

Perhaps instead of thinking of identity as an already accomplished fact, which

There is obviously common ground between this emergent conception of cultural

Enactment and the use of texture in

Rethinking the theory of the linguistic sign

Narrative and nation building

A voice for indigenous peoples

Hybridisation as a strategy of accommodation and

relations of production of agricultural crops and knowledge, along gender lines.11

lawyer-judge whose Quechua discourse is peppered with Spanish-derived legalisms

TEXTUALISING HISTORIES AND

Peruvian Independence Day

Our society has become a narrated society in a threefold

History and memory

Commemorative ritual as text

to meaning and the possibility of a new synthesis of structuralist and performative

The official commemoration

The central commission of festivities for the 166th anniversary of national

In these moments the women of the association of bread-bakers of Ocongate are

challenged by the ways in which subjective reminiscence intervenes in the process of

The clearest or most articulated counter-statement made by the people of Ocongate

In these circumstances, it is not surprising that there is often a sense of failure

ACKNOWLEDGMENTS Earlier drafts of this chapter were presented to the Department of

Somos libres, seamoslo siempre, seamoslo siempre

Largo tiempo el peruano oprimido la ominosa cadena arrastro,

La humillada serviz levanto,

15. The original words of this speech were as follows: