Professional Documents
Culture Documents
Olivia Harris - Alterities Kinship and Gender - A Companion To Latin American - Cap 14
Olivia Harris - Alterities Kinship and Gender - A Companion To Latin American - Cap 14
Olivia Harris - Alterities Kinship and Gender - A Companion To Latin American - Cap 14
to Latin
American
Anthropology
Edited by
Deborah Poole
FLACSO - BibJiciees
ik Blackw ell
P u b lish in g
2 0 0 8 by Blackwell Publishing Ltd
except for editorial material and organization 2 0 0 8 by Deborah Poole
B L A C K W EL L PU BLISH IN G
350 Main Street, Malden, MA 0 2 1 4 8 -5 0 2 0 , USA
9 6 0 0 Garsington Road, Oxford O X4 2 D Q , U K
The right o f Deborah Poole to be identified as the author o f the editorial material in this work has been
asserted in accordance with the U K Copyright, Designs, and Patents Act 1 9 8 8 .
All rights reserved. No part o f 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, except as permitted
by the U K Copyright, Designs, and Patents Act 19 8 8 , without the prior permission o f the publisher.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand
names and product names used in this book are trade names, service marks, trademarks, or registered trade
marks o f their respective owners. The publisher is n ot associated with any product or vendor mentioned in
this book.
This publication is designed to provide accurate and authoritative information in regard to the subject mat
ter covered. It is sold on the understanding that the publisher is not engaged in rendering professional
services. I f professional advice or other expert assistance is required, the services o f a competent professional
should be sought.
1 2008
G N 5 6 4 .L 2 9 C 6 4 2 0 0 8
3 0 6 .0 9 8 dc22
2007032294
A catalogue record for this title is available from the British Library.
T he publishers policy is to use permanent paper from mills that operate a sustainable forestry policy, and
which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices.
Furtherm ore, the publisher ensures that the text paper and cover board used have m et acceptable environ
mental accreditation standards.
Introduction 1
D eborah Poole
Part I: Locations 9
Olivia Harris
Given the immense cultural, ethnic and class diversity o f Latin American societies,
kinship and gender relations in the region are, to say the least, heterogeneous. In this
chapter I explore the productive qualities o f this heterogeneity, bearing in mind that
in an important sense heterogeneity is the ground o f kinship and gender at the most
general level. The point is worth emphasizing for several reasons. First, the resurgent
anthropological interest in kinship has tended to downplay difference and bounded
ness in favour o f more open-ended links. Second, an expanded understanding o f the
range o f means by which people create long-term bonds, and the agency involved in
making and maintaining them, necessarily brings with it an emphasis on relatedness
over unrelatedness, inferiority over exteriority, and potentiality over closure (see
Carsten 2 0 0 0 ). Similarly, approaches to gender have been refashioned in recent dec
ades such that the focus on femininity and masculinity as basic principles o f difference
or opposition has given way to notions o f performance, becom ing, and androgyny
(Strathern 1 9 8 8 ; Butler 1990).
To make a synthesis o f anthropological studies o f Latin American kinship and gen
der is clearly impossible. However, as I read around the subject my interest was
aroused by a number o f issues, some o f which preoccupied me in the past, and some
o f which came out o f the vastness and diversity o f the regional literature. In the end
I have chosen to focus on four themes: the colonial and religious dimensions o f inclu
sion and exclusion across Latin America; the house as a means for understanding
indigenous social forms; the significance o f affinity and alterity in South American
kinship; and the ways that difference is expressed through gender. Each o f these broad
themes points to the idea o f otherness by asking how boundedness is construed and
how alterity constitutes an aspect o f sociality, although they do so, as we shall see, in
very different ways.
T h e history o f Iberian colonization and domination casts a long shadow over the
self-knowledge o f Latin Americans, and equally over the attempts o f anthropologists
to understand the varieties o f social forms. Where these approximate what is familiarly
ALTERITIES: KINSHIP AND GENDER 277
European or Christian, anthropologists have all too often assumed that American
lifevvays were derived from European avatars, manifesting a sameness that was
nonetheless distorted and sometimes disturbing. Conversely, where anthropologists
found more obvious heterogeneity, they often supposed that it was a response to the
experience o f colonial or neocoionial domination. Thus they used to take for granted
that the apparent lack o f social organization in native eastern Amazonia was due to
acculturation and the demoralizing effects o f contact with outsiders, or that the sys
tem o f neighbourhoods (barrios) and patron saint feasts in the Mesoamerican and
Andean highland communities was colonial in origin. It has therefore proved more
complex in Latin America than elsewhere to question the applicability o f models of
relatedness derived from European cultural practice, given the exceptional depth of
colonial history in the region.
How, then, are heterogeneous Latin American persons produced through the
workings o f kinship, social hierarchy, gender and sexual classifications? What are the
processes o f inclusion and exclusion that differentiate people and groups? Race cer
tainly remains a powerful operator in Latin America societies today (see Wade, this
volume). B ut how do racial and ethnic binaries operate in the production o f persons?
Insofar as they identify an excluded other, how does that exclusion work at the level
o f the person? H ow do those who are mixed identify themselves in this way, and for
how many generations? Methodologically similar questions can be asked in contexts
that are closer to the classic anthropological understanding o f kinship and gender. For
example, are women and men thought to be different kinds o f beings? In the making
o f persons is there a bias in favour o f the m others or o f the fathers contribution, o f
matrilateral or patrilateral ties? From a different perspective, where affines are symbol
ized as threatening outsiders, do people produce themselves simultaneously as full
human beings and as dangerous and external? The question o f alterities, o f the con
stitution o f difference in the face o f potential sameness, and vice versa, is central to
this chapter.
Scholarly debate concerning kinship and social organization is usually confined
to particular culture areas - the indigenous worlds o f Amazonia, the Andes, Mesoa-
merica, the circum-Caribbean, the Afro-Americans o f the Caribbean and the Atlantic
seaboard - and even within these some regions are considered to be more exemplary
than others. T hen there is the rest o f Latin America, the demographic majority, its
practices and values supposed to derive mainly from Europe. In this chapter I might
have attempted a synthesis o f the diversities o f Latin America, but as a specialist on the
Andean region I have found it illuminating to follow some o f the debates on different
regions, and to let them speak to each other, taking as a given that there has been far
more historical contact and interchange than we yet have evidence for. A contrast is
apparent between European concepts grounded in legal norms, monotheism and the
search for pure, bounded categories, and forms o f Latin American personhood that
are less absolute, defined more by obligation or work than by birth, or organized
around practices that belong not so much to the domain o f kinship as to that of
religion. Nonetheless, while the contrast between the Euro-Christian and the
Native American has continuing salience, equally important are characteristics that
override it: the enduring competition for status, and the forms o f exclusion and dif
ferentiation in personhood that are and were generated by social class, whether the
elaborate hierarchies o f the great kingdoms and empires, the caste boundaries of
colonial society, or the racialized classes o f contemporary Latin America.
C a t h o l i c I n c l u s io n s a n d E x c l u s io n s
The first case I have chosen to focus on is how American, African and European per
sons became redefined in relation to each other through the prism o f colonial hierar
chy and Catholic Christianity. A fundamental aspect o f the conversion o f the New
World was the attempted globalization o f Spanish and Portuguese kinship and gender
norms. A vast literature was generated in the 16th and 17th centuries on questions o f
heresy and idolatry, but the onslaught on notions o f relatedness and sexuality, the
project to recategorize personhood and inheritance through civil and canon law, was
every bit as profound, and given the greater importance attached to eradicating idol
atry, it is striking how much priestly energy was devoted to these issues. The confes
sionals that were written to guide new American Christians in the finer details o f their
faith reveal as much or more about how Europeans sought to impose their kinship
and gender values and practices. For example, Molinas Confessional o f 1569 devoted
only 1 percent o f its teachings to idolatry, in comparison to 15 percent to sins o f the
body (Gruzinski 1 9 8 9 :1 0 0 ; see also Barnes 1992).
T h e aim o f the first 12 Franciscan friars, who arrived in M exico in 152 4 , was to
contribute to the rapid evangelization o f the entire world in order to hasten the
approaching apocalypse (Leddy Phelan 1 9 5 6 ). Unlike Jewish notions o f a Chosen
People set apart from the rest o f humanity, the Christian ideal at its most radical
(for example St Pauls there is neither Jew nor Greek, there is neither slave nor free
man, there is neither male nor female, for you are all one in Christ Jesus in Galatians
3 :2 8 ) aimed to transcend alterity altogether by the conversion o f all peoples into a
non-differentiated community o f G o d . Conversion, in other words, was presumed
to put an end to exteriority altogether, and thereby bring history to a close. It is
ironical, then, that in Spain o f the Reconquista, those populations who refused to
jo in the Christian community were subject to exclusion, in the most violent and
extrem e forms: expulsion or death. Those whose conversion was thought to be
strategic rather than sincere were labelled as impure in contrast to the clean
blood o f long-established Christian families, and subject to investigation by the
Inquisition.
The symbolism o f blood was deployed as a means o f tracking a non-Christian back
ground over several generations. It undermined the open-ended Christian message o f
salvation, since those who converted retained the stain for as long as any record or
memory o f their previous religious affiliation remained. There were attempts to apply
the idea o f impure blood to the new American converts, but it was used more as an
exclusionary device among the colonizers themselves, against those with Muslim and
especially Sephardic Jewish forebears (Wachtel 2 0 0 1 ; Gose 2 0 0 2 ).
Among the American converts, non-Christian practices such as polygamy, and mar
riage between kin up to the fourth degree, were progressively outlawed. Clerics
sought to ascertain which o f a mans wives was legitimate and to compel him to
accept her and only her, by flogging if necessary (for Yucatan see Clendinnen 1987:58).
They sought to undermine established relationships between seniors and juniors and
between parents and children, and to convert women into nurturing and feminine
stereotypes o f European motherhood (for northern Mexico see Gutierrez 1991:7 4 -7 8 ).
In general, preference was given to the patriline, especially in regulating inheritance
to high office among ruling elites, and at the same time the legal position o f women
was systematically downgraded (Silverblatt 1 9 8 7 ; Kellogg 1 9 9 5 ).
Post-Tridentine values included the dogma that wives obey their husbands at all
times; and also the idea that procreative sexual relations within marriage were a debt
that must be paid, while all other sexual activity constituted an offense against the Sixth
Commandment on adultery (Lavrin 1 9 8 9 ). Male homosexuality, the unnameable sin
(elp eca d o n efan d o), was outlawed since it was not only contrary to G ods law, but also
identified with Islam, and hence politically treasonous as well (Seed 2 0 0 1 :1 2 1 ).
The successful application o f these new rules was to be monitored and reinforced
through the practice o f confession. Foucault (1 9 8 1 ) saw the sustained probing through
the confessional o f the penitents sexual activities - and also desires and fantasies - as part
o f a process o f individualization, an attempt to create persons who were fundamentally
detached from their social context. How successful these efforts were in the New World is
obviously questionable, as the work o f 20th century anthropology helps to reveal. In
practice, most Americans did not confess at all, or misunderstood the purpose o f confession
to the point o f undermining it altogether. And contrary to the Tridentine emphasis on the
sinners free will, the belief that external forces influence behavior - whether other people
or cultural traditions more generally - did not disappear (Gruzinski 1989).
The conversion o f the Americans to Christianity was not, however, to be furthered
by intermarriage with the Europeans. As early as 149 8 , Queen Isabella issued a decree
permitting 3 0 Spanish women o f good b irth to accompany Colum buss third voy
age in order to help colonize the New World and instill Spanish morality. The Emperor
himself gave the noble daughter o f the C ou nt o f Aguilar to Hernn Corts to marry
(Konetzke 1 9 4 5 ; Miller 1991:20). In the case o f Portugal, well into the 18th century
it was not uncommon for girls to be shipped o ff to Brazil against their will, to ensure
the reproduction o f a respectable ruling elite (Russell-Wood 1 9 7 8 ). Nonetheless,
there were also a certain number o f strategic marriages between Spanish men and
high-ranking, wealthy American women (Burkett 1 9 7 8 ), and also between Spanish
women and American noblemen and royalty (Carrasco 1 9 9 7 :9 0 -9 1 ; Choque 1998).
Stoleris research on colonial rule in South and Southeast Asia in the 19th and 20th
centuries, whether Dutch, French or English, demonstrates how rigorously racial dis
tinctions between colonizers and colonized were maintained, through tight control
over both European women and the m etis w ho were born o f unions between E uro
pean men and local women (Stoler 1 9 9 7 ). In 16th century Spain and Portugal, how
ever, the concept o f raza had different connotations. While continued attempts were
made to maintain clear boundaries between colonizers and colonized, in practice
these were more typically honored in the breach. Hence the uniqueness o f Latin
America as a postcolonial population.
The Spanish initially referred to themselves as Christians in the New World as
they had at hom e, to distinguish themselves from Moors and Jews. As the Americans
converted to Christianity en masse, there was no legal reason to bar them from frill
membership in the community o f Christians o f Iberian origin (H erzog 2 0 0 3 ). How
ever, the social difference which at first had been expressed in terms o f religion, and
thence also o f blood, quickly became one o f status. American social hierarchies
were telescoped and downgraded. While elite families retained more o f their status
and privileges than has often been assumed (Restall 19 9 7 ), most Americans (known
universally as Indians) became plebeians, while all Spanish in the New World aimed to
identify themselves as gentry and therefore as social equals (Seed 1 9 8 8 :2 1 ). Despite
population loss, especially in central M exico, the native Americans remained numeri
cally dominant, worked mainly in agriculture and mining, and paid the poll tax ( tributo)
unless they were part o f the small surviving elite. The African population worked
mainly in the urban centres, often as domestic servants. Since as slaves or ex-slaves
they were not descended from free vassals o f the King, and were also natives o f another
place, they could never become full legal persons (H erzog 2 0 0 3 :1 5 9 ). The same was
broadly the case also in the Portuguese dominions (Schwartz 19 8 5 ).
Although there was no legal bar to marriages between Indians and Europeans, social
pressures ensured that in practice there were very few. On the other hand, unions were
com monplace, and as a result the children they produced were overwhelmingly ille
gitimate. Hernn Corts had fathered a child with Malintzin (M alinche), the Nahua
woman whose linguistic skills were so central to the enterprise o f conquest, but his
marriage was to a Spanish noblewoman. Thus the separation in the New World between
legal marriage and the multitude o f children created outside marriage, and generally
between different castes, was symbolically inaugurated in the name o f Christian
monogamy. Fo r many commentators the informal and unequal relationship between
Corts and Malintzin lies at the heart o f Latin American identity to this day, and
reflects the profound contradictions in the values that were brought from Europe.
Fo r some two hundred years, to be o f mixed parentage (m estizo or m ulatto) by
definition meant to be illegitimate in the Spanish colonies (Bouysse-Cassagne 1996).
Only in the 18 th century did significant numbers o f Spanish women for the first time
seek to marry mestizos, as some o f this category became prosperous with the growth
o f commercial society. But as the desire for intercaste marriages increased, so did the
formal restrictions (Seed 1 9 8 8 :1 4 8 -1 4 9 ). Even the Church, which had previously
supported such marriages when based on free choice, fell into line. In this changing
social climate, a more rigorous ban on interracial marriage was proclaimed in the
1 7 7 8 Royal Pragmatic, although given the legal status o f Indians it could only be
applied unambiguously to marriage between those o f African descent and others.
Although the Pragmatic did not fully achieve its stated aims, it signaled an increasing
and m ore explicit consciousness o f race and racial origins (Martinez-Alier 1974;
Socolow 1 9 8 9 :2 3 4 -2 3 5 ). Similarly in Brazil, although interracial and interclass mar
riages were not illegal, social pressure ensured that most marriages were between
social, racial and economic equals (Da Silva 1984).
W hile the Iberians disapproved o f marriage between different castes, they accepted
as norm al and inevitable the extramarital unions between higher status men and lower
status women that produced ever more complex mixes o f different groups (see Wade,
this volume). The Indian ruling elites inevitably saw in such practices a threat to their
own social order, not least because mestizos were exempt from paying the Indian
tribute. Thus the Peruvian nobleman, Guarnan Poma de Ayala, in his great work o f
1 6 1 5 , denounced in uncompromising language not only the hypocritical sexual
behavior o f Spanish officials and priests who raped Indian women or kept them as
concubines, but also the sexual freedom o f Indian women who flaunted their liaisons
with other kinds o f men: The author has seen many whorish indian women with
their mestizo and mulatto babies, wearing short skirts, ladies boots and hats. Even
though they are married, they consort with Spanish and black men. And others refuse
to marry an indian man (Guarnan Poma 1980:1128).
ALTERITIES: KINSHIP AND GENDER 281
T h e H o u s e in M e s o a m e r i c a n a n d A n d e a n S o c i e t i e s
I suggest that social differences in native American philosophies were not until recently
formulated in racial terms, and in this section I explore a model o f relatedness that
places less emphasis on birth than those emanating from Europe. Having said that,
the reproduction o f status does o f course typically draw on birth as a major source o f
distinction in American cultures as well as European. In the pre-Christian past, the
most extreme strategy by which high-status Americans reproduced themselves as a
distinctive category was probably that o f the Inkas, whose claim to divine origins was
reinforced by the marriage o f royal siblings, from whose sons and daughters the heirs
o f the following generation were selected. More generally, elites across the Americas
maintained their separateness through strategic exogantic marriages and patrilineal
inheritance.
At the same time, it seems that elites and commoners were bound together through
particular kinds o f ties, and I draw on discussions o f the house, and house socie
ties, to explore the constitution o f social groups in M esoamerica and the Andes.
Central to the developing literature on the house is the recognition that kinship cat
egories that have conventionally been viewed as incom patible, o r at least distinct, can
be found in combination or coexistence, in part reflecting status differences between
members o f a single aggregate which is organized around elite families or lineages
(Lvi-Strauss 1983; Carsten and Hugh-Jones 199 5 ; Waterson 1995). The continuing
debates over whether Mesoamerican and Andean descent and inheritance should be
characterized as patrilineal, bilateral, or bilineal, and whether or when endogamy is a
marriage rule, can be sidestepped by following the lead o f this literature, as well as
Schneider (1984), and privileging indigenous categories, recognizing that the consti
tution o f social relations and groups may involve different kinds o f practice. For exam
ple, I suggest that in Mesoamerica and the Andes, social units combining commoners
with elites were historically defined especially through religious and ancestral cults,
and after conversion to Christianity translated into the worship o f patron saints.
In 16th century M esoam erica and the Andes, generic term s were adopted by
the Spanish administrators to refer to social units w hich were also identified in
Spanish as p a rc ia lid a d (a division o f a larger w hole) or b a r r io (neighborhood): for
exam ple, altpetl and ca lp o lli in the Nahua-speaking regions, c b in a m it or cah in
M aya-speaking regions, and ayllu in the Andes. Populations subject to the Mexica
(o ften known as Aztecs) were obliged to pay tribute on a rotating basis, as allocated
to each unit. In the more centralized Inka state, local groupings had been
reorganized according to an overarching decimal system, including a division into
upper and lower moieties, by means o f which tribute labour was allocated. As the
Spanish to ok over this highly efficient system, they generalized the use o f ayllu to
refer to fiscal units under their respective heads, and reserved the term parcialidad
mainly to refer to the moieties.
Part o f Spanish policy was to resettle the new subject populations in villages mod
elled on the Iberian grid plan. In the Andes these villages were in some cases founded
on the sites o f Inka-period setdements, and because ayllu organization was fundamen
tal to the colonial labour service ( m ita ), the previous organization under the govern
ance o f the local nobility was largely retained. As a result, ayllu structures have
remained clear in some places even in the 20th and 21st centuries, with named ayllus
occupying the same lands over a period o f five hundred years or more (although the
precise boundaries are o f course subject to change and contestation).
In Mesoamerica, by contrast, a less centralized political field meant that each
locality retained more independence, even when tied into tributary relations with the
M exica state o f Tenochtitln (Lockhart 1991; Clendinnen 1 9 9 1 ). The Spanish
brought devastating population loss to central Mexico, through both disease and
war, and the Church played an important role in reorganizing settlements. These
congregaciones were frequently divided into neighborhoods (barrios), each with its
own patron saint and confraternity (co frad a) organized to celebrate its feast. It was
long assumed that these settlement towns, later redesignated as municipios, were
fundamentally Iberian in conception and identity, marking a profound break with
the past. This apparent contrast pervaded mid 20th century scholarship. Leading
anthropologists and historians in the Andean region (Luis Valcarcel, Jos Mara
Arguedas, John V. Murra, R. T. Zuidema, Nathan Wachtel) celebrated the continui
ties between past and present, while Mesoamerican scholars mainly emphasized dis
continuity and loss. The orthodoxy was that cultural practice and social forms in
M esoamerica were the product o f the colonial order, and retained very little o f their
pre-Spanish roots. Drawing on Ricards account o f the spiritual conquest which
asserted that the early Franciscan friars had initiated a profound transformation o f
M exican culture (Ricard 1 9 3 3 ), and expanding the work o f other North American
anthropologists (e.g. Tax 1 9 5 2 ), Eric Wolf established an influential model o f the
closed corporate peasant community, colonial in origin, into which indigenous
populations were typically concentrated (1957; 1 9 5 9 :2 1 3 -2 2 0 ). Martinez Pelez
(1 9 7 1 ) similarly declared that the cultures and social practices o f Guatemalan Indians
were substantially the product o f colonial rule.
T h e assumption o f an absolute break inaugurated by the Spanish Conquest went
together with a sense o f lack, powerfully evidenced in anthropological studies o f kin
ship. Try as they might, few anthropologists found more than occasional lineages, or
what they perceived to be fragmentary traces o f an earlier patrilineal system (Guiteras
H olm es 1 9 5 1 ; Pozas and Aguirre Beltrn 1954). In the Andes too, early studies o f
kinship took a similar course (Stein 196 1 ; Vzquez and H olm berg 1966). The pre
vailing evolutionist assumptions encouraged the representation o f highland Latin
America as a region where, even before the Spanish administration, social organization
ALTERJTIES: KINSHIP AND GENDER 285
based on large kin groups had given way to the principle o f territoriality (H unt 1976;
Nutini 1 9 7 6 ). As Wolf wrote for Mesoamerica:
The diligent ethnologist may still find, among the Otomi-speakers on the fringes o f the
Valley o f Mexico, hamlets based on common descent in the male line and enforced
marriage outside the community; or patrilineal kinship units sharing a common name, a
common saint, and a measure o f social solidarity among the Tzeltal-Tzotzil-speakers of
Chiapas, though there too they have lost their former exogamy and common residence
which they possessed in the past. But these examples remain the fascinating exceptions to
the general rule that, among Middle American Indians as a whole, common territoriality
in one community and common participation in communal life have long since robbed
such units o f any separatist jurisdiction they may at one time have exercised. (1959:220)
to ancestors o f the ruling elite and associated with particular landholdings. They were
sufficiently important that even when populations were forced to relocate their settle
ments by the Inka or the Spanish, the original ancestral shrines remained sites o f
annual pilgrimage (Howard-Malverde 1 9 9 0 ; D el R io 1 9 9 7 ). The encomienda certifi
cates that granted the use o f Indian Labour to individual Spaniards make it clear that
ayllus were identified with a leader, and grouped together under the broader leader
ship o f a noble lineage. The relatedness within such groups was mediated through the
shared land that fed them and whose ancestral shrines they provided for.
These criteria cannot o f course apply to comparable units today, since the ruling
elites have disappeared, reclassifying themselves as mestizo or ladino, and the ancestral
cults were a particular target for the priests who sought to eradicate idolatry. A few
anthropologists have indicated that the house is a key organizing concept for under
standing 20th century kinship practices in M exico (Monaghan 1 9 9 6 ; Sandstrom
2 0 0 0 ), and others working in the Andean region have also noted the importance o f
house imagery (Mayer 1 9 7 7 ; Arnold 1 9 9 2 ; Gose 1 9 9 4 ; Yapita and Arnold 1 9 9 8 ).
However, beyond individual dwellings, a contemporary practice similar to pre-
Christian ancestral cults in the Andes and Mesoamerica is the organization o f saints
day celebrations. Numerous studies o f barrios and municipios in 20th century
Mesoamerica, and o f ayllus or peasant com munities in the central Andes, signal the
importance o f rotating ritual obligations to the functioning o f the unit, and demonstrate
how traditional authority structures were linked to the obligation (cargo) to celebrate
the feasts o f the saints. This complex o f cargos, cofradas and the saints day/ms-has
conventionally been seen as evidence that todays indigenous communities are o f
colonial origin. However, these southern European Catholic institutions may have
been accepte'd with alacrity because they could be mapped onto established practice.
Carlsens study o f Santiago Atitln provides evidence o f such a process. H e notes
the imagery o f vegetable growth and decay in present-day cofradas such that the
elders (principales) are identified as the trunk and the ancestors as roots, with clear
pre-Christian roots. B ut the bundle cult, whose origins lie in the Maya Classic
period, in the form it is practiced today is associated with Christian saints (Carlsen
1997). In similar vein Hill and Monaghan suggest that the patron saint o f each canton
in Sacapulas today is like the founding ancestor in the myth o f origin (1 9 8 7 :1 3 ) , and
Monaghans (1 9 9 5 ) study o f Mixtee Nuyoo today argues that Nuyootecos realize the
great house that is their community through cargo service. Arnold ( 1 9 9 3 ) proposes
for the Qaqachaka ayllus o f highland Bolivia that the patron saints, associated with the
founding o f the pueblo, are a Christianized form o f the mummified ancestors. Finally,
Urton has emphasized the central importance o f patron saints and their fiestas in the
articulation o f ayllus o f Paqariqtambo (southern Peru), suggesting links with pre-
Christian cults (1 9 9 0 :9 9 ).
Earlier generations o f anthropologists recognized the importance o f the fiesta sys
tem, but were hampered by the conventional divisions o f ethnological knowledge.
Thus Tax and Hinshaw, writing o f the Maya in the midwestern Guatemalan high
lands, note: Perhaps the institution most im portant in binding the household to the
community is the civil-religious hierarchy, and traditionally a sure criterion o f o n es
acceptance o f and by the community has been community service through this insti
tution (1 9 6 9 :8 8 ). This observation is made under an overall heading o f political
and religious organization. Whether we see the civil-religious hierarchy as a form o f
kinship is open to question, since members o f such groups were not necessarily
genealogically related, but it surely constitutes a fundamental ground o f relatedness in
the region.
These units today (barrios, municipios, peasant communities, ayllus), associated with
particular landholdings, are often strongly endogamous, but endogamy never seems to
be a rule (Hunt and Nash 1969; Hickman and Stuart 1 9 7 7 ; Hill and Monaghan 1987;
Snchez Parga 1990; Ossio 1992). On the contrary, where it is politically or socially
disadvantageous, other marriage practices take precedence (e.g. Skar 1998; Spedding
1 9 9 8 ). In the central Andes it is striking how rarely ayllus are imagined as groups o f kin
in spite o f the high rates o f endogamy recorded in some cases (although Isbell 1978:105
documents an exception in Ayacucho, southern Peru). Land is associated with particu
lar ancestors (and more o f them male than female) who worked it and handed it on to
their descendants, and these forebears are remembered in rituals. When there is great
pressure on resources, exclusionary forms such as those emphasizing patrilineal inherit
ance may prevail (Abercrombie 199 8 ; Arnold 1998:28). B ut common ayllu member
ship requires people to cooperate and to exchange labour in highly formalized ways, to
mourn their dead, and to celebrate the patron saint or saints together.
It is through meeting these obligations that people have the right to work the land
(G ose 199 4 ; Harris 2 0 0 0 ). The emphasis on duties to the collective is extraordinarily
strong in both Mesoamerica and the Andes, and fall personhood has until recently in
important senses been a function o f membership o f the group. As such, the decline -
or outright abolition - o f collective duties in recent decades, and the growth o f legally
constituted individual property rights in land, strike at the very heart o f indigenous
American understandings o f personhood, relatedness and identity.
The commensality at the heart o f ritual practice is a powerful statement o f the
bonds constituted and reproduced through nurturing and the giving o f food
(Weismantel 1995). Similarly, the territoriality o f these units, long associated by
evolutionary thought with the political domain, may be the grounding o f a
relationship realized through the work o f agricultural production and consumption.
Seen from this perspective, the question o f whether, and to what degree, units such as
the calpolli, the altpetl, cah, chinamit- or ayllu were or are kinship groups ceases to be
salient. Endogamy, where practiced, is a reiteration o f solidarity and shared interests
within such units, but where circumstances dictate, they can easily absorb outsiders,
so long as they fulfill their obligations in the form o f labour and ritual and festive
contributions; those who did not, or do not, fulfill these obligations are expelled.
T here is, then, often a pragmatic quality to the attribution o f kin ties.
The centrality o f religious cult in this understanding o f relatedness may also help to
explain the importance o f godparents in these societies. Compadrazgo is generally
assumed to be an institution o f Mediterranean origin, but the speed and extent to
which it was adopted in Mesoamerica and the Andes suggests that this supremely
optional form o f kinship also had American roots, corresponding to the fluidity and
openness o f local cultures o f relatedness. Compadrazgo ties can be created in many
different ways, and on the basis o f many different relationships. They can be between
close kin, virtual strangers, or patrons and their clients. In some areas anthropologists
have noted that individuals may have as many as ten different godparents, and power
ful patrons may have compadres and godchildren running into the hundreds (Laughlin
1 9 6 9 ; Nutini 198 4 ; Spedding 1998).
However, it would be distorting to emphasize only fluidity and the potential for
inclusion. The same features o f these social groups are also a means by which opposi
tion, competition and alterity may be articulated. The fact that incomers might, and
may, become part o f a particular cult- and Iand-group does not mean that their differ
ential status would be quickly forgotten. In both the Andean and Mesoamerican regions,
for example, the prevailing dual organization in at least some cases originated precisely
as a means o f differentiating conquerors or in-migrants from the established population.
Dual divisions - usually between upper and lower moieties - remain an important way
in which boundedness and opposition are expressed within a larger landholding unit.
W ithin such groups too, for all their potential open recruitment, certain relation
ships must have embodied differentiation. In the past status distinctions were key
markers o f alterity. In recent times - at least in the Andean region - it is arguably rela
tions o f affinity that most obviously play out antagonism and opposition within any
land-group.
A f f i n i t y a n d A l t e r it y : A m a z o n ia a n d t h e A n d e s
W o m e n , M e n , a n d D if f e r e n c e
16th and 17th centuries came from the Catholic Iberian peninsular, this new diffusion
across the globe o f gender and kinship values and practices originates from Protestant
northern Europe and America. These globalizing values are once again presented as a
civilizing force, and are used to negotiate the faultlines o f existing kinship and gender
relations. How, and how far they will produce a radical transformation o f Latin
American sociality is yet to be determined.
REFERENCES