YALE PUBLICATIONS IN THE HISTORY OF ART, 30

G eorge L. H ersey, ed ito r

STUDIES IN
ANCIENT AMERICAN
AND EUROPEAN ART
The Collected Essays of George Kubler
edited by

Thomas F. Reese

YALE U N IV E R S IT Y PRESS
New Haven and London

Copyright O 1985 by Yale University. All rights reserved.

This book may not be reproduced, in whole or in part, in any
form (beyond that copying perm itted by Sections 107 and
108 o f the U.S. Copyright Law and except by reviewers for
the public press), without w ritten permission from the publisher.
Designed by Nancy Ovedovitz and set in Garam ond N o. 3
type by T he Com posing Room o f Mich., Inc. Printed in the
United States o f America by Murray Printing Company,
W estford, Massachusetts.

Library o f Congress Cataloging in Publication Data

Kubler, G eorge, 1912
Studies in ancient American and European art.
(Yale publications in the history o f art ; 30)
Bibliography: p.
Includes index.
1. A rt America. 2. Indians Art. 3. A rt, European.
I. Reese, Thom as Ford. II. Title. III. Series.
N 6501.K 83 1984
700
84-13216
ISBN 0-300-02662-5 (alk. paper)
T he paper in this book meets the guidelines for perm anence
and durability o f the C om m ittee on Production G uidelines
for Book Longevity o f the Council on Library Resources.

10

Contents

Authors Preface

XI

Editors Preface

xiii

Editors Introduction

xvii

An analytical table of contents appears at the

beginning of each o f the four parts. The sources
of the rdeles are noted there.
I. C O L O N IA L A R T
Introduction
Prologue: Colonial Transformations in Ancient Italy
1.1 Some Etruscan Versions of
Corinthian Ceramics
9
The Focillon Legacy: The Interactions ofForm and
Technique in tbe Colonial Arts of the United States,
Latin America, and West Africa
1.2 The Rebuilding of San Miguel at
Santa Fe in 1710
16
1.3 Ucareo and the Escorial
27
1.4 The Machine for Living in
Eighteenth-Century W est Africa
31
Culture Contad, Anthropology, and A rt History
the Forties: Case Studies in the United States,
Mxico, Paraguay, and Per
1.5 Two M odes o f Franciscan
Architecture: N ew Mxico and
California
1.6 [Selections from] T he Q uechua in
the Colonial World
1.7 [Selections from] The Tovar
Calendar: An Illustrated Mexican
Manuscript ca. 1585

in

34
39

51

1.8 Santos, An Exhibition of the "

Religious Folk A rt of N ew Mxico

61

Post-Panofskian Studies of Form and Meaning in

the Sixties: Renaissance and Disjunction in Latin
American A rt
1.9 On the Colonial Extinction o f the
Motifs of Precolumbian A rt
66
1.10 Indianism, Mestizaje, and
Indigenismo as Classical, Medieval,
and M odern Traditions in Latin
America
75
1.11 N on-Iberian European C ontri
butions to Latin American Colonial
Architecture
81
Studies in Latin American Urbanism: Some
Different Vieiving Distances and Focal Points
1.12 The Unity o f Cities in the
Americas
1.13 The Colonial Plan o f Cholula
1.14 O pen Grid Town Plans in Europe
and America, 15001520

88
92
102

Fine and Plain Arts: Sociological, Economic, and

A rt Historical Perspectives on Latin American Art
from Precolumbian Times to the Nineteenth Century
1.15 The Arts, Fine and Plain
111
Figures
following page 118
II. S P A IN , P O R T U G A L , A N D E U R O P E
Introduction
125
The Focillon Legacy: The Interactions ofForm and
Technique in Late Medieval Spanish Architecture
II. 1 A Late Gothic Com putation of Rib
Vault Thrusts
126

VI
The Morphology ofTypes: Uique Forms in
Seventeenth- and Eigbteenth-Century Spanish
Architecture
II.2 Camarines in the G olden Age

Contents
III.2 Toward Absolute Time: Guano
Archaeology
136

Disjunction and European Transfer Points in the

Sixteentb Century: Formerm dung, Extended
Classes and Trace-Elements at the Escorial
II. 3 Francesco Paciotto, Architect
140
11.4 Palladio and the Escorial
151
11.5 Galeazzo Alessi and the Escorial
157
Beyond Connoisseurship: Ttvo Paintings at Yale.
Studies in Attribution, Identification, Seriation,
Dating, and Replica Masses
11.6 The Portrait o f H ernando Cortes
at Yale
162
11.7 The Portrait o f Fray Juan de San
Bernardo A ttributed to Valdes
Leal
166
Beyond Iconography: Studies of the Status of the
Artist and of the Dynamic Interrelationships
between Text and Image in Seventeenth-Century
Painting
11.8 Vicente Carduchos Allegories of
Painting
176
11.9 Three Remarks on the Meninas
184
Iconographical Programs, Text and Image, and
Philip II: Studies in Italian Numismatics, Spanish
Painting, and Portuguese Architecture
11.10 A Medal by G. P. Poggini
Depicting Per and Predicting
Australia
189
11.11 The Soul ofSt. Philip [Philip II] by
Murillo
193
11.12 The Claustral Fons Vitae in Spain
and Portugal
204
Figures
following page 208

III. ANCIENT AMERICA

Introduction

217

The Beginnings: A rt History and Archaeology at the

Institute of Fine Arts and at Yale in the Thirties
and Forties
III. 1 The Cycle o f Life and Death in
M etropolitan Aztec Sculpture
219

225

Architecture as Space: A rt Historical Contributions

to Diffusion, Periodization, and Chronology in the
Fifties
111.3 The Design o f Space in Maya
Architecture
242
111.4 Polygenesis and Diffusion:
Courtyards in Mesoamerican
Architecture
251
Post-Panofskian Studies in Form and Meaning
the Sixties: Iconography without Texts
111.5 Precolumbian Mural Painting
111.6 The Iconography o f the Art of
Teotihuacan
III. 7 Iconographic Aspects o f Architec
tural Profiles at Teotihuacan
and in M esoamerica

in
256
263

275

Intermezzo: A n Agenda for Research and Ancient

America Revisited
111.8 Early Architecture and Sculpture
in Mesoamerica: Com m entary on a
Paper by T. Proskouriakoff
280
111.9 The Styles of the Olmec Colossal
Heads
288
Disjunction, Transfer Points, Time Frames,
Iconographic Programs, and Interrelationships
between Text and Image in the Seventies: Studies in
Maya A rt
III. 10 Vicos Idea o f America
296
III. 11 Mythological Ancestries in Classic
Maya Inscriptions
301
111.12 The Clauses o f Classic Maya
Inscriptions
308
111.13 The Paired A ttendants of the
Tem ple Tablets at Palenque
316
111.14 The D oubled-Portrait Lintels of
Tikal
322
111.15 Mythological Dates at Palenque
and the Ring N um bers in the
Dresden Codex
327
III. 16 Aspects o f Classic Maya Rulership
on Two Inscribed Vessels
334

Contents
Theoretical Reflections on Continuities and
Discontinuities in Precolumbian A rt and History
* III. 17 Renascence and D isjunction in the
Art o f M esoamerican Antiquity
351
Bibliography
361
Figures
following page 372

IV. METHOD A N D THEORY

Introduction

377

Homage to Henri Focillon

IV. 1 H enri Focillon, 1881 1943
IV.2 The Teaching o f H enri Focillon

378
381

On Artists, Historians, Scientists, Evolution, and

Paradigms
IV.3 Style and the Representation of
Historical Time
386
IV.4 Com m ent on Vanguard Art
391

Ancient America Revisited

IV .5 Period, Style, and Meaning in
Ancient American Art
IV.6 H istory or Anthropology of
Art?

The Shape o f Time Revisited

IV. 7 A Talk with G eorge K ubler [An
Interview by R obert Joseph
Horvitz]
IV.8 Towards a Reductive Theory of
Visual Style
IV.9 The Shape of Time Reconsidered

Vil

395
406

413
418
424

A Bibliography o f W orks by G eorge Kubler

(to 1982)
431
437
Index

Introduction

Ancient America, which contains seventeen anieles,

opens with two sections representing K ublers
teaching and scholarship on Precolumbian art be
fore the publication o f The A rt and Architecture of
Ancient America in 1962. Like a submerged reef,
these studies em erge only occasionally above the
surface, but are known to experienced observers
who know how they inform and guide his work in
other fields. t h e b e g i n n i n g s joins an article on
Aztec Mxico with another on Preconquest Per,
echoing similar alignments in K ublers studies on
sixteenth-century Mxico and the colonial
Quechua. The first reflects his studies with Spinden between 1936 and 1938; the second his collaboration with Yale colleagues B ennett and
Hutchinson after 1940. a r c h i t e c t u r e a s s p a c e
contains two articles that represent, together with
the first article of the subsequent section, a rejected 195051 draft o f Ancient America in which
Nuclear America was treated as a unit with main
divisions by architecture, sculpture, and painting
that were divided in turn into topical discussions.
Both apply art historical concepts about architec
ture as space to the analysis o f Precolumbian art.
The last four sections o f part III represent
Kublers continuing contributions to Precolum
bian scholarship following the com pletion o f A n
cient America in 1959. M ost deal with iconographical issues that were given m om entum by (1)
Proskouriakoff s breakthroughs regarding the his
torical nature o f Maya art, (2) Panofskys Renaissance and Renascences in Western Art, which appeared in 1960, and (3) personal questions about
the limits o f iconography raised in The Shape of
Time, p o s t - p a n o f s k i a n s t u d i e s i n f o r m a n d
m e a n i n g contains three articles that propose new
methods for reconstructing meaning at Panofskys

second level, w here texts are absent. The first

treats form and meaning simultaneously; and the
second and third reconstruct meaning through the
use o f configurational analysis to minimize the
dangers frequently engendered by the disjunctive
nature o f the bonds between form and meaning.
i n t e r m e z z o , the fourth section, includes two ar
ticles that reassert the im portance o f art historical
perspectives and offer new deas about problems
first raised in Ancient America, d i s j u n c t i o n ,
T R A N S F E R P O I N T S , T IM E FR A M ES , I C O N O G R A P H IC P R O G R A M S , A N D

I N T E R R E L A T I O N S H I P S BE

the penultim ate fifth

section o f part III, consists o f seven articles o f the
1970s that extend many o f Panofskys concepts
about renascence and disjunction. They parallel
K ublers work on European subjeets and study the
problems o f iconographic programs, text and im
age, disjunction at transfer points, and the nature
of Maya thought about time. They were natural
extensions o f K ublers investigations of the 1960s,
when he defined the m ajor iconographical configurations o f Teotihuacan in The Iconography of the
A rt of Teotihuacan, 196567, and o f Classic Maya
art in Studies in Classic Maya Iconography, 1967
69. In the new studies, he turned to the study of
the condnuities and disjunctions that occurred
where these two civilizations met. Indeed, he attem pted to lcate the exact historical moments at
which the Mexican dynasties transferred their authority and cults to the Maya area. H e had long
been interested in the junctures w here cultures
met, but, with the im portant exception o f the
ChichnTula transfer point, he had not studied
them previously in M esoamerican art, probably
because o f the rudim entary State of knowledge
about Maya history before 1960. t h e o r e t i c a l
TW EEN t e x t a n d i m a g e ,

217

A ncient Am erica

218

R E F L E C T IO N S O N C O N T I N U I T I E S A N D D I S C O N -

the sixth and final section, consists o f

an article that represents K ublers most extensive

t in u it ie s

discussion of the question o f renascence and dis

junction in Precolum bian art, the central issue in
most of his work after 1960.

III. 1
The Cycle of Life and Death
in Metropolitan Aztec Sculpture

Among the great achievements o f plstic art the

world over, Aztec sculpture is to be accounted as
one of the climactic events, as one of the points
along the ligne des hauteurs of which our late, beloved teacher H enri Focillon spoke so eloquently.
This sculpture, however, is known most imperfectly to the students of plstic form. It has suffered long oblivion for a variety o f causes. After its
initial misfortune o f being discovered during the
Counter-Reformation, it passed from the cabinets
de curiosit of the eighteenth century into the ethnological collections of the century o f Darwin and
Ruskin, always just beneath the attendon o f the
historians and critics o f art.1 In an age when artistic
preferences were for verisimilitude and heroic
morality, Aztec sculpture satisfied none of the
touchstones o f taste.
Today its position in the history of art is tentative and problematic, for it incorporates an inde
pendent aesthetic, devoid o f any historical filiation
whatsoever with the vast linked perspectives of
Asiatic and European art.2 Y et it is precisely here
that the inestimable valu of ancient American art
may be assessed in that an American humanity,
clinically isolated, so to speak, from all extracontinental influences of an historical nature, achieved
an expression in the language of visible form
which is an ndex to all humanity, in our widest
possible intelligibility, in our most stable, our
most universal aspect, beyond the local and specif
ic limitations.3
M etropolitan Aztec sculpture may be categorically differentiated from all other ancient
American art by its striking conquest o f certain
aspects o f em otional expression. The m odern stu-

dent can disregard the barriers o f time and culture

in the examination o f a specimen such as the head
of a dead man, in the National Museum of Mxico
(fig. III-l). The em otional contagion from these
forms, expressive of the pathos o f organic dissolution, is inescapable. Regardless o f time and cul
ture, the work has an authority derived from mature sculptural tradition, from rich sensibility, and
from empiric observation. A t the other extreme,
the serpent heads o f Aztec workmanship display a
raw vitality unprecedented in American art, secured by reorganization and emphasis o f significant particulars (fig. III-2). These vales of Aztec
sculpture are lacking in the generalized, idealizing
art of the Mayas of Southern Yucatn and
Guatemala. In Maya art, the repertory of human
expression contains types analogous to those of
Buddhist art, limited by the contempladve, introverted figures o f the stelae on the one hand, and
on the other by the apotropaic countenances of
monsters. Between these extremes lie the many
variants of local style and of iconographic types,
but in general, human expression lacks diversity
and subtlety o f inflection. A rough equivalent for
the expressive pow er o f Aztec sculpture is to be
found only in the plstic art o f the Mochica potters
of the north coast o f Per. There, however, the
concern with expressive content and with the val
es o f affecdve states o f being is perhaps a byproduct of the art o f detailed, accurate portraiture.
In general, Mochica pottery tends rather toward
caricature at a low level o f generalization.
To analyze the sculpture o f the Aztecs requires
some preliminary definition o f the term Aztec itself. It signifies both a culture and a political entity
219

220

A ncient Am erica

that were roughly coterminous. A dherence to the

political structure was through m em bership in a
loose tribal confederacy held together by military
forc. U pon occasion, however, as with the people
of the republic of Tlaxcala,4 an ethnic group participated in the culture w ithout belonging to the
political organism. The boundaries o f the Aztec
culture-state extended from northern Mxico
deep into Central America, with isolated enclaves
as far south as Nicaragua and Panama.5 Communal
agrarian living was the basic structure o f the society; originally heterogenous and local, the political
control carne to be centered at the time of the
Conquest by Spain in the hands of M ontezuma
(Motecuzoma), who converted rule by council
into the forms o f absolute dynastic governm ent.6
Tenochtitln, the island city in the great intervolcanic Valley of Mxico, became the administrative m etrpolis during the fifteenth century.
Its maintenance was secured by tribute levied
from the subject provinces.
The principal traits o f Aztec culture appear to
have been widely diffused throughout Middle
America rather earlier than the military conquest
and political consolidation o f the area by the armies of Tenochtitln. The Nhuatl, or Aztec lan
guage, served as an administrative lingua franca.
Human sacrifice was assigned a cardinal position
in the religious life o f the culture, and ceremonial
warfare for the purpose o f securing sacrificial victims became a dom inant them e in Aztec behavior.
An intricate ritual calendar wras observed for the
worship o f a populous, syncretic assembly o f supernaturals.7 Pictographic annals recorded the his
tories of the com ponent tribes.8
The singular fact, nevertheless, is that Aztec
sculpture cannot be said to have assumed the form
under which we know it until the middle of the
fifteenth century. Before then, its constituent ico
nographic and ceremonial elements can be identified as the property o f scattered ethnic groups;
then, precipitately, the process o f integration occurred.9 O f m onum ental, m etropolitan Aztec
sculpture, the preparatory, experimental stages
have not been identified. T here is no evidence that
any o f the Aztec pieces m entioned or illustrated
here was produced, say in the fourteenth century.
Tenochtitln was founded only c. 1325,10 and for

another century its society lacked the resources,

the leisure, and the stability to produce monu
mental sculpture. It cannot be said to have become
a metrpolis until the second quarter o f the fif
teenth century, after the subjugation o f its rivals,
Atzcapotzalco and Texcoco, in the Valley o f Mx
ico. In fact, the only specimens of metropolitan
Aztec sculpture bearing any suggestions as to the
date o f their manufacture belong to this late peri
od in Aztec history, certainly not before the reign
of M ontezum a the Eider (144069).
W hat the position o f these datable pieces in the
history o f Mexican form may be, now evades definition. This much is certain: that Aztec work of the
fifteenth century rests upon a m ore ancient sculptural tradition, perhaps derived largely from
Southern Mxico and dependent upon craftsmen
nourished by older civilizations, such as Olmec,
Totonac, and Mixtee. W ith them came technique
and formal command, but the expressive power
was supplied from within Aztec culture itself. The
situation is perhaps analogous to that o f the origins
of Gothic sculpture in the lie de France: new
sculptural vales were associated with the emergence o f a new political entity, but the main
doeuvre was drawn from the older late Romanesque traditions of Burgundy, Languedoc, Provence, and Lombardy.
The conditions under which these works were
produced in Tenochtitln may be partially reconstructed, in spite o f a confusing and fragmentary
documentation. N o individual artistic personalities can be identified. Sculpture was the projection o f communal solidarity. The monumental
specimens were the work o f many individuis.
Metal tools were still a novelty, and metal itself
was principally dedicated to ornamental purposes.
The technology remained that of a people in the
lithic horizon o f material equipm ent. The heavy
labor of quarrying and transportation was proba
bly achieved by tribute-laborers. T he actual carving was done by a m em ber o r members o f a craft
organization o r community, such as those of
Atzcapotzalco (goldworking) or Culhuacan (pot
tery manufactures). T he labor itself was accounted
as a form o f tribute, and the finished work was
assigned to agiven cult and its priesthood. M ostof
the specimens discussed here were recovered

The Cycle of Life and Death in Aztec Sculpture

from the great cult center, with its scores of tem
ples, which stood upon the site o f what is now the
Cathedral plaza o f Mxico City (the Zcalo) and
upon the areas just northeast o f the plaza.11
The correct iconographic identification o f these
pieces remains problematic. The great Franciscan
ethnographer o f Aztec culture, Fray Bernardino
de Sahagn12 drew information from Indians who,
when describing the attributes, costumes, and appurtenances o f the supernaturals, referred not to
stone figures such as ours, but to cult-figures made
of less perm anent materials, to the god-impersonators who played so im portant a role in Aztec
ritual life, and to the pictorial representations in
the cdices. In general, m oreover, Aztec religin
would seem to have centered less about the stone
cult-figure than we would expect from the great
number of surviving examples. The living ritual,
the procession, and the ceremonial itself were
more immediate than the inert, inaccessible fig
ures hidden within the shrines. In addition, the
stone cult-figure usually achieved individuality
through masks, costume, and attributes; divested
of these, the statue lacked religious individuality.
What, then, is the expressive content o f this
sculpture? We shall find its dom inant themes in
the representations o f animate forms. It will be
seen that a striking difference separates the pieces
chosen here for illustration. O n the one hand, we
have works representing the human figure (figs.
III-l and III-3 to III-8); on the other are those
showing plant or animal, but mainly animal forms
(figs. 111-2 and III-9 to III-12). It will further be
seen that radically different modes of expression
were tapped for these two classes o f representa
tion. W here human beings are concerned, vitality
and animation are at a mnimum; where animals
are portrayed, their vital principie is given an exaggerated, maximal expression.
Human beings are shown in the flaccid postures
of relaxation and distention (fig. III-3). The muscles of the body are loose and unstretched. The
facial expressions are those o f surrender and resig
naron, o f the undoing o f all tensin. Thus in the
numerous standing figures of men and women;
their postures, proportions, and gestures report
the most static, passive, and unalive aspects o f the
human form (fig. II1-4). The gestures are soft and

221

empty; the posture suggests that of the sleepwalker or ghost, suspended between distinct states
of animation, and really belonging to neither (fig.
III-5). In certain other examples, the human form
is associated with the symbols o f death, such as the
fleshless skull, the skeletal joints, and the vacant
eye-sockets (fig. II1-6). Elsewhere, we find the
plstic representation of an im portant rite in Aztec
culture. A human being is shown, manifestly alive,
but wearing the flayed skin o f another human
being as a costume (fig. III-7). Such figures pertain
to the cult o f Xipe Totee, whose rites were associ
ated with the renewal of the earths fertility. 13 In
the standing figure o f the Museum of the Ameri
can Indian in N ew Y ork City (fig. III-8), the mini
mal vitality o f the w earer is countered by the em p
ty, dead flaccidity o f the garm ent o f human skin. A
state o f tensin exists between the garm ent and its
wearer, between surface and substance, between
appearance and reality. All such representations
figure a state o f being which is neither life or
death, butinterm ediary between them. In general,
then, Aztec sculpture usually alludes to the human
figure either directly with the signs and symbols of
death or with a repertory o f expressions suggesting the extinction o f life and the proximity of
death. Rarely can one identify an explicitly vitalist
concept o f the human form. Upon the evidence of
the sculpture alone, w ithout further examination,
there emerges the idea o f Aztec humanity as a
quiescent, passive race, obsessed with the symbolism o f death. It is singular, for instance, that
among anthropom orphic figures, no apotropaic
specimens occur, no frightening guardian figures
such as we know from Asiatic art, and no Medusas
or Gorgons. In the portrait o f Aztec man by his
own sculptors, there is lacking the daemonic con
cept of humanity. The image of man is rarely disturbed by m onstrous metamorphoses.
To return for a m om ent to the iconographic
problem: the question may arise w hether it is in
fact the portrait of man with which we are dealing.
May it not rather be the portrait o f deity, the likeness o f an assembly o f supernaturals? The experience o f other cultures and styles teaches that the
image o f deity is the portrait o f perfected man.
Elsewhere in the history o f sculpture, the images
of anthropom orphic supernaturals incorprate

222

A ncient America

those expressions of the social attitudes that are

expected from the human celebrants o f the cult in
question. The gods o f the G reeks show the harmony of spirit and body that was the ideal o f life. The
figures of the saints on the portis o f the churches
of the twelfth and thirteenth centuries reflect the
faith and works expected o f Christian communicants. In the seventeenth century, the ecstatic pos
tures and enraptured expressions of the saints
served as models for the exercises prescribed by
devotional literature. In each age, then, the portrait of man, perfected according to the ethical
objectives o f his society, is to be found in the im
ages o f the gods, the saints, and the supernaturals.
The mom ent, however, that one turns from the
search for the figure o f mankind to the world of
animal form, a vitalism of striking intensity is
revealed in virtually every Aztec image of an ani
mal or plant. The nonhuman aspects o f vitality
seem to become the sculptors subject matter. All
those aspects o f animal energy which cannot be
duplicated or simulated by human beings are the
themes o f representation.
Typical is the artists preoccupation with snakes
and serpents. T heir sinuosity and their deadly
striking power are manifestations of a vitality alien
to humanity. The them e o f the coiled serpent, p re
pared to strike, is among the most common in
Aztec sculpture (fig. III-9). An elaboration of it is
found in the numberless representations o f feath
ered serpents (figs. III-2, 111-10). The symbolic
meaning o f the concept is poorly understood. As a
vital expression, however, the feathered serpent
synthesizes two modes o f movement; the flight of
birds and the legless undulation o f snakes. N either
of these is human, but they both excite w onder
and, often, fear. The variations upon this double
theme of feathered sinuosity are numberless, and
the great stone reptiles that adorned the walls surrounding the tem ple precinct in Tenochtitlan are
among the most splendid examples o f Aztec sculptures (fig. III-2).
The claws and the dentition o f the great felines
were likewise seized upon by the sculptor as forms
incorporating the daemonic, inhuman powers of
nature. In the large sacrificial vessel commonly
assigned to the cult o f Tezcatlipoca (fig. III-l 1),
the dynamism o f the entire body of the tiger is

maximal: each stylization contains caricatural exaggeration drawn to the brink o f abstraction. The
result is neither caricature or abstraction, but reduction to the single impression of daemonic ani
mal vitality.
An extraordinary figure o f a pum a in the Na
tional Museum o f Mxico (fig. III-12) likewise expresses the savage, mindless vitality of the great
felines. The formidable teeth are bared; the eyes,
once inset with polished stones, are heavyrimmed, watchful voids. The coat of the animal is
matted and coarse. If human beings are displayed
at moments o f surrender or lessening of vitality,
the animals are exhibited in postures suggesting
the utm ost defense o f life and the instinctive reflexes of defense against attack upon life. Among
less ferocious animals, the toad is shown just be
fore his leap, at a m om ent when his energies are
collected and focused. A gigantic porphyry grasshopper also exhibits his preparation for sudden
and surprising m ovem ent (fig. III-13).
Plant forms were frequently represented by the
Aztec sculptor. A basalt figure o f the nopal cactus
emphasizes the fleshy rigidity o f the mature form.
A porphyry calabash shows the ripe, perfected
fruit (fig. III-14). Thus, w hether plant or animal be
represented, the most m ature and vital phase of its
existence is selected for study by the craftsman.
The process o f organizing forms is the same in all
instances. The sculptor isolates from reality those
traits that suggest the expression he desires to
achieve; these traits then are developed to a point
beyond caricature, yet short of abstraction, at
which expressive power reaches a mximum.
With humans, then, expression is static and
moribund. W ith animals, it is dynamic and vital.
To what relationships in the structure of Aztec
thought does such a peculiar assignment of ex
pressive vales correspond? We may infer from
the evidence o f the sculpture alone that man was
regarded as a phenom enon distinct both from the
supernaturals and from the animals. The metaphysical powers o f the supernaturals were expressed by means of relatively abstract symbols
and attributes. Such were the smoking mirror of
Tezcatlipoca, symbolizing omniscience, the conch
shell of Quetzalcoatl, perhaps signifying his relationship to birth and genesis, the rattle staff of

T h e Cycle o f Life and Death in A ztec Sculpture

Xipe Totee, or the costum e and facial paint of
many other deities. These various powers lay be
yond the reach of man and nature. The animals, in
turn, were distinguished by the manifestations of
feral vitality peculiar to each genus. The nature of
mankind, finally, found expression in the attitudes
of resignation, surrender, and sacrifice.
The recorded literature o f the Aztecs yields
many confirmations o f this structure o f the cos
mos. The relationships among mankind, the supernaturals, and the rem ainder o f the animate
world were conceived as those of mutual interdependence, originating in the creation o f the physical universe by the supernaturals. In the Aztec
narratives o f genesis, the gods created the world
from their own blood. Thereafter the super
naturals themselves were dependent upon the cre
ated world for their further sustenance, which was
composed of sacrificial offerings o f the human
heart and of human blood. H enee the perpetual
repayment o f the archetypal blood gift through
human sacrifice became m ankinds obligation. In
return for their sustenance, however, the super
naturals bestowed annual renewal o f life upon the
animate world. T he hymn to Yacatecutli expresses
the relationship:
Y o proporciono el sustento,
pero ahora el alim ento que tom o
es el agua del corazn,
que habis venido trayendo para m
a travs de las arenosas m ontaas.14

In another poem , the raingod, Tlaloc, speaks:

Ah, yo he sido creado,
mi deidad es venerada con sacri
ficios de sangre y con ellos es
festejada.
Yo produciendo la lluvia
me he revelado com o d io s.15

Thus the supernatural donors o f life needed

mankinds collaboration to maintain the vital
rhythm o f the universe. Since mans donation of
his own blood, however, was the act o f his free
will, the maintenance o f the universe was contingent upon mans proper discharge of the blood
debt. H enee the ritual of human sacrifice may be
assessed as the rite that maintained the equilibrium of the universe. The ceremony of blood-

223

offering assumed many different forms. All members of the society regularly donated blood drawn
from the ears, the tongue, and other body parts.
At frequent ceremonies men were sacrificed by
removal o f the heart (fig. III-15), by flaying, by
gladiatorial sacrifice, and by other means. The recruitm ent for sacrifice was achieved mainly
through ceremonial warfare against other tribes,
conducted for the express purpose of securing
prisoners for sacrifice. As the volume o f human
sacrifice increased, so was the fertility of earth
and animals augmented.
The humanistic, M editerranean valuation of
human life as an entity for fulfillment and selfrealization was totally alien to Aztec culture. Sahagun recorded the standards whereby the godim personator was selected for sacrifice as
Tezcatlipoca. The passage reveis the Aztec concept of beauty.16 The m ore perfect, the more
promising the specimen o f the race, the more fitting was he regarded as a subject for offering to the
gods. Thus the concepts o f death and human perfection were intimately associated. Sacrificial
death was the appropriate culmination to the
flawless individual existence. The fulfillment of
the individual lay, not in self-realization or in the
exercise o f congenital gifts and acquired training,
but in the ritual surrender o f life itself.
These eyelieal exchanges o f vitality yield an interpretation for the peculiar distribution o f expressive vales observed in the sculpture. Man
was the consum er o f plants and animals, but he
was also the producer of divine sustenance
(human blood). The supernaturals consumed
human blood and produced the vitality of plants
and animals. The raw, mindless animation of the
physical world was transm uted by humanity into a
substance suitable for the nourishm ent of the
gods. Mankind was distinguished from the re
mainder of creation by a godlike responsibility,
expressed in forms of the highest, even repulsive
austerity, since man himself was the eucharist.
NOTES
1 R obert G oldw ater, Primitivism in Aiodern Painting (New
Y ork, 1938), chapters I and II.
2 A. V. K idder, Looking Backward, Proceedings of the
American PhilosophicalSociety, LX X X III (1940), 52737.

224

A ncient Am erica

3 H enri Focillon, The Life of Forms in A rt (New Haven,

1942), p. 8.
4 Diego Muoz Camargo, Historia de Tlaxcala (Mxico,
1892).
5 G eorge C. Vaillant, The Aztecs of Mxico (N ew York,
1942), p. 220.
6 Jos de Acosta, Historia natural y moral dlas Indias (Mx
ico, 1940), pp. 5 6 9 -7 0 .
7 H erm ann Beyer, El llamado calendario Azteca (Mxico,
1921 ).

8 See Anales del Museo Nacional de Arqueologa y Etnologa,

poca 1, III (1886), for reproductions.
9 Hans Dietschy, Mensch und G o tt bei den mexikanischen Indianern, Anthropos, X X X V -X X X V I (1 9 4 0 -4 1 ),
3 2 6 -4 0 .
10 Vaillant, The Aztecs, pp. 9091.
11 Ignacio Alcocer, Apuntes sobre la antigua Mxico-Tenochtitln (Tacubaya, 1935).
12 Fr. Bernardino de Sahagn, Historia general de las cosas de
Nueva Espaa, 5 vols. (Mxico, 1938).

13 Marshall H. Saville, T he Aztecan G od X ipe Totee, Iri

dian Notes and Monographs, V, no. 2 (1929), 151-74.
14 A fter the Aztec recorded by B. de Sahagn, Historiad* ta
cosas de Nueva Espaa, ed. F. del Paso y Troncoso (Madrid,
190507), f. 281. Translation in Angel MaraGaribay, La
poesa lrica Azteca (Mxico, 1937), p. 17.
15 Garibay, La poesa lrica Azteca, p. 19, after Sahagn, His
toria, f. 274.
16 Eduard Seler, Einige Kapitel aus dem Geschichtswerk des
Fray Bernardino de Sahagn (Stuttgart, 1927), pp. 91951.

[On current views of the meaning o f Aztec human sacrifices,

see my preface (A rt of Aztec Mxico. Treasures of Tenochtitln,
H . B. N icholson et al. {Washington: N ational Gallery of Art,
1983], pp. 1415) and the conference volume now in pressat
D um barton Oaks, as The Aztec Templo Mayor, as well as the
recent volum e by E sther Pasztory, Aztec A rt (New York,
1983). g k J

III.2
Toward Absolute Time:
Guano Archaeology

terpretation o f the strata has been made by Pro

fessor H utchinson: he concludes that the entire
deposition, 47.4 m. thick on Central Chincha, cannot have taken m ore than three or four thousand
years, or can it have taken less than six centuries.
The short period is out o f the question for oceanic
and biological reasons, for if deposition began in
the thirteenth century, other changes would be
known. W ithout considering the archaeological
findings, and only on the basis of the fine stratifications evident in the photographs, Professor
Hutchinson attributes the beginning of the large
deposits, on the main Peruvian guano islands, to
the first or second millennium b . c . Between these
theoretical limits, the archaeological record sug
gests a date for the initiation of the deposits that is
roughly intermediare. The archaeological record,
however, does not depend upon biological or climatological evidence. It has been worked out independently of other findings, and its fit with
Professor H utchinsons argum ent is not dependent upon any part of that argument, but only
upon the depth relations among well-documented
objects.
Such objects are few. To be useful, they must be
accompanied by some record o f the depth at which
they were discovered on a specific island; and the
object itself, or a serviceable illustration, must be
available for study (Appendix A, Class A). All ma
terials found in unstratified environm ents (Ap
pendix A, Class B), or at unrecorded, vague, or
improbable depths (Appendix A, Class C), have
been discarded from the present treatment. Since
it is most desirable, however, that owners of ob
jects from the guano islands seek to recover more

The Peruvian records indicare that the guano is

lands (fig. 111-16) yielded, between 1826 and
1875, several thousand artifacts o f Preconquest
origin. These are probably only a fraction of the
archaeological material actually discovered during
the commercial destruction o f guano caps. Most of
it has been lost; several hundred specimens in the
museums of Europe and America still retain some
record o f their origin; and a small group of pieces,
with which this paper is principally concerned, are
known to have been discovered under stratified
and depth-recorded conditions.
Such evidence is not ideal for study today, seventy years after the com plete destruction o f the
immense and ancient guano deposits of the Chin
cha, Macabi, and Guanape island groups. Y et the
accounts o f a few independent observers and od
photographs o f the guano stratification (fig.
III-l7) record inform ation that, when assembled,
suggests the rough outlines o f an absolute timescale for Peruvian prehistory. T he biological histo
ry of the Peruvian guano islands, exhaustively
studied by Professor G. E. H utchinson in a monograph1 from which this paper is an offshoot, supports the attem pt to establish an absolute timescale for the artifacts.2
Photographs taken in 1860 on N orth and C en
tral Chincha Islands, and now owned by the A m er
ican Museum o f N atural H istory, show the stratifi
cation of the guano stacks remaining at that tim e.3
The photographs reveal both wide and fine stratifications of such remarkable regularity (fig. III-l 7)
as to dispose o f the contention that the lower
layers of the guano were subject to great mechanical compressions and deform ations.4 The in
225

A ncient Am erica

226

specific information concerning their holdings,

Appendix A lists not only the materials from unstratified environm ents (Class B) but also the discoveries at unrecorded, vague, or improbable
depths (Class C); m ore exact inform ation is desirable upon these specimens as well.
For the time being, reliably docum ented guano
artifacts o f Class A are the following. These short
descriptions are supplem ented in Appendix A.

Chincha Islands
1. An armorial slab o f stone, found in 1847 by
William Bollaert beneath 18 feet o f guano (fig.
III-18 and item A 1.1, Appendix A). Present
whereabouts unknown.
2. W ooden staff ending in a rough anthropom orphic head. Found before 1861, 1520 feet be
low the surface o f the guano (fig. III-19 and
item A 1.2, Appendix A). Peabody Museum,
Salem, Mass.
3. W ooden staff, ending in helm eted, nude
female figure, seated, holding cup, and wearing
large earplugs. Found before 1873, beneath 33
feet o f guano, (fig. 111-20 and item A 1.3, A p
pendix A). Present whereabouts unknown.
4. Set of ten silver fishes, found before 1867, be
neath 32 or 34 feet o f guano (fig. 111-21 and
tems A 1.413, Appendix A). American M u
seum o f N atural H istory, No. 1010.
5. Rude wooden figure o f geom etrical style,
found before 1873, beneath 32 or 35 feet of
guano (fig. 111-22 and item A 1.14, Appendix
A). Present whereabouts unknown.
6. Three clay objects, one w ooden staff, and a
stone anthropom orphic slab, found before
1873, beneath 62. feet o f guano (fig. 111-23 and
items A 1.15-19, Appendix A). Present where
abouts unknown.

Guaape Islands
1. A wooden staff, ending in a knob surrounded
by four human faces, found before 1870, be
neath 27 feet of guano (fig. 111-24 and item A
2.1, Appendix A). British M useum, Christy
Collection N o. 7008.

2. A penguins body, resting upon a piece of cloth,

found between 1867 and 1873, beneath 32 feet
of guano (fig. 111-25 and item A 2.2, Appendix
A). Present whereabouts unknown.

Macabi Islands
1. Three w ooden figures, found on N orth Maca
bi, January 18, 1871, and certified by thegovernor of the island as having been found under
60 English feet o f guano (fig. 111-26 and tems
A 3 .3 -5 , Appendix A). Formerly in Hamburg,
Johanneum N aturhistorisches Museum, pre
sent location unknown.
Clearly only the Chincha Islands have yieldea
depth-recorded artifacts in numbers that permir
discussion. But it will be seen that the objects and
the depths recorded from the other, more northerly island groups, although few in number, are
not inconsistent with the argum ent that can be
developed from Chincha Islands material.
Certain artifacts from the Chincha Islands art
closely related to industries and styles already
known from the archaeology of the Peruvian
mainland. Some com m ent on these connections
and absences o f connection will be needed to pre
pare for the discussion o f depth-relations among
the objects.
1.
The Bollaert armorial slab (fig. 111-18) is of
colonial manufacture. The cartouche working of
the outline of the shield is o f N o rth Italian origin,
in the early part o f the sixteenth century.5 The
shape itself was widely diffused throughout Eu
rope and America before 1575. The inscription
yields no further clues, for it is barely legible. It is
carved in an approximation to Gothic letter that is
of little use in dating, for this letter was current in
Spain at least until the decade o f the 1560s.6 The
quarterings, like those o f many Spanish-American
grants o f arms, contain topographic allusions. The
upper right quadrant shows a long-billed bird,
probably a pelican. The lower right quadrant displays waves and three islands, o f which the central
one bears a horseshoe-shaped nest, characteristic
of the guanay nests o f the Peruvian guano islands
(cf. fig. 111-30). These bearings can refer only to

Tow ard A bsolute Time: Guano A rchaeology

227

the Chincha Islands, which are three in num ber

mand, such as the one carried by the Inca master of
and have long been populated by large colonies of
the roads, as portrayed by Felipe Guamn Poma
pelicans.
de Ayala, ca 1613.12 The head carved on the hanSince the arms refer to the islands, they proba
dle is rude and direct, worked in a manner that
bly mark the ownership under which the early
seeks to suggest natural form by rounded pas
colonial exploitation o f guano was conducted. Besages. I can find nothing in Preconquest Peruvian
cause inhabited islands were always crown properwood carving that resembles this long-headed,
ty, the arms cannot have been granted to an indi
modeled versin of human shape. Colonial, rather
vidual.7 They can have signified only crown ow
than Preconquest, work is likely.
nership, under a specific corregimiento, or govern3. The helm eted, nude wooden figure, rement, with its administrative center on the
produced by T. J. H utchinson, is known only by
mainland.
this wood engraving (fig. 111-20). N o dimensions
This administrative center was surely the port
are given, but the piece is evidently the head of a
on Paracas Bay, variously known as Sangallan and,
staff and therefore probably between 5 and 10
later, Pisco (fig. 111-27). T he history o f this settleinches high. The sex, clearly female, is at variance
ment is relevant. Cieza de Len, writing before
with the masculine gear of helm et and large ear1550, spoke o f Sangallan as a port,8 and described
plugs. T he carving o f the body forms is abrupt and
the nearby Chincha Islands, without being aware
schematic, although not w ithout anatomical acof their guano deposits. In 1572, Viceroy Toledo
curacy in the rounded flesh surfaces. The sharply
recommended to the Council o f the Indies that a
indented outlines o f the figure and the squat body
town be founded in Pisco Valley; by 1576, shipproportions recall Inca sculpture distantly, such as
ping was received at Puerto de Pisco; and in 1587
the sodalite figurines13 from Pikillajta in Cuzco
the harbor settlem ent was visited by English piValley, or the large diorite head found 8 meters
rates.y Early in the seventeenth century, the town
beneath the present level of the Jesuit church in
possessed churches (Franciscan Recollects, found
Cuzco.14 A nother suggestion o f late (that is, fifed 1602; San Juan de Dios, founded 1634); in
teenth century) date is the elongated, concave1640, its ame was changed to San Clem ente de
flaring form of the cup the figure holds between
Mancera; and in 1687, the city was wrecked by
her hands. Such cups in metal are familiar from
earthquakes and tidal waves. Pisco was moved two
north coast sites in Inca and immediately pre-Inca
miles inland to its present site in 1688.10
associations.15 O n the whole, the style of the fig
The style o f the arms and of their carving sug
ure cannot be assimilated either to the Coastal
gests the period between 1575 and 1600. If the
styles of consummate naturalism, such as Mochica
association with Pisco harbor is valid, they are
w ood16 and pottery, or the geom etrizing styles of
probably the arms (otherwise unknown) of the set
Tiahuanacoid and Middle period associations. It
tlement founded by order of Viceroy Toledo, ca.
seems to relate most closely to the wood carvings
1572, and destroyed in 1687.11 For working purof a late, pre-Inca period, such as those from
poses, a date 15751600 is in agreem ent with the
Mrquez, at the m outh o f the Chilln Valley
style of the arms and their carving, as well as with
(Dept. Lima). Among the staffs figured by Baesthe history of settlem ent on Paracas Bay. Within
sler, one (fig. 187) shows the legs carved and
the span 15751600, the earlier years agree best
crossed in a m anner closely similar to that of our
with the history o f settlem ent at Pisco and with the
figure 111-20.1 Thus a date in the fourteenth or
energetic governm ent o f Viceroy Toledo.
fifteenth century seems reasonable.
2.
The Salem staff, 27 inches long, bears a
4. The ten silver fishes collected by Squier are
available for study at the American Museum of
rudely carved head at the thick end (fig. III-19).
Natural H istory (fig. 111-21). It is possible that
The stick is too light for a club (IV 2 inches mx
these objects, each with a perforation near the
imum diameter) and too short for a cae or digging
mouth, were originally sewn to a cloth as ornastick. The form is rather that o f a baton of com-

228

A ncient America

ments. Several show m ore elabrate tooling.

These differences might suggest separate groups
were it not that the objects are reliably attested as
having been found together. In common, howev
er, their style may be related to that o f a num ber
of metal objects o f late pre-Inca period, reported
from the coast and catalogued by Baessler.18
5. The w ooden torso with head, figured by
Hutchinson, shows a rude, geom etric style (fig.
111-22) that may be compared to the abstract rep
resentations o f the Tiahuanacoid period on the
coast. The size is unknown. The face is treated as
half an ellipse. It contains lozenge-shaped eyes and
a box m outh with five teeth. The fat plae o f the
face is surm ounted by a squarish hat with peaks at
the corners. The hat is o f a form known first in the
Tiahuanacoid period, as in the headdresses figured
by Baessler and B en n ett.19 The lozenge eyes and
box m outh with five teeth compare closely with a
figure of unknown provenience illustrated by
Seler.20
6. Five objects are shown by H utchinson in the
find reported at 62 feet beneath the guano on the
Chincha Islands (fig. 111-23).
Specimen A 1.15 is evidently a pottery vessel,
representing a seated human with a rope around
his neck. The form is common in Mochica pottery
of the north coast. The wood engraving suggests
the white paint on a red ground that is characteristic of many Mochica wares. The human figure
represents a prisoner, as may be inferred from
other specimens in which the hands are roped to
gether behind the figures back.21
Specimen A 1.16 represents a dog vessel. Unfortunately, its color cannot be determ ined. Its
most striking singularity is the combination of a
flaring neck vent with carrying loop. This assem
bly is lacking in standard Mochica wares, where
the flaring neck never appears with a carrying
loop, to my knowledge. Such an assembly does
appear repeatedly in pre-Mochica, north coast
wares, such as Salinar;22 it disappears from stan
dard Mochica wares o f all classes; and it reappears
in Late Chimu vessels.23 T he modeled form of the
dog, on the other hand, strongly recalls Mochica
rather than Chimu techniques o f detailed and descriptive sculpture.24 H enee, A 1.16 offers a
choice: if it can be proved that its period is Late

Chimu, then a large part o f the argument of this

essay collapses, for the argum ent is based upon an
assumption that the five pieces in figure II1-2 3 are
all contem poraneous within two or three generations. But the modeling is far closer to Mochica
than to Chimu, and the neck-and-loop assembly is
known in pre-M ochica Salinar style.25 These affinities will allow a Mochica or pre-M ochica Iden
tification for item A 1.16 until better evidence for
the Chimu dating is discovered.
Specimen A 1.17 is evidently a clay fragment of
unknown use and style. Possibly an egg or a fruit
shape is intended. The light color and precise
shape again suggest Mochica origin.
Specimen A 1.18 is a w ooden staff, slotted at the
handle and bearing a crouching animal or human
figure. It is almost the exact analogue of one fig
ured by Baessler as coming from Chancay.26 In the
absence of any systematic knowledge about Peru
vian wood carving, it is impossible to assign this
piece or its analogues to a well-defined period. No
compelling arguments forc an identification ei
ther as Mochica or post-Mochica. The Chancay
provenience of the piece figured by Baessler can
not be taken as a telling argum ent against Mochica
style, for Mochica specimens have long been
known from the south coast o f Per,27 although
never properly evaluated as evidence for Mochica
distribution.
Specimen A 1.19 is described as a stone idol.
It is clearly a thin pate, shaped as an ithyphallic.
possibly herm aphrodite, human. The head is
marred by flaking, so that little can be observed of
its style, although the half-ellipse o f the face is
clearly and strongly defined. N othing like it in
mainland archaeology is known to me. The unworked, downward tapering base suggests use ai
an upright slab, inserted into the ground.
The depth-recorded objects from other island
groups than the Chinchas are not numerous
enough to warrant detailed discussion. The one
group o f finds, however, from the Macabi Islands,28 under 60 English feet of guano, is closeiy
enough related to mainland archaeology to merit
treatm ent (fig. 111-26). All specimens, of which
illustrations were made, are of wood.
N um bers 1 and 3 are very closely related to the
wooden figures found by H utchinson in Caete

Toward A bsolute Tim e: Guano A rchaeology

Valley (fig. III-44).29 N um ber 1 is a standing,
nude male prisoner, as in the Caete Valley specimen. N um ber 3 is a seated figure, badly worn and
almost identical with its Caete Valley analogue.
N um ber 2, a seated, nude, wooden prisoner, is
nearly a replica o f the Chincha Island Mochica
pottery vessel representing such a prisoner (fig.
111-23). N ow all three Macabi specimens from
Hamburg closely resemble the wood carvings in
the British M useum (figs. 111-28,111-29) cataloged
as coming from the Macabi Islands.30 The advantage of this comparison comes from the fact that
the British M useum woodcarvings were found
with a pottery fragment of unmistakable Mochica
manufacture (no. 8 on fig. III-28).31 Apart from
the well-modeled realistic style o f the wood carv
ings, the Mochica pottery association confirms
their identification as Mochica work.
The depth-relations among the Chincha Island
finds may now be treated. The m ethod will be
open to innum erable questions, so that an effort
must be made to set forth all the assumptions upon
which the m ethod is based. The m ethod itself is
rudimentary in its simplicity. The depth of a find is
taken as a direct indication of the time at which it
was deposited on the islands. For the Chincha Islands, it is assumed, from the evidence of photographs and early surveys o f the guano caps,32
that the guano was laid down in horizontal layers,
at a regular rate on all parts o f the islands; in other
words, that the islands were completely and steadily populated by ubiquitous bird colonies,33 and
that their production o f guano, if subject to varia
tions, varied cyclically in regular short- and longterm phases. This assumption is supported by the
flat, table-top character o f the rock surface of the
three main islands, N orth, Central, and South
Chincha. Second, it is assumed that over secular
periods the ratio between deposit of the guano and
surface erosion by wind remained constant. Third,
the human exploitation of guano, from Mochica
times to ca. 1840, was not only regular but negligible, never approaching a large-scale ecological disturbance for the birds. (See note 43.) Fourth, deposition ceased on the Chincha Islands during the
1840s, with the beginning o f industrial exploita
tion.34 Fifth, it is assumed that mechanical com-

229

pression of the guano at deep levels in the cap was

insignificant, as suggested by od photographs (fig.
III-17).35 Sixth, it is assumed that the deposit of
artifacts occurred at the surface o f the guano cap
rather than in pits or shafts. This assumption is not
susceptible o f proof. Seventh, we assume that
depth-recorded finds, regardless o f horizontal
positions on the islands, are equally and uniformly
significant o f tim e-relations.36 Finally, in generalizing upon the depth-relations among finds, we
assume that the date o f deposition is in all cases
roughly contem poraneous (within a generation)
with the date of manufacture. Proof here is also
lacking.
The mximum depth of the guano cap may be
taken as 44.7 m., by averaging the deepest soundings taken on the three islands o f the Chincha
group during 1853.37 Since it is certain that depo
sition ceased altogether about 1850, this year may
be taken as a terminal date. Among the depthrecorded artifacts, the critical specimen is the
Bollaert slab, for an absolute date is easily ascertained by stylistic and historie evidence. This date
has been fixed between 1575 and 1600, with some
likelihood that the years around 1575 correspond
with the manufacture and installation of the arms
on the island. Therefore 1575 will be taken as the
year corresponding to the 18-foot, or 5.5 m.,
depth at which the slab was found. H enee, 275
years elapsed during the accumulation o f 5.5 m. of
guano. Dividing depth by time, we find that the
annual thickness o f deposit was on the order of 20
mm., in the compacted, eroded, and layered condition imposed by the presence o f the bird colonies.
At this rate the accumulation of guano in the
deep stacks would approach 2 meters per century.
The initiation of the deposits then is seen to fall
roughly in the fourth century b . c . (see accompanying chart).
The remaining depth-recorded artifacts from
the Chincha Islands may now be given their secu
lar positions by extrapolation. The depth of 10.4
m. for the silver fishes places them, and the ar
tifacts found at similar depths, in the fourteenth
century a . d . It may be pointed out that nothing in
their style is inconsistent with such a dating in late
pre-Inca tim e.38 The depth o f 19.8 m. for the

230

A ncient Am erica

Profile of
cenfer stack
Chincha Islands

ro

c
O
o
O
iroo

i
ro
o
o
o
?l

1800-1850
18th c.
5m.--------- 17th c.
16thc
15th c.
10m .--------- 14th c.
13th c.
12th c.
11th c.
10th c.
20m.-------- 9th c. |
8th c.
7th c.
6th c.
5th c.
4th c.
3rd c.
2d c.
1st c.
1stc. B.C.
40m.
2d c. B.C.
3rd c. B.C.
4th c. B.C.
44.7m.
1m .----------

5.5m. Bollaert slab

6.1 m. Salem staff
10.05m. Helmeted female
. 10.4m. Silverfishes
10.7m. Geometrical figure

19.8m. Mochica vessel

our slab. If 1535 is taken, then 5.5 m. accumulated

in 315 years, at a rate o f 17 mm. annually, or 1.70
m. per century. At this rate, the Mochica-associ
ated artifacts at 19.8 m. would fall in the late seventh century, which is not unreasonable, by existing relative estimates. But the silver fishes (fig.
111-21) at 10.4 m. and the nearby objects would
have been left on the islands in the thirteenth cen
tury, which is perhaps too early, especially for the
helmeted female staff head (fig. 111-20), with its
strong, late pre-Inca or early Coastal Inca associa
tions.
If the Bollaert slab was deposited ca. 1640 (an
im portant date in the history of Pisco town) the
guano accumulated at the rate of 2.6 m. per cen
tury, with an initial date in the second century a . d .
H enee any date later than 1640 for the Bollaert
slab places the Mochica-associated find (19-8 m.)
at an impossibly late time, after the twelfth century
A .D .

A.D.

Rock

Mochica-associated artifacts reported by H utchin

son brings us to the ninth century a . d ., a dating in
full agreem ent with the most recent estmate of
relative north coast chronology.39
That the same rate o f secular accumulation
holds for certain other island groups is suggested
by the well-documented find o f Mochica artifacts
under 60 English feet o f guano on the Macabi Islands. (See p. 235.)
It should be pointed out that datings other than
1575 are of course possible for the Bollaert slab
(fig. III-18). But the other datings are confined
within fairly narrow limits. The lower limit is the
Spanish Conquest, ca. 1535, when early European
settlem ent on Paracas Bay might have produced

The argum ent on which this essay is based

that depth directly indicates the age o f guano ar
tifacts under certain conditions should not depend at any point upon the biological history of
the guano islands, if it is to bring confirmation to
the results of such a biological history. Another
line of evidence, fortunately, permits an independent definition o f both lower and upper limits.
As to a lower limit at 1535 or thereabouts, the
possibility is dispelled by early colonial writings
that reveal Spanish unawareness o f the guano resources of the Chincha Islands. Cieza de Len,
writing before 1550, spoke o f these islands and
incorrectly described the guano caps as sand hills.
H e wrote that the group near Sangallan (in Paracas
Bay, the Chinchas and nearby rocks), a little ovei
four leagues from the coast, were seven or eight
other small islets, some high and others low, uninhabited, and w ithout wood or water, tree, shrub.
or anything else, except seis and sand hills [are
nales']. . . . T he Indians, according to their own
account, used to go to these islands to make sacri
fices, and it is presum ed that great treasure is bur
ied on them. . . . This fascinating account is our
earliest European m ention of Indian visits to the
islands, and it proves that before 1550 the Indians
had abandoned the habit, and that guano from the

Toward A bsolute Time: Guano A rchaeology

Chinchas was still unknown to Europeans. But
Cieza knew of guano and its use as fertilizer, for in
another passage o f the same work he spoke of the
Indians use of guano from the islands off the coast
of Tarapaca, several hundred miles farther south
than the Chinchas.40 Later writers repeated Ciezas remarks w ithout further verification. Lpez
de Velasco, the em inent cosmographer, writing
before 1574, used the same words, like Reginaldo
de Lizrraga, who, ca. 160510, followed Cieza
closely.41 These writers were aware of guano and
described its use at Arica and on the Tarapaca
coast in general.
Two sources, however, indicate that guano from
the islands off the central coast carne into commercial use between 1560 and 1580. Pedro de Avendao, writing in 156465 o f the central coast of
Per, spoke o f many vessels being loaded at the
islands and mentions Indian disputes over ownership.42 Jos de Acosta, writing ca. 158090, de
scribed the guano islands of the central coast briefly and specified that the material was used for
fertilizer in Lunaguana Valley, just south of the
Rio Caete. The source may have been either the
Asia island group or the Chinchas, for Lunaguana
Valley is roughly equidistant from the two.43
The upper limit for the Bollaert slab may be
fixed by reference to the economic history o f the
port on Paracas Bay, namely, Pisco H arbor. As
pointed out earlier, the formal foundation o f such
a harbor town occurred ca. 1572. The rapid developm ent o f the great mercury mines at Huancavelica, however, soon necessitated the founda
tion o f another seaport nearer the natural outlet
from Huancavelica. This was found in Chincha
Valley, well to the north of the harbor at Pisco. By
1576, Pisco was forbidden to ship mercury, and
with this interdict Pisco declined as a harbor. In
1587, the unhappy town was sacked by English
pirates; all mercury was passing through the port
in Chincha Valley, and Pisco itself was subordinated by viceregal order to the corregimiento located at Valverde de lea, some 70 kilometers from
Pisco.44 N o t until 1640 did Pisco H arbor revive
briefly, when Viceroy Mancera attem pted to reestablish its activity.45 But new disasters soon befell Pisco. In th e I640s, clouds o f guano dustblow -

231

ing over the regin made the site uninhabitable,

and in 1687 an earthquake and tidal wave led the
settlers to abandon the seashore and build anew
two miles inland.46
In brief, Pisco flourished best under Viceroy
Toledo in the 157Os, before the creation of the
port in Chincha Valley and before the incursions
of pirates. The economic history of Pisco and the
beginnings of colonial guano exploitation allow
the Bollaert slab to be assigned to the decade of
the 1570s. The date is further confrmed by the
style o f the carving. (See p. 227.)
Various other aspects o f the guano island mate
rial m erit discussion, aspects that do not relate to
absolute chronology. T he artifacts suggest a spe
cial guano island apparatus and iconography. Ci
eza, Garcilaso, and Frezier mention ceremonies
and sacrifices.47 The num ber of wooden staffs sug
gests an undefined activity specially significant on
the islands (items A 1.2, A 1.3, A 2.1, B 2.02, C
3.1, C 4.11, etc). G reat layers o f cloth deposits, as
reported by various writers, probably answer to
ceremonial requirem ents (items C 3.7, C 3-01).
The presence o f many nude prisoner figures in
wood and pottery (items A 1.15, C 3.10, etc.),
suggests a penalty for abuse or misappropriation
of the islands and their resources, especially in the
Mochica period. Headless female mummies (item
C 3.01) and traces o f domestic architecture (items
B 2.1, C 3.16) point to sustained habitadon. Finally, a Mochica vessel (fig. 111-30) describes the
visitof balsa-rafts to the islands. A human being, in
hierade posture, suggests ceremonial behavior.48
T hroughout this repertory, however, certain
styles o f mainland industries are conspicuously
lacking. The wares o f Nazca Valley and the textiles
of Paracas pennsula are not known from the
Chincha Islands. On no island group has any spec
imen of Tiahuanacoid style been recorded. Also
lacking from the record are wares such as lea Geometric and Chancay. That the ceramics o f Nazca
and lea valleys should be lacking is astonishing, in
the presence o f Mochica artifacts that are usually
supposed not to extend so far south. In any case, a
revisin of our concepts of Mochica extensin is
clearly indicated (note 27). Such a revisin at once
suggests the hypothesis that the success of the

232

A ncient Am erica

Mochica cultural pattern may have depended in

part upon a technique o f guano fertilization of arid
coastal valleys. Real proof, however, is at present
lacking. Coastal Tiahuanaco peoples may have de
pended more upon highland connections for subsistence than upon intensive coastal agriculture.
Nazca and lea valley peoples never achieved widespread cultural domination over their rem te
neighbors. The Mochica pattern was the only
coastal style, to present knowledge, that achieved
distribution, however, sporadic, throughout the
entire system o f north, central, and south coast
valleys.
The significance of this approach to absolute
time, in relation to the early cultures discovered
by J. B. Bird in northern Chile49 (Shell and Fish
H ook culture on fossilguano at Punta Pichalo) and
in relation to climatological history in the South
Pacific, has been treated elsewhere by Professor
H utchinson.50 H ere it is necessary only to comm ent upon the contracted time-scale, in terms of
the Mochica finds at 6062 feet. If these are ob
jects of Late Mochica style, then the tenth to thirteenth centuries are open for post-M ochica and
pre-Inca cultures. If the objects are not terminal in
Mochica style, but incipient or Middle, the timespan available for M iddle-period styles is substantially reduced, perhaps to a period covering no
more than the twelfth and thirteenth centuries.
The answers to this and other questions were
probably destroyed unwittingly during the commercial destruction o f the guano deposits.51
The truth remains that several assumptions on
which this work is based are arbitrarily taken. But
it is also true that the evidence o f the depth-re
corded artifacts on the Chincha Islands, without
depending upon these assumptions, shows an
order in depth that corresponds to order in time.
At any m om ent, examples o f depth-recorded ar
tifacts, now unknown to me, may come to light. If
such artifacts depart from the observed correspondence between depth and date, then the hypothesis will have to be modified or abandoned. But at
present no reliably recorded object violates the
time-depth correspondence. For the present,
therefore, the dates contained in the chart on p.
230 can be considered reasonable, with Mochica

falling between the seventh and eleventh cen

turies.

APPENDIX A
In order to catalog the objects and yet to leave room in
the enumeration for future expansin, a decimal classifkation is used. T he prefixes A, B, C, designare the
character o f the find as depth-recorded (A); under unstratified conditions (B); at vague, improbable, or unrecorded depths (C). T he numeris before the decimal
point designare the island group: Chinchas (1.); Guaape (2.); Macabi (3-); Lobos (4.). Following the deci
mal point, a numeral other than zero indicates that a
drawing, a photograph, or the object itself is available
for study. If a zero follows the decimal point, no visual
evidence is known. In groups o f objects constituting a
single find, each object is cataloged by a separate
number where visual evidence is at hand.
A l . l . The Bollaert slab (fig I I I -l8), discovered on
North Chincha Island under 18 feet o f guano in 1847,
has been lost to view since 1860. W hen the drawing was
published in 1859, the slab was about to be presented
to the British M useum , but no record o f such agift has
been discovered.
The drawing was shown in 1946 to Albert Van de
Put, F.S.A., Late Keeper, Library, Victoria and Albert
M useum, London. Mr. Van de Put could not assign a
coherent m eaning to the inscriptions. The style o f the
armorial carving, however, led him to write (in litt.,
Feb. 18, 1946): With regard to the date o f the slab, the
shield shape is of, no doubt, Italian (Tuscan?) derivation, first quarter o f the X V I century. . . . Mr. Van de
Put points out that the carver has not gone very far
with the cartouche working o f the outline, which tends
to the baroque but is rather severely plain in comparison with what was being turned out by the third
quarter o f the X V I century in Europe . . . , and he
concludes that it seem s probable that a date about
15751600 is indicated.
Inquines addressed to Father Rubn Vargas Ugarte
and to Dr. Jorge Zevallos in Per during 1944 tailed to
yield any information concerning the grant o f munici
pal arms to the port o f Pisco. Similar inquines in Spain
remained fruitless. It is certain, however, that Pisco wa>
granted arms, and it is Iikely that these arms, when
located, will correspond with those o f the Bollaert slab.
Quarterings with topographic allusions are not uncommon among Latin American municipal arms (e.g., Ptzcuaro, M xico).

Toward A bsolute Tim e: Guano A rchaeology

Bibliography: Bollaert 1860, p. 149; London Ilus
tra ted Times, March 5, 1859, p- 157; M ontoto 1928;
Garca Carraffa 190236.
A 1.2. L. W. Jenkins, D irector o f the Peabody Mu
seum, Salem, Mass., describes this object (in litt., July
8, 1944) as follows, . . . agod-stick orclub said to have
been found 1520 feet below the surface [on the
Chincha Islands] and given to the Essex Institute by
Capt. J. B. King in 1861. It is about 1 V2 in. in diam. and
27 in. long. A head is crudely carved on one end, and
the stick tapers to a point at the other end (fig. 111-19).
A 1 3 . W hen discussing the Chincha Islands, T. J.
Hutchinson reproduced an engraving o f a wooden staff
head, carved in the form o f a nude, seated, and hel
meted female (fig. II1-20). The legend accompanying
the engraving states that it is a w ooden idol found at a
depth o f 33 feet under guano. The text offers no fur
ther details.
Bibliography: T. J. H utchinson 1873, vol. I, p. 108.
A 1.413. Ten silver fishes w ere given to E. G.
Squier by H enry Swayne, a landowner in Caete Valley, who obtained them from Juan Pardo, the Italian
master o f a coasting vessel before 1867. Pardo declared
they were extracted from beneath 32 feet o f guano on
the Chincha Islands. At the same time there was found
the body o f a female, lacking the head, which, however, was discovered at som e distance from the skeleton.
The chest, breasts, and ribs were covered with thin
sheets o f gold . This quotation is from Swaynes letter
to Squier. In a penciled catalog o fh is collection, Squier
recorded the depth as 34 feet. It is not explicitly stated
that the fishes and the skeleton w ere parts o f one single
find. The fishes were acquired by the American Mu
seum o f Natural H istory in 1875 and cataloged as no.
1010 (fig. 111-21).
Bibliography: Squier 187172, pp. 5152; Am eri
can M useum o f Natural H istory, MS catalog 1, vol. I
(All Parts o f the W orld), p. 33; T. J. H utchinson 1873,
vol. I, p. 129A 1.14. T. J. H utchinson, discussing the Chincha
Islands, reproduced an engraving o f this object (fig.
111-22) with the legend, w ooden idol found at a depth
of 35 feet under the guano. The accompanying text
mentions a w ooden idol discovered at a depth o f thirty-two feet.
Bibliography: T. J. H utchinson 1873, vol. I, p. 105.
A 1 .1 5 -1 9 - T hese five objects (fig. II1-23) from be
neath 62 feet o f guano on the Chincha Islands are
known only through the w ood engraving reproduced
by T. J. Hutchinson.
Hutchinson also reproduced another engraving,

233

showing eleven objects o f M ochica style, in his discussion o f the Chinchas. H e described them as relies o f
household gods and regal em blem s, taken from a depth
not known to me, but very, very deep. . .
The resemblances to the Mochica pieces from Macab in the
British Museum are striking and were commented
upon by H utchinson in a footnote.
H utchinson further stated that on the Chincha Islands, w e have had excavated from a depth o f sixty feet
in the guano regala o f the od Kings, stone and wooden
idols, as well as, now and then, som e pieces o f gold .
The present whereabouts o f the Hutchinson collections is unknown. In 187475 these collections, con
sisting o f at least five cases, were exhibited in London at
the Bethnal Green Museum. Correspondence with the
Victoria and Albert M useum, which ultimately ab
sorbed the Bethnal G reen M useum, has yielded no
trace o f the collections.
Bibliography: T .J. H utchinson 1873, vol. I,pp. 104,
1 0 6 -0 7 ; 1874, pp. 3 1 3 -1 4 ; 1875, pp. 13, 325.
A 2.1. This portion o f a staff or baton (fig. 111-24) was
found 27 feet beneath guano on South Guaape Island
before 1870. Exhibited at the Ethnological Society,
London, in 1870, by Josiah Harris, the specimen is now
in the British Museum. There it has been identified (in
litt.) by Adrin Digby o f the British Museum, Depart
ment o f Ethnography, as no. 7008 in the Christy
Collection.
The knob head is surrounded at the neck by four
small human faces in relief carving. The hair on each,
centrally parted and beginning at the eyebrows, recalls
Inca metal figurines from the Valley o f Cuzco (cf.
Schmidt 1929, p. 390).
T he Harris staff is almost exactly identcal with an
other one in the British M useum, registered there as
no. 7431, and in our catalog as tem C 3-13, from the
Macab Islands.
Bibliography: Nature, 187071, vol. III, p. 59;
Squier 187172, p. 48; Gonzlez de la Rosa 1908, p.
42; The Journal o f the Anthropological Institute o f Great
B ritain a n d Ireland, London, vol. I, p. 39, fig. 2.
A 2.2. Betw een 1869 and 1873, not long after the
beginnings o f commercial exploitation o f the Guaape
Islands, numerous artifacts were encountered. T. J.
Hutchinson wrote that at a depth o f thirty-two feet
under the guano . . . has been found the body o f a
fiattened Penguin, with a piece o f cloth underneath.
. . . Several idols have been discovered here likewise,
and the Chinese workmen have turned up gold ornaments, which, o f course, were at once appropriated and
partitoned, according to their ordinary usages in cases

234

A ncient America

o f treasure trove. O f the penguin, Hutchinson noted

that in its flattened position it is only half an inch in
thickness. . . . T he iliustration (fig. 111-25) accompanying this passage is not clear enough to allow any
inferences concerning the style or technique o f the
cloth.
Bibliography: T .J. H utchinson 1873, vol. II, p. 129.
A 3 - 3 5. Onjanuary 18, 1 8 7 1, eight or nine wooden
figures were encountered on N orth Macabi Island at a
depth o f 60 English feet. On February 28 the governor
o f the Guanape and Macabi groups signed an affidavit
attesting the conditions o f the find. This certifcate was
accompanied by a photograph (now lost) o f the objects.
A Lima merchant, W ilhelm Scheel, acquired the pieces
and donated them to the Johanneum Naturhistorisches
Museum in Hamburg. Only three specim ens arrived at
their destination, where they were exhibited in 1873
74. The first, no. 1 on figure 111-26, is reported as 73
cm. high; no. 2 as 37 cm.; and no. 3 as 33 cm. T he other
specimens lost in transit were five or six in number,
meist sitzend mit untergeschlagenen Beinen, darunter
mindestens noch 3 mit dem um den Hals geschlungenen Stricke.
Bibliography: Boln 1874, p. 93; Virchow 1873, pp
1 5 3 -5 4 .
B 2.1. An important group o f artifacts was encoun
tered on South Guanape Island in 186364. The find
was made by J. P. Davis o f Boston, an engineer then
engaged by the Peruvian governm ent to survey the
guano stacks o f the north coast. Davis found a wooden
female figure ca. 1 foot high (fig. 111-31); three pottery
figures ca. 4 inches high; and two water jars, o f which
one portrayed a fish. The discovery occurred at the
edge o f a cliff about 4 5 0 feet high, among the ruins o f
what appeared to have been a stone hut, covered to a
slight depth by h u anu." E. G. Squier noted that the
large w ooden figure (fig. 111-3 1) was so com pletelv
saturated with the salts o f h ua n u that it has very nearlv
the specific gravity o f marble.
Bibliography: Squier 187172, p. 53.
B 2.01. In 1871 the report reached London o f the
discovery o f artifacts on N orth Guanape Island. The
editor o f N ature published the report in the following
terms: beneath forty feet o f guano, a cavity was come
upon, which, on the removal o f the guano, was found tu
be a cave, leading downwards further forty feet. . . .
It . . . contained . . . well-preserved sea-fowl and other
birds, lizards eggs. . . . In many cases, the color o f the
eggs is preserved. T he cracks and fissures in the walls o f
the caves were found filled with solidified ammoniacal
salt. Two pieces o f earthenware vases were found, bear
ing figures, also two gold earrings, and a bundle o f

medicinal herbs tied up in woven cloth. The report

also mentions rude representations o f the human fig
ure, cut in very hard w ood. It seem s clear that the
entrance to the cave was sealed by 40 feet o f guano, and
that the objects therefore antedated this stratum.
Bibliography: N ature, 1871, vol. IV, p. 394, Lon
don; Squier 187172, p. 48; Gonzlez de la Rosa
1908, p. 42.
B 2.02. T he governor o f the Guaape Islands re
corded a number o f finds on South Guaape before
1873. W e ow e this record to the diligence o f M.
Gonzlez de la Rosa, who sent a questionnaire concern
ing guano archaeology to the officials supervising the
industry. The questionnaire was sent from Paris be
tween 1869 and 1872. In 1873, Jos Mara Garca replied from South Guaape that objects had been en
countered at depths o f betw een 3 and 4 meters. H e
mentioned dolos y utensilios de madera negra y chon
ta, los primeros representando un hombre en cuclillas,
con los brazos cruzados sobre el pecho y descansando
en la cabeza cuadrada en un palo redondo, de 8 cen
tmetros de dimetro y de metro y m edio de largo.
There were also bastones largos o varas, algunos tall
ados, y canaletes, que son una especie de remos cortos
con pala ancha, tallada esta con figuras de distinta clase
de pescados. Garca noted further that neither human
remains or objects o f stone had been found.
Bibliography: Gonzlez de la Rosa 1908, pp. 4 0 41.
C 1.1. G old objects from the Chincha Islands were
exhibited in N ew York at the m eeting o f the American
Ethnological Society in 1859. The pieces were reported
to have been found with thirty feet or more o f [guano]
resting upon them . A sheet gold figure in the find can
be illustrated (fig. 111-32). N o record o f the appearance
o f the other specimens has been found. T hese orher
objects were a gold, tubelike extensin to the figure
illustrated (fig. 111-32), and two gold cups o f the weight
o f five dollars each. Figure 111-32 weighed a little less
than twelve gold dollars.
The pieces w ere exhibited by Thomas Ewbank, who
prepared a drawing o f each piece for his communication to the Society. They were then the property of
Trevor & Colgate, the Wall Street bullion dealers, and
had been in the latters possession for about a year.
Bibliography: Squier 187172, p. 49.
C 1.23. In the Marine Museum at Mystic, Connecticut, are two specimens o f Inca pottery from the
Chincha Islands (fig. II1-36). They were found during
or before 1860. The first is an aryballos, about 8 V2
inches high and lightish brown in color. The second is a
blackware vessel, shaped like a six-pointed star. Descriptions, drawings, and information concerning the

Toward A bsolute Tim e: Guano A rchaeology

pieces were provided by Cari C. Cutler, Secretary o f
the Marine Historical Association in Mystic.
The pieces formerly belonged to Captain G eorge B.
W endell o f the ship Galatea. Mr. A. O. V ietor o f the
Yale Library informed me that this ship was engaged in
the CallaoN e w York trade during 185960. Captain
W endell acquired pieces from Commander Jones o f
the Peruvian navy at the Chincha Islands on O ctober 6,
1860, with the information that they carne from 200
feet under the surface o f the guano. A letter from
Commander Jones to Captain W endell gives 300
feet. (Communicated by Captain W endells son, Ar
thur R. W endell.) That these depths are improbable
em erges from the fact that the guano caps never ex
ceeded 56 varas (168 feet).
C 1.4. A golden figure o f a female (fig. 111-33), from
the Chincha Islands, is in the Museum fr Vlkerkunde, Berlin. As part o f the Gretzer C ollection (no.
31760) it entered the m useum in 1907. This collection,
formed by a merchant o f Lima during thirty-three years
in Per, contains objects acquired roughly betw een
1870 and 1900.
The piece is 15.5 cm. high and weighs 16 gms.
Bibliography: Schmidt 1929, p. 367, no. 6, and p.
593; Anonym ous n.d., p. 9 3 \Jahrbuch der Koeniglichen
preussischen Kunstsammlungen, vol. X X X III, p. 90,
1907.
C 1.523. During 1872, Louis Agassiz, aboard the
Hassler, stopped in Paracas Bay and acquired a number
o f objects from the Chincha Islands. The specimens
certainly known to have com e from the Chincha Islands
are 18 objects o f silver, now in Peabody Museum, Har
vard University under the accession number 73-6-30.
N o depths were recorded and the style o f the pieces is
generally amorphous. I ow e photographs o f these spec
imens to the courtesy o f J. O. Brew and Donald Scott.
Bibliography: Peabody M useum 1874, pp. 2 0 22;
Proceedings o f the California Academy o f Sciences, vol.
X X X III, pp. 2 5 7 - 5 8 , 1 8 6 8 -7 2 ; Agassiz 1880, pp.
3 9 4 -4 0 0 .
C 2.1. A human figure o f unspecified material and
technique was collected at the Guanape Islands in
1826. This is the earliest recorded find o f artifact char
acter on the guano islands. That such finds were com
mon before 1848 is suggested by Pickett (1848, pp.
3738): In the guano deposets [sic], and far below the
surface, ancient tools and instruments are frequently
found. . . .
The original o f the G oodw in figure was long kept at
W esleyan University, M iddletown, Connecticut, but it
can no longer be found there. Fortunately, a east was
made for the American Antiquarian Society in W orces-

235

ter. The Society in 1910 effected an exchange with

Peabody Museum, Harvard University, and the east is
now in Cambridge (accession 10-47, specimen 76832).
C 2.2. A helm eted, cup-holding w ooden figure (fig.
111-34) from the Guaape Islands is in the Museum fr
Vlkerkunde, Berlin. It stands 118 cm. high and it en
tered the museum in 1890 with the Maier Collection.
The forms may be compared with those o f item A
1.3, although the carving o f fish forms on the flesh parts
is lacking in the more naturalistic treatment o f the
latter.
Bibliography: Schmidt 1929, pp. 41 3 , 599.
C 2.01. Am ong the most puzzling reports o f
Guaape Island guano archaeology is the brief notice in
a London magazine in 1871, describing gold orna
ments and other objects . . . cloth . . . said to have
paintings o f animals and symbols, o f which the colors
were well preserved . . . a stratum o f w oolen rags, five
feet thick, and reaching over a mile in extent. Gonz
lez de la Rosa, after inquiring about this discovery, did
not find reason to question its authenticity.
Bibliography: Athenaeum, London. no. 2267, April 8,
1871; Squier 1 8 7 1 -7 2 , p. 48; G onzlez de la Rosa
1908, p. 44.
C 3 .1 6. During 1868, a survey o f the Macabi Islands yielded traces o f habitation. In the ruins o f a
dwelling (otherwise not described) were found the fol
lowing objects:
1. O ne golden vase, ca. 15 cm. high.
2. Four w ooden idols (fig. 111-35, nos. 1, la). (C
3.1.)
3. O ne w ooden pedestal, suitable for a base for one o f
the idols.
4. Tw o wooden containers, one nearly square, re
sem bling a boat, and divided in two compartments. This piece measured .05 m. wide, .04 m.
high, 25 m. long (fig. 111-35, no. 3a). The other
vessel, o f cylindrical shape on a narrow foot,
stood .05 m. high and contained a brown, ochreous
powder. (3.2-3.)
5. O ne staff with a carved knob (fig. 111-35, no. 2). (C
3.4.)
6. Tw o necklaces, one o f figs (fig. II1-3 5, no. 5), the
other o f cacao beans (fig. 111-35, no. 4). (C 3 .5 -

6 .)
7. Several bone amulets o f rude manufacture, one
painted red, another carved to resemble a seals
head.
8. Two reddish pottery statuettes.
9. Tw o spondylus shells, one filled with vicua wool.
10. O ne fragment o f net cloth.

236

A ncient Am erica

11. O ne fragment o f cord or string, dyed alternateiy

red and white.
12. Fragments o f a cloth o f red and white squares.
T hese specim ens, to judge from the style o f the
wood carving, were all o f M ochica manufacture. They
represent the m ost com plete known assembly o f
Mochica dom estic artifacts anterior to the discovery o f
intact M ochica graves by W. D . Strong in 1946 (Strong
1947, pp. 4 5 3 -8 2 ).
During 1947 I sought to lcate these objets in the
M use Prhistorique et dEthnographie o f Bordeaux.
Because o f the war, the collections o f this museum
were not available for inspection. But it is not impossible that further records o f the discovery are preserved
in Bordeaux.
Bibliography: D ulignon 1884, pp. 15758, pl. V;
D elfortrie 1871, p. 508.
C 3-716. During 1871, Josiah Harris exhibited at
the Ethnographical Society in London a collection o f
objects from the Macab Islands. T he collection was
discovered during N ovem berand D ecem ber, 1870, on
N orth Macabi, in an area o f 80 square yards, at depths
o f 14 to 43 and more feet. Cotton rags were also shown.
On N orth Macabi, these had extended many hundred
yards, at an average thickness o f five feet, and below a
deposit o f several feet o f guano.
According to Adrin D igby, Curator o f the Depart
ment o f Ethnology, British Museum, the collection was
presented to the British M useum on Septem ber 15,
1871, by Josiah Harris and A. W. Franks, where the
objects o f pottery and wood were registered as nos.
7 4 1 6 -4 7 . Several have been illustrated by T. A. Joyce.
Mr. Digby kindly prepared the follow ing extraer from
the British Museum registers:
7416. Tw o fragments o f large convex terracotta vessel with m oulded figures in high relief on con
cave side. . . . Presented by Josiah Harris,
Esq.
7417. Fragment o f rim o f large terracotta vessel (?),
having a seated human figure with right arm
outstretched, broken off; features obliterated; found in the guano. (N ew ell C oll.) Pre
sented by A. W. Franks.
7 4 18. H ollow figure o f red earthenware in form o f a
male captive, hands tied behind back . .
forms a vase . . found in the guano. (N ew ell
Coll.) Presented by A. W. Franks.
7419- Broken red earthenware vase, representing a
captive; found in the guano. Presented by
Josiah Harris, Esq.
7420. Seated w ooden figure, legs crossed, left hand

7421.

7422.

7423.

7424.

7425.

7426.

7427.

7428.

7429-

7430.

7431.

on one knee; right hand broken off; snake or

cord around neck; figure hollowed; the lower
end round and solid; found in the guano.
(N ew ell Coll.) Presented by A. W. Franks.
H alf figure carved in w ood, hollowed, originally dark red, wearing sleeveless shirt; snake
encircles neck . . . ; found in guano. Present
ed by Josiah Harris, Esq.
W ooden vase in form o f a seated male figure,
hands tied behind back; cord twisted around
neck, down back, and bound round the
wrists; found in guano. (N ew ell Coll.) Pre
sented by A. W. Franks.
Imperfect w ooden vase in form o f a male fig
ure, hands bound behind back; snake twisted
round his neck biting end o f penis; on cylindrical tenon. (N ew ell Coll.) Presented by A.
W. Franks.
Solid wood carved figure o f a male captive;
originally dark red; hollowed . . . back and
part o f side broken o ff and lost . . . ; found in
guano. Presented by Josiah Harris.
Flattened male figure carved in w ood, color
originally dark red; fat hands on breast;
lozenge-shaped eyes filled with white pigment; figure is split; found in guano. Present
ed by A. W. Franks.
Figure o f a female, carved in w ood, originally
dark red; fat hands on breast; eyes and mouth
filled with whitish pigment. (N ew ell Coll.)
Presented by A. W. Franks.
Figure o f a female, carved in w ood, originally
dark red color; fat; eyes and mouth filled
with white pigment; hands on breast; found in
guano. (N ew ell Coll.) Presented by A. W.
Franks.
Rough figure carved in solid, heavy wood,
originally o f dark red color; found in guano.
(N ew ell Coll.) Presented by A. W. Franks.
Roughly carved standing male figure; arms
appear to have been tied behind; cord around
neck; solid, cut out o f heavy wood. (Newell
Coll.) Presented by A. W. Franks.
Upper end o f staff, in heavy red wood, in
form o f winged human figure wearing head
dress with fat semicircular front; playing pandeau pipes. (N ew ell Coll.) Presented by A.
W. Franks.
Mace o f heavy dark red wood; cylindrical
shaft . . . pear-shaped knob encircled by four
human heads; in good preservation; found in
guano. Presented by Josiah Harris, Esq.

Tow ard A bsolute Tim e: Guano A rchaeology

7432. H eavy w ood staff, lower part broken off;
spheroidal knob, four heads encircling the
base; a throne with canopy and seated figure,
eight small figures, etc. . . . (N ew ell Coll.)
Presented by A. W. Franks.
7433- H eavy w ood staff, dark red; cylindrical shaft,
lower end sawn off, upper end o f large cir
cular knob encircled by four heads surmounted by a throne with seated figure, etc. Pre
sented by J. Harris.
7434. H alf o f upper end o f a mace carved in wood,
representing a throned figure . . . ; found in
guano. (N ew ell Coll.)
7435. U pper end o f a w ooden staff in form o f a
double cone with broad central flange; four
taces encircle the base; cylindrical shaft,
cut off. (N ew ell Coll.) Presented by A. W
Franks.
7436. Part o f a bulb-shaped w ooden object, originally red, with collared ends; three horizontal
lines engraved on the bulb; found in guano.
(N ew ell C oll.) Presented by A. W. Franks.
7437. U nopened spine-covered bivalve shell found
in guano with the other objects. (N ew ell
Coll.) Presented by A. W. Franks.
7438. Valve o f a bivalve shell, outside covered with
spines; coral band round inner margin o f lip; a
smaller shell had woven cloth adhering to
outside. (N ew ell Coll.) Presented by A. W.
Franks.
7439. O blong piece o f rubbed-down shell pierced
with four holes in pairs at each end; surface
colored with verm illion paint. (N ew ell Coll.)
Presented by A. W. Franks.
7440. T w o fragments o f a series o f halves o f oval
brown berries strung on a twisted brown fiber
string; found in guano. Purchased from the
widow o f Captain N ew ell and presented by
A. W. Franks.
7441. Tassel made o f narrow braids o f chocolatebrown colored fiber fastened across at intervals with twisted strings o f yellowish cotton.
(N ew ell Coll.) Presented by A. W. Franks.
7442. Cloth in which gold was found in pieces about
the si2e and shape o f threepenny pieces;
oblong, closely w oven cotton; edges selvaged; covered with stitching. Presented by J.
Harris, Esq.
7443. Piece o f thick w oven mat o f cinnamon col
ored w oolen strings; found in guano. Present
ed by Josiah Harris.
7444. Fragments o f w oven cinnamon colored w ool

237

en cloth . . . ; found in guano. Presented by

Josiah Harris.
7445. Fragment o f w oven, cinnamon brown colored
cloth . . . ; found in guano. Presented by
Josiah Harris.
7446. Fragments o f soft, closely w oven dark brown
cotton cloth, several times folded and a stitch
passed through to keep them together; found
in guano. (N ew ell Coll.) Presented by A. W.
Franks.
7447. Closely w oven, cinnamon colored cotton bag
with two longitudinal slits in closed end; dou
bled eight times and a stitch passed through to
keep it together; found in guano. (N ew ell
Coll.) Presented by A. W. Franks.
In this listing, no. 7416 (fig. 111-28, no. 8) = C 3.7;
no. 7418 (fig. 111-28, no 7) = C 3.8; no. 7420 (fig.
111-28, no. 5) = C 3.9; no. 7422 (fig. 111-29) = C 3.10;
no. 7427 (fig. 111-28, no. 6) = C 3.11; no. 7430 (fig.
111-28, no. 3) = C 3.12; no. 7431 (fig. 111-28, no. 1) = C
3.13, no. 7432 (fig. 111-28, no. 2) = C 3.14; no. 7433
(fig. 111-37) = C 3.15; no. 7434 (fig. 111-38) = C 3.17;
no. 7435 (fig. 111-28, no. 4) = C 3.16.
Item C 3.13 closely resem bles item A 2.1 (fig.
111-24) but cannot be the same object, for A 2.1 is
registered at the British M useum as no. 7008. All spec
imens excepting C 3.13 are clearly o f Mochica man
ufacture. They closely resemble the Chincha Island
specimens figured by Hutchinson (fig. 111-39), but are
clearly different from H utchinsons, for this writer
noted the resemblances him self in a footnote.
Bibliography: Joyce 1912a, fig. 36, pl. VII, p. 42;
Joyce 1912b, pl. VIII, opp. p. 86; Nature, vol. IV, p.
196, 1871; T. J. H utchinson 1873, vol. I, p. 107.
C 3.01. On the Macabi Islands before 1873, Jos
Mara Garca reported the discovery o f pottery vessels
o f various sizes, som e representing birds. Large clay
containers held sheet gold figures and masks. Vast
quantities o f cotton cloth in bad condition were encountered, as well as many female, headless mummy
bundles. Several wood staffs appeared, som e 6 0 90
cm. long and 5 cm. in diameter, rounded at the head and
tapering to a point. All these objects were m et in the 4 5 m. level, as well as above and below this level.
Gonzlez de la Rosa supposed that the headless female
mummies pertained to virginal sacrifices and that the
islands were a sacred cemetery.
A mummy from the guano islands is preserved in the
Horniman Museum, London, but the writer was unable
to ascertain whether it is headless and female.
Bibliography: Gonzlez de la Rosa 1908, pp. 4 2 44.
C 4 .1 12. Charles W iener, who traveled in South

238

A ncient America

America betw een 1875 and 1877, formed and published a large collection o f archaeological materials.
Among these, many specim ens are figured as coming
from the Lobos Islands. It is not clear which group
W iener had in mind. N o such islands are described in
the text o f his travels. T he map o f his travels, however,
shows the Islas de Lobos o ff Chiclayo, in a position
corresponding to the group now known as the Islas
Lobos de Afuera, a well-known guano station (fig.
III-l 6).
The occurrence o f any w ooden artifacts on either
the Lobos de Afuera or Lobos de Tierra groups is most
improbable. The guano on these islands is in general
poor in nitrogen and has evidently undergone extensive leaching. The occasional w et conditions implied by
this fact would militate against the preservation o f
wood. T he artifacts in question are far more Iikely to
have com e from Macabi (G. E. Hutchinson).
W ieners objects (figs. 111-40111-43) are all o f
Mochica style and compare closely to the Mochica
specimens from other island groups.
The items are as follows:
C 4.1 Seated pottery prisoner (fig. 111-41).
C 4.2 W ooden vessel rim fragment, with m odeled,
running warrior, as on C 3.7 (fig. 111-40, left,
center).
C 4.3 Seated w ooden prisoner (fig. 111-40), top, left).
C 4.4 Staff head with standing, nude, male prisoner
(fig. 111-40, top, center).
C 4.5 Seated w ooden prisoner (fig. 111-40, top,
right).
C 4.6 Seated w ooden prisoner; flaring neck at top o f
head; hands bound behind back (fig. 111-40,
right, second from top).
C 4.7 Seated wooden prisoner (fig. II1-40, right,
th ird from top).
C 4.8 Seated w ooden prisoner (fig. 111-40, bottom,
left).
C 4.9 Standing w ooden warrior (fig. 111-40, bottom,
center).
C 4.10 Mummy bundle head in wood (fig. II1-42).
C 4 . 11 Staff o f w ood, the head carved as a throne and
canopy, containing a seated figure (fig. 111-43)C 4.12 Staff o f w ood similar to C 4.11 .

Items C 4.9 and C 4.4 are closely similar to the spec

imens figured by T. J. Hutchinson (fig. 111-44), from
Caete Valley. The others may be compared to the
specimens in figure 111-39.
It is possible, but unlikely, that W iener copied

H utchinsons engravings, changing the proveniences

and adding new details. The W iener collections in
Paris, as I have been informed by H. Reichlen and John
Rowe, do not include these objects, and they are not
enumerated in the registers o f the W iener collections.

APPENDIX B
Dr. G. Evelyn H utchinson o f Yale University has added the following remarks:
The guano deposits o f Per consisted o f the faeces of
sea-birds which had undergone a slight diagenetic
change in conformity with the great aridity o f the country. The modern bird colonies apparently produce from
8 to 10 cms. a year under the favorable conditions afforded by the best islands. Taking the lower estmate of
the rate o f increase in thickness o f a deposit, it would
require about 593 years to produce the thickest deposit
recorded, on Central Chincha Island.
The photograph o f the section on Central Chincha
Island, taken by Captain Merriman, shows a vertical
section o f the deposit in which twenty-two bands can be
counted. It is evident from the height o f a man standing
on the cliff that the thickness o f these bands in aggregate was 150170 cms., or each band was 6 .8 7.8 cms.
thick. T he stratigraphy therefore suggests annual deposition at the same rate as today. There is, however,
som e evidence from the low er part o f the photograph
and also from contemporary descriptions, that thinner
layers occurred. The stratification appears, notably
from a pair o f photographs published in the Boletn de la
Compaa Administradora del Guano, vol. XI, pp. 472
and 483, to have been remarkably horizontal, at least
on the N orth Island. There is som e evidence, to be
presented in full in a forthcoming monograph, that the
nature o f the bird colonies had changed during the peri
od immediately prior to exploitation in the middle of
the nineteenth century and that this change had led to a
loss o f guano, through the activities o f burrowing birds.
Part o f the discrepancy betw een the archaeological and
stratigraphic estmate o f the rate o f accumulation is per
haps due to wind erosion at a time when burrowing
birds had becom e com m on. The archaeological evi
dence from the Bollaert slab, which anyhow may have
com e from near the periphery o f the deposit where its
thickness was reduced, therefore tends to give vales
for the rates o f accumulation o f guano which are too
low. The result o f wind erosion would be to make any
od object appear older when measured in terms of the
time taken to cover the Bollaert slab. Professor Ku'olers chronology, therefore, is most unlikely to be over
contracted; if it is in error it is Iikely that it is not con-

Tow ard A bsolute Tim e: Guano A rchaeology

tracted enough. As his argument stands his conclusions
appear inherently probable.
The full data will be available shortly in a Survey of
Contemporary Knowledge of Biogeochemistry, III:
The Biogeochemistry of Vertebrare Excretion, to ap
pear in the Bulletin of the American Museum of Natural
History [XCVI (1950) e d . ] . 52

NOTES
1 Forthcom ing in the Bulletin of the American Museum of
Natural History (G. E. H utchinson 1948). See the abstraer
by Professor H utchinson o f the relevant portions o f this
paper, proffered here as A ppendix B.
2 T he possibility has occurred to various writers, but it has
never been developed because o f the lack o f a suitable
biological history o f the islands. M anuel Gonzlez de la
Rosa took the trouble, betw een 1869 and 1872, to send a
questionnaire on archaeology to thegovernors o f the vari
ous islands. His paper (Gonzlez de la Rosa 1908), however, expressed skepticism that valid findings on chronology
could be derived from the guano islands. T he paper by
Luis G am arra D ulanto (1942) is similarly inconclusive.
Biological writers, on the other hand, such as M urphy
(1936, p. 292), have been deterred from drawing reasonable conclusions by their estim ates o f immense antiquity,
irregular deposition, and mechanical com pression o f the
lower layers o f the guano caps. T he views are discussed at
length by G. E. H utchinson in his monograph.
3 Tschudi (1 8 6 6 -6 9 , vol. V, pp. 3 7 4 -7 5 ) remarks that in
1840, when he first visited the Chincha Islands, few ships
were in evidence, and the birds w ere still to be num bered
by the millions. In O ctober 1858, however, the ships and
the w orkers w ere so num erous that no birds w ere to be
seen. Only the great central stacks w ere left on the islands.
4 M urphy 1936, vol. I, p. 289. Professor H utchinson pro
vides the following com m ent to this passage: An even
m ore impressive piece o f evidence o f a general horizontal
stratification o f great regularity is to be obtained from the
photographs o f the N o rth Island deposit in 1853, pub
lished in the Boletn de la Compaa Administradora del
Guano, vol X I (1926), pp. 472 and 483, illustrating reprints o f papers o f Raimondi. These will be republished as
a com posite in my m onograph. O th er com m ents by Pro
fessor H utchinson in the following pages will be followed
by the initials G. E. H.
5 A lbert Van de Put, Late K eeper o f the Victoria and A lbert
Museum and authority on Spanish heraldry, has suggested
(in litt., Feb. 18, 1946) ad a te about 1 5 7 5 -1 6 0 0 . Inqui
n es addressed to Per yielded a theory to the effect that
the stone was perhaps a funeral m onum ent left upon the
islands by a Flemish prate in the late sixteenth or early
seventeenth century. (Communicated by Jorge Zevallos,
C hief of the Seccin H istrica at the Archivo Nacional in
Lima, in litt., N ov. 9, 1944). O f significance is only the

9
10

11

12

13

14
15
16
17

18

19

20
21

22
23

239

independent agreem ent, concordant with my view, upon a

possible late-sixteenth-century dating.
G othic letter was used in the inscriptions on the founda
tion stone o f the Escorial in 1563. See the account by Fray
Juan de San G ernim o 1845, p. 23.
Individuis were not granted land by the crown, but only
the labor o f Indians resident upon specifed lands. See
Zavala (1935) for a treatm ent o f the entire question.
Cieza de Len 1864, p. 27. A Spanish town named San
gallan existed as early as 1535, seven leagues up the river
from the sea. Jim nez de la Espanda 1881, p. xvii; Calancha 1638, p. 235.
G obernantes del Per 1 9 2 1 -2 6 , vol. IV, p. 126; vol. V ,p.
486; vol. X, pp. 3 3 5 -3 7 .
Paz Soldn 186263, pp. 565ff.; M ontesinos 1906, vol.
II, p. 258; Prez de T orres 1749, p. 10; Anonymous
194 0 -4 3 .
Prolonged investigation and correspondence with Peruvian and Spanish authorities failed to yield any description
o f the municipal arms o f Pisco. Surely such arms were
granted, ca. 1572; renew ed in 1640; and perhaps granted
again in 1688.
G uam n Poma de Ayala 1936, fol. 356. G uam ns colo
nial officials all carry long, slender staffs, or thick, knobby
clubs.
Such sodalite figurines, known from the central coast of
Per, are made o f material quarried in Boliva. (See
Ahlfield and W agner 1931, and Valcrcel 1933, pp. 21
48.) Valcrcel (1933, pp. 28, 33) m entions sodalite figu
rines like those o f Pikillajta from lea and Nazca valleys as
well as from the Paracas pennsula.
See A nonymous 1933, pl. X V II, 499, and pp. 8 1 -8 2 .
Baessler 1906, pl. 20, fig. 320; Hamy 1897, pl. LIV, 1 5 4 57 (Ancn).
Mochica staffs are figured by Strong 1947, pp. 466, 478,
481. Cf. also Squier 1 8 7 1 -7 2 , p. 54, figs. 11-13.
Baessler 190203, vol. I, pl. 33. T he staff, 161 cm. long,
has a cross-legged figure 15 cm. high on the handle. Cf.
also Schmidt 1929, p. 413.
Baessler 1906, nos. 237, 298, 460, 461. Baesslers materials in general are o f a style that may be identified as
Late, in the sense that they are either Incaic or immediately pre-Incaic. T he Squier fishes may also be compared
to the Late Chimu pressed blackware representations on
pottery o f the north coast, e.g., Baessler 1 9 0 2 -0 3 , vol. II,
pl. 72, figs. 2 6 2 -6 3 .
Baessler 1 9 0 2 -0 3 , vol. IV, fig. 408 (featherwork); Bennett 1946a, pl. 43 (Coll. Jo h n Wise, hat in velvet technique).
Seler 1893, pl. 13, fig. 5 (Macedo Coll.).
As in Schmidt (1929, p. 135), from Chim bte, or in the
specimen in the M useum o f the American Indian (5/
1746), 0.243 m. high from Chim bte. Cf. Tello 1938, p.
16.
Tello 1938, fig. 2 and p. 65; Larco H oyle 1944, p. 12.
As on the blackware'specimen from Lambayeque, figured

240

24
25

26
27

28
29
30
31

32
33
34

35
36

A ncient Am erica

by Bennett (1946, pl. 49, d), or in the pressed blackware

specimen from Chiclayo figured by Schmidt 1929, p. 212
(23 cm. high).
Cf. Tello 1938, p. 171.
It is absent from the Cupisnique wares discovered and
published by Larco H oyle (1941). But the dog vessel may
belong to a style that is still undiscovered on the mainland.
Baessler 1 9 0 2 -0 3 , vol. I, pl. 33, fig. 188. T he figure is 8.5
cm. high, on a staff o f 68 cm. overall length.
Mochica artifacts from Ancn are fairly num erous at the
Museum o f the American Indian: nos. 5/1732, 8/4538,
14/4540, 14/4544, 14/7671, 15/1449; from near Lima,
7/2721, 7/2724, 7/2735, 7/2740; lea, 14/4570. Peabody
Museum, H arvard, has a Mochica specimen from Pisco or
lea (75572, Farrabee collection, 1909). For Caete Valley
see T. J. H utchinson 1873, vol. 1, pp. 13839.
See Appendix A, items no. A 3.35.
T .J. H utchinson 1873, vol. I, p. 138, figure o n p . 139 (our
fig. II1-44).
See A ppendix A, items no. C 3 .7 -1 6 .
W iener (1880, p. 580) figures another set o f wood and
pottery specimens as coming from the Lobos Islands (fig.
II1-40). These correspond so exactly with the H am burg
and British M useum specimens that they might be taken
for parts o f the same find. W ieners collections are housed
at the M use de lH om m e in Paris. T he curators there
were unable to identify W ieners figures with specimens
at present in Paris. O n p. 617, W iener figures another
Mochica prisoner vessel, as from the Lobos Islands (fig.
111-40). O n p. 650, a w ooden head, very much like that of
H am burg no. 2, is shown with the same provenance (fig.
II1-42). O n p. 687, w ooden staffs (fig. 111-43) appear from
the Lobos Islands, closely similar to the Mochica staffs
figured by Joyce (our figs. 111-37,111-38) and T .J. H u tch
inson (our fig. 111-39). Also o f the same style are the staffheads brought to Bordeaux from Macabi (fig. 111-35).
Canas 1854.
The photographs of 1853 and 1860 support such an assumption. See the treatm ent by G. E. H utchinson (1948).
Stevenson (182529, vol. I, p. 357) observed only sev
eral small vessels constantly em ployed at the Chinchas in
1825. In 1840, Tschudi (1 8 6 6 -6 9 , vol. V, pp. 3 7 4 -7 5 )
saw ships arrive at the islands only rarely. B ut in O ctober
1858, a great flotilla was steadily engaged, and no birds
were in evidence, although Tschudi had seen millions in
1840.
Contrast the views held by M urphy 1936, vol. I, p. 292.
Among many unproved assumptions, this is the weakest.
Caas (1854, p. 25)gives areas and volumes o f guano for
each island. A rea divided into volume yields a mean depth
of 12.12 m. This suggests the lenticular form of the guano
cap, thinning toward the periphery o f the island. (See
M urphy 1936, vol. I, pl. 16.) B ut the depth recorded finds
were made late in the removal o f the cap and therefore
near the deep center, in stacks such as the one shown in

37

38
39
40

41

42

43

44

45

46

47

figure III-17. T he only exception is the Bollaert slab,

found in 1847.
If the B ollaert slab was under an atypically thin peripheral layer, all the postulated dates based on its use are
too od. It would be as well to point out that this possibility, which is the greatest potential source o f error, can
only lead to a false expansin o f the chronology. (G. E. H.)
Caas 1854, pp. 2 0 -2 3 . This surveyors depths are given
in Peruvian varas (.8475 m.). H is deepest soundings:
N o rth Chincha, 50 varas; Central Chincha, 56 varas;
South Chincha, 52 varas.
On the Inca conquest o f the valleys o f the south coast of
Per, under T opa Inca ca. 1470, see Rowe 1946, p. 207.
Strong 1947, pp. 4 5 9 -6 4 .
Cieza de Len 1864, pp. 27, 26566 (M arkham translation). This work was first published in 1552. The writer
has verified M arkham s translation o f sand hills in the
Antw erp edition o f 1553 (p. 14).
Lpez de Velasco 1894, p. 492; Lizrraga 1946, p. 88.
H errera y Tordesillas (1601, p. 60) is another geographer
who repeats Cieza w ithout verification. It is likely that
Garcilaso de la Vega, who left P er in 1560, also knew no
more than Cieza (1553, fol. 102).
Apuntiamientos (sic) para el buen gobierno del Per, MS
cited by Jim nez de la Espada 1881, A ppendix 2, p.
cxxxviii.
Acosta 1940, pp. 32829. A much later w riter, in 1792,
reports that Chincha Island guano was mainly used in
Chancay Valley (U reta 1792, p. 220). Cf. Tschudi 1847,
pp. 2 3 9 -4 2 . Chancay Valley had consum ed ca. 33,00036,000 fanegas (cwts.) annually since 1790.
Asia Island is perhaps an unlikely source; the island
had little guano in the last century, as its form does not
perm it much accumulation. T he Chincha Islands are far
more likely, particularly since U reta (1792) and Humboldt noted that boats regularly brought guano from the
Chinchas to Chancay. (G. E. H .)
Foundation o f Pisco ca. 1572: G obernantes del Per
192126, vol. IV, p. 126 (hereafter GP). Pisco forbidden
to ship mercury: G P, vol. V, p. 486. Pisco sacked by pi
rates and subordinated to lea: G P, vol. X, pp. 312, 335
37. All mercury shipped through Chincha Harbor, GP,
vol. X, p. 271. Cf. Prez de T orres (1749, p. 10) on
Chincha H arb o r ca. 1588, when Pisco shipped only rhe
wines o f lea. O n one m ercury shipping route, see
W hitaker 1941, p. 14.
M ontesinos 1906, vol. II, p. 258. T h e ame was changed
to San Clem ente de Mancera, and the town was given
jurisdiction over the pampa de Chincha.
Clouds of guano dust: Cobo 189093, vol. I, p. 180 (writ
ten 1652 b ut probably referring to events of the I640s)
Anonymous 1 9 4 0 -4 3 , vol. X IV , p. 30. This last source
describes the quake and flood o f 1687 (vol. X III, p. 38).
Cf. Paz Soldn 1 8 6 2 -6 3 , pp. 565ff.
Cieza de Len 1864, pp. 26566; Garcilaso de la Vega

Toward A bsolute Tim e: Guano A rchaeology

48
49
50
51

1609, fol. 102; Frezier 1717, p. 235; cf. Baessler 190203, vol. II, text, pls. 7273: in the Peruvian myths seis
played a part, because the inhabitants o f the seabord [sic]
believed that after death their souls w ere carried by them
to the island o f G uano.
Photograph o f the specim en in the museum at Hacienda
Chicln kindly com municated by Rafael Larco Hoyle.
Bird 1943.
G. E. H utchinson 1948.
It is far from unlikely that, once aware o f the importance
o f guano archaeology, curators here and abroad will discover appreciable num bers o f depth-recorded artifacts
not known to me. G reat Britain, however, has been thoroughly searched; Professor G. E. H utchinson and I addressed some 300 letters during 1944 to museums in
G reat Britain. The answers, at that difficult time, were

241

astonishing in their prom pt and thorough attention to the

problem . Only two collections yielded relevant material
(items C 3.716, C 3.01).
52 G. E. H utchinson, 1948
[It may be noted for the record that dating objects by guano
stratigraphy anticipated the carbn-14 results with some accuracy, but the layered condition was destroyed in the nineteenth century, and its association with ancient objects was
unique in the Peruvian islands. G uano archaeology was
stillborn, and nothing new is known to me since this publica
tion. A. W. Franks, who gave twenty-one items to the British
Museum in 1871, is probably Sir Augustus W ollaston Franks,
for whom, as its ow ner in Britain, the medieval Franks Casket
in the British M useum was named in 1867 when he gave it
there. g k

III.3
The Design of Space
in Maya Architecture

The notion o f elaborated shelter so far dominates

our concepts o f architecture that we often find it
difficult to think o f building in its monum ental
sense, apart from shelter. As m onum ental form,
architecture com memorates a valuable experience
by distinguishing one space from others in an am
pie and durable edifice. Such an edifce does not
need to end o se rooms: it may suffice to cancel
space by solid masses, or to inscribe space with an
otherwise useless system o f lines and shapes. The
simplest monum ental modes are the precinct, the
hut, the cairn, and the path. The precinct marks off
an area to be reserved; the hut endoses it in part;
the path signis a direction; and the cairn marks a
point by elevation. From precinct to stadium
forms one typological series; from hut to cathedral
another; from path to arcade-lined boulevard an
other; and from cairn to pyramid still another. The
combinations o f path, precinct, hut, and cairn
yield all the possibilities o f m onum ental architec
tural form, not only in terms o f solids, but also in
terms of the space bathing those solids.
In this context, the study o f the form o f ancient
American architecture requires concepts somewhat different from those customary in European
architectural history.
ROADS
Although we commonly think o f a roadway in
function o f a transportation system, it is not necessary to regard all roadlike construction as trans
portation devices. Transportation is merely one of
the possible and common functions o f a roadway.
1. It may serve to connect nothing, as on the

pampas o f the south coast o f Per, w here the immense netw ork o f abstract lines goes nowhere.1
The lines describe zom orphic and abstract fig
ures; they climb precipitous slopes w ithout regard
for any travelers com fort, and they termnate
abruptly after miles o f progress for no evident reason. The form o f these lines is clearly monu
mental, to inscribe some meaning upon the inhu
man and hostile wastes o f nature. For unknown
reasons, the lines indcate directions, as thegraphs
of a forgotten but once im portant human activity.
In this sense, their monum entality is as evident as
that o f a Maya stela or pyramid.
2. A roadway may serve singly and entirely as an
elem ent in the formal com position o f space. In this
event, the roadway need connect nothing, any
more than an esplanade connects. The roadway
serves as a form o f oriented longitudinal space,
directing attention from nowhere in particular to a
point of interest. Such is the case at Polol in the
Peten Maya area, where a roadway 40 meters
wide, bounded by parallel walls ca. 150 meters
long, forms an esplanade entrance into the central
plaza.2
At Teotihuacan (fig. 111-46), the entire site can
be regarded as such a roadway, running northsouth for over a mile, to provide a formal axis for
the lateral developm ent o f platform and building
groups. The roadway connects nothing; it only afords axial ord er.3
3. Roadways may connect places, although this
is not their primary purpose, which may be the
production o f agreeably ordered space. Such is the
case in a num ber o f Maya sites.
At Uaxactun, G roups A and B face one another
242

T he D esign o f Space in Maya A rchitecture

from south to north across a ravine spanned by a
graded causeway.4 The distance between the centers of the two groups is about 250 meters. The
plaza of G roup A is about ten meters higher than
Group B; G roup A is the m ore ampie and wider
enclosure. Thus the roadway connects U-shaped
courts which open toward one another in a reciprocal relation.
At Nakm, G roup E (the subordinate northern
enclosure) is connected with G roup A to the south
by a wide roadway. It is about 25 m eters wide and
it runs between long, discontinuous, and narrow
platform mounds in parallel rows. In this case, the
two groups cannot be regarded as reciprocating,
for the whole assembly has bifold rotational symmetry (Z-form), with the vistas in the roadway
blocked at both ends.5
Ixkn displays a variant on the pathway site. A
roadway about 750 m eters long runs north and
south connecting the temples on two small hills.
Midway between them and opening on both sides
of the axis of the roadway is a connected system of
small courts.6 At Labn, in the Puuc, a n o rth south causeway likewise connects two open court
systems 175 m eters apart.
4. Roadways serving to connect distant places in
the service o f an established system o f communication are familiar in Per, since the Classic stage
if not earlier, and in the Maya area, where elevated
and perm anent roads connect various ceremonial
centers o f northern Yucatn.7

243

roughly the early and the late shapes o f Maya

architecture.
In group design, this type of pyramid, with stairs
on all four sides, has four equal fronts, o f which
each suggests and requires an adequate spatial environment. Thus the radially symmetrical pyramid
is the most monum ental and commanding of pyra
midal forms, for all approaches to it are o f equal
valu, and the form tends to induce further radial
symmetry in its general setting among other edi
fices. It is a rare form in America, used sparingly
even by the Maya architects, who were most partial to it.
2.
Virtually all other large freestanding pyra
mids in M esoamerica and in the Andean area display only bilateral or m irror symmetry. The axis of
reflection is usually a ramp or staircase. All parts of
the edifice faithfully m irror one another to right
and left o f this axis. At best, this form commands
two environm ents instead o f four (Balakbal,
Structure V I),11 and usually only one, depending
on the num ber o f staircases. If there are staircases
front and rear, the building demands front and
rear approaches. If there is only a front staircase,
only the front approach requires spatial design.
The Pyramids o f the Sun and Moon at Teotihua
cn (fig. III-193) have but one approach.12 The
large pyramids o f Tikal (fig. III-197) are o f this
same type. But at Tikal, the rear and sides are of
such steep and forbidding setback angles that the
single-front approach, already indicated by the
flight o f stairs, gains compelling authority.13

PYRAMIDAL PLATFORMS
The pyramid, as an elem ent in group design, had
complex possibilities, according to emphasis on
special approaches; abutm ent with other edifices,
axial relation to other elements, elaborations in
plan, and so on. H ere we shall consider the freestanding single pyramid in relation to the space
surrounding it.
1.
In Shepards term s,8 the plan o f Pyramid EVH-sub at Uaxactn has radial symmetry.9 The
axes of reflection for the fundamental portion are
the four staircases and the four inset corners. If we
disregard its temple building, the pyramid called
El Castillo at Chichn Itz is exactly identical in
symmetrical form .10 These two edifices represent

P R E C IN C T S
The necessary counterpart in group design to the
freestanding pyramid is the precinct. If the free
standing pyramid functions as a great cairn in calling attention to the site, the pyramid also requires
boundaries to the special place it celebrates.
These boundaries may be m ere walls or ditches,
as in the fence o f hexagonal basalt columns at La
Venta in Mxico (fig. III-45).14 M ore commonly,
in Mesoamerica, extended platforms, sometimes
with buildings, served the limiting function. The
limiting platforms (fig. III-194) defined the basic
measure o f the space o f an enclosure; within the

244

A ncient Am erica

enclosure, or at its edge, the pyramid rose high to

mark the valu and importance of the place.

GEOMANTIC GROUPS
Frans B lom 15 first identified the form and function o f an assembly o f edifices in the Maya area,
serving as a m onum ent to the most significant
horizon positions o f the rising sun. In G roup E at
Uaxactun, the assembly consists of a pyramid facing east; across the court from it, three temples
occupy a northsouth terrace. From the eastward
facing point o f observation on the pyramid stairs,
the rising sun was seen to em erge over the northernm ost temple on the solstice m orning o f June
21; over the central temple on the equinox mornings of March 21 and Septem ber 21; and over the
southernm ost tem ple on the solstice morning of
D ecem ber 2 1 .16 This particular assembly is common from Ixkun to Ro Bec in the central Maya
area: twelve and possibly eighteen sites show it.
Ruppert supposes that their function in the origi
nal instance may have been astronomical but
provincially and decadently became merely ritualistic. Ricketson properly associated such assemblies with ritual and geomancy, rather than with
observational astronom y.17
Signifcant to us is the fact that the collocation of
pyramidal forms about a plaza obeyed relationships of celestial order and thus reflected the cos
mos in the spatial relationships o f the ritual center.

BALL-GAME COURTS
A ball-game o f ritual character, played with a rubber sphere on special courts by opponents wearing
distinctive costum e and protective armours, has
been an outstanding trait o f M iddle American
culture since rem te antiquity. The game was
played in the southw estern U nited States,
throughout Mesoamerica, and in the A ntilles.18
There is no evidence o f it in the Andean area.
As an architecturai form, the ball court (fig.
111-49) is most common in the Maya area. Essentially it is two elongated and parallel pyramidal
platforms o f specialized profile. The playing arca
consists of the alley between the platforms, two
end-fields, and the inner surfaces o f the platforms.

Types and subtypes throughout Mesoamerica are

defined by the combination o f vertical and sloping
planes of the platforms.
In plan the ball court displays biaxial radial symmetry. It is the m ost eloquent of all forms in displaying the M esoamerican concern for space design among edifices. T he surface of the pyramidal
platforms are meaningful, not only as the boundaries o f masses, but also as the boundaries of a
volume o f space between the masses. Profiling,
measurements, proportions, and symmetry of
forms all are governed by the requirem ent of
space design for a specific purpose. In the monu
mental frame o f reference, ball courts compose a
close-knit assembly o f precinct, pyramids, and
pathway, often enriched by buildings at the periphery (Chichen Itza).

BUILDINGS
Many ancient American buildings rise upon long,
shallow plans. They may en d o se a chain of narrow
rooms, as at the Maya site o f Nakum. Structure
D 19 there forms the Southern side o f the great
plaza. The building is more than 400 feet long and
only 30 feet wide, with about forty-four rooms,
opening north or south by narrow doorways. This
building type, in innum erable variants, is extremely common throughout Mesoamerica and
the Andean area. Sometimes it resembles the
Greco-Romn stoa, as at Chichen Itza (fig. 111-50)
or Pachacamac, consisting of a long wall with columns in front o f it, and roofing between wall and
columns. Usually the extended block buildings
rest upon platforms; the platforms allow for terraces in front o f the building and for staircases
connecting the terraces. Building, terraces, and
steps produce distinct yet related modes of spatial
organization.
Such edifices served mainly as precinct or boundary markers for specific areas and subareas of the
ceremonial center. O ften the extended block
building, as at the H ouse of the Governor in Ux
mal, is subordinated to a court on one facade.20
With the other, opposite facade, it dominates an
implicit court on its own ampie terrace.
Unusually revealing o f the Mesoamerican architects reticence in the design o f interior spaces is

T h e D esign o f Space in Maya A rchitecture

his handling of the quadrangular building enclosure. If we restrict the discussion to true buildings, relatively few examples are available at Uax
actun, Tikal, Palenque, Piedras Negras, Uxmal,
Kabah, Chichn Itza, Mitla, and Tula.

OPEN-CORNERED ENCLOSURES
At Uxmal the principal courts display rectangular
design, achieved by unconnected buildings. In
both the N unnery and the Palomas, the main entrance axis bisects the long sides o f the rectangle.21
There is m irror symmetry of left and right both in
the courtyards and in each o f the com ponent
buildings. C rner exits, lightly marked on the
south side, are left between the buildings of the
Nunnery group. The N unnery platform level rises
by several terraces from south to north, and the
north range forms a closed rear wall to the court.
In the Palomas G roup, the court is a place of passage to a second courtyard dominated by a
pyramid.
If the outer contour o f the plans of the two
courts is considered, distinct types are evident.
The Palomas contour tends to form a true rec
tangle. The N unnery contour approaches cruciform shape. The cruciform contour reappears at
Kabah, also in the Puuc district, in two quadrangles known as the Codz Poop and the W est
Group.22 At Mitla (fig. 111-48), G roups E, F, J, and
K all are open-cornered courts o f cruciform
contour.
G roup F at Tikal23 shows the composite features o f both the N unnery and the Palomas at Ux
mal. Structure 76 forms an open-cornered quadrangle with Structures 74 and 77, upon a plan of
cruciform contour, but Structure 75 joins the
quadrangle in a rectangular contour, as at the
Palomas.

CLOSED-CORNER ENCLOSURES
The enclosing ranges o f buildings m eet at the corners to form a closed court in the Maya area at
Palenque in the building called the Palace.24 H ere
the solution is indecisive, for only the northern
corners are closed, w hile the Southern corners re-

245

main open. T he contour of the entire court forms a

true rectangle, as at the Palomas G roup in Uxmal.
At Piedras Negras, in Courts 1 and 2,25 the cor
ners of the courts are open, although the designer
disposed o f the flexible double-range system used
at Palenque, with parallel vaults and piered facades. Fie refrained, nevertheless, from continuing this system around the corners.
M ention should be made o f unvaulted Maya
buildings in this context. At Uaxactun, Structure
A -IX 26 is a houselike edifice upon a high platform.
The ruins of its outer wall surround three sides o f a
court about sixteen m eters wide. Further details
are not known, but the form suggests wooden
roofing and supports upon a many-chambered
plan.
This disposition is repeated in two courts at San
Clem ente in the Peten with the unusual feature of
a connecting interior stairway leading from the
lower to the upper court. H ere again, the contour
is rectangular.27
At Mitla (fig. III-48),28 two forms of the closedcorner quadrangle appear: the first, represented
by Courts G, H , I, A, B, and C, has the cruciform
outer contour in plan because of the fact that the
enclosing ranges touch one another only in their
inner corners. In these buildings, the designer still
conceived the ranges as separate entities without
real connection in plan and structure. H e achieved
only the closing o f the crner exits, but not the
structural and formal unifkation of the court as a
whole.
C ourt D at Mitla, on the other hand, is a closedcorner quadrangle o f true rectangular contour. Internally, the ranges still fail to connect at the cor
ners. It is as if the open-cornered ranges o f the
Palomas G roup have simply been pushed together
until their corners met.
Courts G, H , I, A, B, and C likewise resemble the
N unnery at Uxmal, in that the ranges o f the opencornered, cruciform -contour group have been
pushed together until the corners met.
The unusual patio building at Chichn Itza
called the M ercado (fig. III-50)29 cannot properly
be considered as forming part of the quadrangle
enclosure series o f the Maya area. It pertains instead to the developm ent of the many-aisled,
colonnaded buildings of the Mexican period in

246

A ncient Am erica

Yucatn, when the designers, under Mexican influence, began to compose the interior spaces of
buildings. The Mercado patio is really a colonnade
of quadrangular form at m odest scale, rather than a
great courtyard enclosure of the type of the N unnery at Uxmal. Its closest parallel is found at Tula in
Structure 1 (fig. III-51).30
Only Chichn Itz and Tula show examples of
complete enclosure by continuous structure,
made possible by the use o f the colonnade. A
hesitating example o f colonnaded enclosure ap
pears at Palenque. But at Tikal, Uxmal, Kabah,
and Mitla, the designers were reluctant to free
themselves from the concept o f enclosure by sep
arare masses. The Maya architect usually thought
of his building as masses. H e composed them
from the outside, juxtaposing them for effects in
space design. T he Mexican architect at Tula, on
the contrary, thought o f a building as a hollow
volume of which the interiors should and could
be carried around corners and through different
levels. The architects o f Mitla thought much like
Maya builders, but they experim ented hesitatingly with colonnaded volumes in G roup E.

2. By the skillful spacing of ranges of buildings,

the architects could achieve the effect of manystoried buildings, either on artificial or on natural
terraces. In the Maya area generally, this device
was common. In the Peten, the acropolis plat
forms provided the necessary terracing. In the Ro
Bec district at the center of the Maya area, natural
contours were used. In the Puuc and Chenes districts, a pyramidal core of artificial origin was faced
with ranges o f buildings simulating many-storied
edifices.33 The same device appeared in northern
and eastern Yucatan, as well as on the G ulf coast of
Mxico, in the Totonac territory centering upon
Tajn, and in Aztec dwelling com pounds.34
3. G enuine superposition o f stories, with the
requisite strengthening o f bearing walls and planned distribution of loads, was common only in the
Andean area. Sporadic cases of true multistoried
building appear in the Maya area.

Genuine superposition is indicative of interest

in the design o f interior volumes and of a high
degree of technological skill. Both characterize
Andean architecture more consistently than that
of Mesoamerica. Occasionally, however, the Maya
architect achieved complex superpositions of an
audacity never attem pted in South America. Such
MULTIPLE-STORIED BUILDINGS
is the case at Palenque. The Palace tower, square
We m entioned earlier the urgent requirem ent of
in plan, rose four stories. An interior staircase,
the American Indian designer for differentiation
cleverly adapted to a central shaft, gave access to
by height. The various ways o f achieving diffour distinct levels o f vaulted corridors resting one
ferences of level are w orth review.
on top o f the o th er.35
Less audacious in structure but m ore massive m
1. Easiest and most natural were the differences
effect is the Caracol at Chichen Itza. Here the
of contour in the building site itself. M onte Alcircular plan consists o f two concentric annular
bn31 in Southern Mxico (fig. 111-47), for invaults surrounding a central masonry pier. The
stance, was a natural pyramid, rising steeply from
pier contains a spiral stairway leading to a vaulted
the valley floor. The architects took advantage of
second-story chamber. The cham ber loads only
its spurs, saddles, shoulders, and abrupt contours
upon the stairway pier. The use of the building is
by levelling and flling w herever necessary to
unknow n.36
achieve striking differences o f level. In the Maya
In general, if a m odern engineer or architect
area, the Usumacinta regin offered naturally terwere restricted to Maya materials and techniques,
raced sites dropping swiftly to the level o f the
he would not hesitare to superimpose several
river. The architects of Palenque, Yaxchiln, and
layers of corbel vaults provided he always were
Piedras Negras used these contours with consumallowed to align all the bearing walls vertically. For
mate skill to achieve graceful variety o f space de
if the bearing walls o f an upper story charge the
sign. Andean terracing is another example o f the
capstone or vault overhang o f a lower story, the
artificial enrichm ent of natural contours, both in
edifice collapses. But as long as the bearing walls
the highlands and on the coast.32

The Design of Space in Maya Architecture

or piers in superposition are vertically aligned, the
stability o f the entire complex is increased, within
the limits o f the crushing strength o f the stone.
The Maya architect may have resorted to super
position m ore often than the ruins allow us to
guess. In one case, he was perhaps careless about
vertical alignment o f bearing walls. A t Xtam pak, if
we may trust M alers drawings,37 the bearing wall
of the third-story vault charges nearly on the capstone o f a second-story vault.
At Tikal, superpositions appear to be common.
The most impressive is Structure 10.38 From
across the south ravine which it overhangs, the
spectator has the illusion o f a five-storied struc
ture; the cross section shows two terraces: the
lower with two stories and the upper terrace with
three.

COLUMNAR SPACES
A satisfactory history o f columnar supports in an
cient America has not been written. A t Teotihua
cn, a colonnaded vestibule appears in Classic time
at the upper level o f the edifice called Los Subter
rneos.39 In Southern Mxico, at M onte Albn,
on the same horizon, columnar supports mark the
facade o f M ound X and other buildings.40 Both at
Teotihuacn and M onte Albn these supports are
square in plan and date from the Classic stage.
In the Maya area, Classic stage sites fall squarely
in two geographical divisions. Colum nar supports
are absent in the Petn and M otagua districts, al
though heavy piers resembling portions o f wall
occasionally separate the doorways o f multichambered buildings, as, for example, at the Copn ballcourt temples. T rue columnar supports appear in
the Usumacinta regin and throughout the Classic
stage sites o f Yucatn proper. N arrow piers and
squarish columns are proper to the facades o f the
Usumacinta regin; round columns characterize
the Puuc district, northern Yucatn, and the East
coast.
It is often implicitly assumed that pier, square
column, and round column represent a development sequence. Actually the three types must be
regarded m ore as geographical peculiarities than
as evolutionary steps in a development. W ooden
supports were used in the earliest stages of build

247

ing history. The concept o f a shafted support is

one o f the oldest o f all building devices. The rep
lica in stone or clay is not much younger. The
signifkant variable in respect to chronology is not
the form o f the shaft but the context in which it is
used.
At the Petn and Motagua sites, the builders
were chiefly interested in the massive design of
space by pyramids and platforms. The chambers
enclosed by Petn and Motagua buildings were
incidental to this overriding interest. The facades
were designed to retain mass and density. The
doorways were m inor incidents in the massive and
sculptural treatm ent o f the temples and palaces
of Copn or Tikal. This concept o f space design by
masses is in fact older than the isolated, coherently
chambered plans o f the Usumacinta regin, the
Puuc, and Yucatn.
At Palenque, Yaxchiln, and Piedras Negras,
space design by masses was a continuing tradition,
but it was enriched by highly articulated buildings,
such as the palaces and temples, with narrowpiered facades and complex cham bered construc
tion, interior buttressing, and skeletal roofcombs.
This entire constellation o f building traits signified, in Late Classic time, a profound shift in
architectural form from sculptural masses to organized interior volumes o f increasing spatial complexity. That round columns are completely lacking in the Usumacinta regin is a geographical and
not a chronological peculiarity, for coeval sites in
the Puuc district display a lavish use o f round col
umns, both freestanding and engaged.41
Ultimately, in the Mexican period of Yucatn,
the round column became the dominant form at
Mayapn, Ake, or Tulum, but it never entirely
displaced the square shaft. The reason for the in
creasing abundance o f round columns in PostClassic Maya architecture is probably to be found
in the spatial context o f their use. As the organization o f interior volumes became m ore and more
spacious, the square column, with its pierlike suggestion o f a mass of wall, was seen to obstruct the
interior spaces with too massive a bulk. The eye
slips around a cylindrical shaft. T he round column
is visually absorbed by the spaces it supports,
while a square shaft arrests the eye with a blocklike
recollection of wall surfaces.

248

A ncient Am erica

In any case, the coherent organization o f chambered volumes requires columnar supports or
piers. If interior spaces are to display continuous
design, they must melt one into another by gradual
transitions. Only the column and the colonnade in
ancient building practice perm itted these subtleties. But the resources o f the column were not
called upon until the desire to organize interior
spaces arse among the architects o f the Classic
stage in the western and northern Maya areas.

CHRONOLOGICAL DEDUCTIONS
Architectural space is distinguished from the
space o f m ere habitation by monum ental order
and imposing size. T he design o f architectural
space in ancient America offers three general
categories:
1. Fenced-in enclosures o f geom etric plan, defined mainly by thin-walled construction.
2. Courtyards and plazas in several levels, formed
by the collocation o f impressive masses such as
pyramids and platforms.
3. C oherent organization o f cham bered and roofed volumes, as buildings and groups of
buildings.
There is reason to hold that class 1 represents
the oldest m ode of space design in Mesoamerica
and in the Andean area. At La Venta, in Veracruz
(fig. 111-45), the great plaza at the north o f the site
is an enclosure roughly 60 x 75 m eters, fenced in
by a close-set row o f hexagonal columns of basalt.
Some o f the columns weigh over two tons; all are
set on the clay floor o f a sunken plaza. The court is
entered by a wide south aperture, flanked by
smaller rectangular enclosures o f the same basalt
hexagons.42 The grandiose character o f this sys
tem firs well with the colossal heads o f the area.
N o other example is known in Mesoamerica.
To date we know nothing conclusive about the
age o f La Venta. Typologically, however, the site
must be regarded, even if o f very late date, as a
survival or vestige o f a primitive m ode o f m onu
mental enclosure, far older, typologically, than the
precincts formed o f pyramids and platforms.
Class 2, o f courts bounded by pyramids and plat
forms, is o f an antiquity going back at least to the

G reat Cult stage, with the greatest development

marked in the Classic stage. O f Classic stage date
are the pyramidal and platform courts o f the Maya
area, from northern Yucatn (Oxkintok and Cob)
to the Usumacinta, Petn, and Motagua sites of
Guatem ala and H onduras. T he acropolis of Copn
is the classic example. M onte Albn, Teotihuacn,
Xochicalco, and Tajn in Mxico are other examples illustrative o f the technique o f designing
space by massive assemblages.
The coherent organization o f chambered vol
umes is a developm ent o f the Classic stage, timidly
proposed at first in massive buildings and gradually amplified during the terminal periods of
American Preconquest history. In Mesoamerica,
the phenom enon is closely linked with regional
variation in the articulation o f columnar supports
from the undifferentiated wall masses of Early
Classic and Pre-Classic periods.

CONCLUSIONS
1. European plan and structure generate such
complex enclosures that the history o f occidental
building has rightly been regarded as a Progressive
conquest o f enclosed spaces. In American antiq
uity, however, the rooms were at all times less
im portant than the masses. The design of a build
ing was secured far less by the enclosure of rooms
in an articulated envelope than by the ponderous
combination o f vast masses, solid throughout,
sculpturally related to one another, and structurally dependent upon simple static accumulations o f building material.
2. The concept o f a unit of architecture differs in
Europe and in ancient America. In Europe, the
sacred building, the m arket, the courts of justice,
the palaces, and the houses are each directly recognizable as isolated and functionally defmite entities. In American architecture such clear distinctions are often impossible. The greatest uncertainty surrounds the identification o f palaces and
tem ples. T he room-clusters that might have
served as residential units usually blend and merge
with other types, such as pyramidal platforms with
tem ple buildings, with ball-game courts, with
sweathouses, colonnades, and courtyards, to the
point at which clear distinctions among building

The Design of Space in Maya Architecture

units become difficult. At Copn, for example, the
vast primary platform covering twelve acres underlies and supports quantities o f secondary platforms; these in turn bear other platforms, pyra
mids, and buildings that merge with one another
to such an extent that Copn can be regarded as
one giant construction o f subordnate parts continuous with one another.
3. A cardinal objective o f the American Indian
architect in all periods and regions was to achieve
differentiation by height. The ceremonial centers
and the cities display a multiplicity of level that
probably distinguished the hierarchic rank of the
vague functions to which the edifices were dedicated.
For example, Tikal in the Petn, the largest and
perhaps oldest o f Maya sites, consists of courts and
plazas arranged in some eight groups upon an area
between two and three miles in diameter. A cross
section from south to north43 through the most
densely built regin reveis the site as a cluster of
acropolis platforms o f artificial construction, sep
arated by natural ravines and by sunken plazas.
Between acropolis and sunken plaza, the plat
forms and buildings rise in graded ranges, leading
from large, lower courtyards to progressively
smaller but higher courtyards, and ultimately, to
the diminutive temple chambers surm ounting the
great pyramidal platforms. Among the various
building groups, the ravines and valleys are spanned by immense causeways, both level and graded,
leading from level to level.
At Copn, on the other hand, w here the level
river valley gave no marked natural differences of
contour, the builders slowly constructed an ar
tificial hill over 30 meters high, covering twelve
acres o f ground, and rising from the river by a now
lost grand stairway.44
The Usumacinta river valley sites, at Piedras
Negras, Yaxchiln, and Palenque, all enjoyed the
rolling contours o f a narrow river valley. H ere the
Maya builders o f the Classic stage could m ost easily achieve the ascending ranks o f courts rising in
hierarchic order from the lowest esplanades and
plazas to the highest temple buildings overlooking
the smaller, more withdrawn and lofty quadrangles.45 Thus, in ancient America, architecture
was endowed with a kind o f dimensionality lacking

249

in much Od W orld building, that is, the artificial

height o f the building grade above the natural site
grade.
4. In general terms, the American designer was
far more sensitive to certain spatial aspects of ar
chitecture than the European builder. It is axiomatic that any building creates a surrounding
space. Strictly considered, the spaces engendered
between and among buildings have not often been
regarded as significant in European architecture.
In American Precolumbian building they were
treated as a primary source o f architectural effect.
The American Indian architect was restricted by
technology to the assembling o f solid masses, but
in the operations of design, he was infinitely more
attentive to their harmonious combination than
the Europeans. This special field in which the
American excelled was the achievement o f large
and rhythmically ordered open volumes.
The open volume, composed by richly articulated surfaces and related by storied changes of
level, is the most strikingly developed formal aspect of ancient American building. Painstaking attention to the spatial forms among edifices is lack
ing at no m ajor architectural site in ancient Amer
ica.46
To summarize, the principal formal aspects of
Maya architecture concern the dominance of
masses over the enclosed rooms, in a system of
poorly differentiated functional building types,
organized by striking differences o f level and
height, and deliberately composed in respect to
the spatial environm ent generated between or
among edifices.
NOTES
1 Kosok and Reiche 1947, pp. 2 0 4 -0 5 ; Reiche 1949, fig.
58.
2 Lundell 1934, pp. 1 7 7 -8 2 , fig. 3.
3 Gamio 1922, vol. I, pl. 8.
4 Ricketson and Ricketson 1937, fig. 198.
5 Tozzer 1913, pl. 33. Shepard 1948, p. 221.
6 Morley 1 9 3 7 -3 8 , vol. V, pt. 2, pl. 196a.
7 Villa 1934, pl. 9.
8 Shepard 1948, p. 221.
9 Ricketson and Ricketson 1937, fig. 33.
10 M arquina 1951, p. 848.
11 R uppert and D enison 1943, pl. 69.
12 Gamio 1922, vol. I, pl. 12.

250
13
14
15
16
17
18
19
20
21

22
23
24
25
26
27
28
29

A ncient Am erica

M arquina 1951, pp. 54144.

Covarrubias 1946, p. 92.
Blom 1924.
Ricketson and Ricketson 1937, fig. 68. M orley 1946, pp.
333 -3 4 .
R uppert and D enison 1943, p. 5. Ricketson and Ricket
son 1937, p. 107.
Schroeder 1955, pp. 156 -6 1 . Blom 1932, pp. 4 8 6 -5 3 0 .
Satterthw aite 1944, pp. 3 -7 .
Tozzer 1913, p. 170, fig. 55.
M arquina 1951, p. 763. H olm es 189597, pp. 8 0 -9 6 .
Ruz Lhuillier 1945, p. 40, holds that the Puuc-type quadrangle at D zehkabtun (Chenes district) is o f the N unnery
type.
M arquina 1928, p. 67.
Tozzer 1911, pl. 29, structures 7 4 -7 7 .
Maudslay 1889-1902. H olm es 1 8 9 5 -9 7 , pp. 1 6 7 -8 6 ,
pl. X X IV .
Satterthw aite 1935. Proskouriakoff 1946, no. 4. Morley
1 937-38, vol. V, p. 2, pl. 202.
Ricketson and Ricketson 1937, p. 34.
Morley 1 9 3 7 -3 8 , vol. III, p. 450; vol V, p. 2, pl. 207.
H olm es 1 8 9 5 -9 7 , pp. 2 2 7 -7 9 .
R uppert 1943, figs. 1 -2 .

30 Acosta 1945, pp. 2364.

31 M arquina 1951, p. 313. Rubn la Borbolla 1953, pp. 134
40.
32 Fejos 1944, fig. 10 (Inty Pata).
33 M arquina 1951, p. 747 (Sayil).
34 Vaillant 1939, pl. 41.
35 H olm es 1 8 9 5 -9 7 , figs. 5 4 -5 7 .
36 R uppert 1935, p. 275.
37 Spinden 1913, figs. 1 4 2 -4 3 .
38 Tozzer 1913, pp. 1 1 2 -1 3 .
39 Gamio 1922, vol. I, pl. 53.
40 Caso 1935, pp. 1 3 -1 4 .
41 Andrews 1942, pp. 26263.
42 Covarrubias 1946, p. 91. Stirling 1940; 1943.
43 Tozzer 1911, pl. 30.
44 Stromsvik 1947, p. 63.
45 Proskouriakoff 1946, no. 4.
46 C ontrast the situation in classical G reek architecture as
analyzed by R. Scranton 1949, p. 252.
jA n addition to the rem arks on colum nar spaces (pp. 247-48)
is my Serpent and Atlantean Columns: Symbols of MayaToltec Polity, Journal of the Society of Architecturai Histo
rian , 41:2 (1982), 9 3 - 1 1 5 . g k ]

111.4
Polygenesis and Diffusion:
Courtyards in Mesoamerican Architecture

The recent revival o f diffusionist arguments for

American antiquity, sponsored by R. H eineGeldern and G. Ekholm, depends mainly upon
visual comparisons to prove the Od W orld origins
of New W orld civilizations in Precolumbian time.
Both Ekholm and H eine-G eldern proposed
Southeast Asian origins for many American architectural forms. I have suggested that for every
one of these an older European parallel can also
be proposed (trefoil corbelled arches, miniature
roofed buildings inside temples, sacred tree or
cross forms, colonnette decorations, Atlantean
supports, and m outh-shaped doorways). The thesis o f Asiatic origins is thereby diluted to include
the entire Od W orld. Thus the Asiatic focus
loses both precisin and m eaning.1 To trace these
themes through long evolutions in America remains difficult or impossible.
To this list of visual parallels I propose now to
add courtyard designs. In order to be consistent
the diffusionists would have to arge from visual
similarities between the Romn atrium house and
the colonnaded courtyard designs o f Mxico and
Yucatn, that the two are historically related. They
would also have to produce Southeast Asian links
for the chain o f diffusion. It is my contention that
the two forms are unrelated, both historically and
visually, being independently convergent rather
than causally related, as I shall prove by a recapitulation of the history o f courtyards in ancient
America. This topic has the great advantage over
others of allowing us to examine many successive
links in coherent traditions of design both in the
Od W orld and in America. This is still not possible with Olmec and similar motifs, upon which the

diffusionists have anchored many o f their contentions.

As an architecturai form, the courtyard
throughout history displays an extremely wide
range o f possibilities. The elements o f courtyard
design are few and simple, yet their expressive
range, as determ ined by the historical examples, is
practically unlimited, like the expressive range of
black-to-white gradations in pictorial schemes.
The courtyard not only docum ents the fundamen
tal modes o f social organization, but also serves to
perpetate and intensify any given mode.
The archetypal basis o f all courtyard compositions, w here the effort is made to mark off a reserved area, is the precinct.2 Every civilization and
every generation produces its own variations upon
this fundamental monum ental mode. Essentially
the precinct is an enclosure open to the sky above;
its scale, height, chambering, and ornam ent affect
the expressive meaning.
The variations may be secured in the boundary,
in the enclosure, or in the environs. The boundary
may be continuous or broken in various ways. The
enclosure may be a raised platform or a sunken
garden or a stairway of terraces, depending upon
the designers desire for focused or diffuse attention in the occupants. In relation to its surroundings a courtyard may reflect a wish either for privacy and retreat or for connectedness with the
environs, by means of gaps and leaks in the bound
ary.
Among all these modulations, the most significant elem ents in the consideration of ancient
American courtyards would seem to be the units
of which the boundary is formed; the intervals or
251

252

A ncient Am erica

gaps between these units; and the differences

among inner and outer floor-levels.
I
A useful sequence among M esoamerican architec
tural groupings emerges from study o f the corners
in courtyard designs. C ourts3 with open corners
characterize the oldest and simplest groupings of
platforms, as at La V enta (fig. 111-45), the Olmec
site in Southern Veracruz, or Oztoyahualco near
Teotihuacn. Closed platform corners, achieved
by continuous construction, first appear during
the Classic era, as at the Ciudadela o f Teotihuacn
(fig. II1-46).
Further examples of open and closed cornered platform com position appear at M onte Albn (fig. 111-47). The m ountaintop acropolis de
fines an oblong court. Its narrow south end is
highest, rising as an asymmetric cluster o f pyra
mids. At the north end, another cluster o f plat
forms surrounds a sunken court separated from
the main plaza by a barrier platform.
The east range is lined with great stairways and
the west side has three freestanding elements. O f
these the northern group, called System IV, and
the Southern end, called M ound M, reflect the
arrangem ent o f the whole acropolis upon a smaller
scale. The assemblage of platforms might be de
scribed as an am phitheater composition, affording
privacy and enclosure to large gatherings of peopie whose attention centers upon a dominant stairway and tem ple at the end or in the center o f the
enclosure. The essential parts are barrier plat
forms with closed corners at the front and sides
and a stagelike elevation either at the rear or in the
center. In orientation and grouping, the entire de
sign bears a striking resemblance not only to the
Ciudadela o f Teotihuacn (fig. 111-46), but also to
the Olmec grouping at La V enta in Southern Ve
racruz (fig. 111-45), where the northward succession of pyramidal elevation, long court, and north
ern precinct are geometrically identical, although
less imposing than at Teotihuacn and M onte
Albn.
A nother similar grouping appears at Tikal in the
Perdido Quadrangle, w here the south end of the
Morley Causeway is w ithout doubt a pyramid and

forecourt grouping. It is also tem pting in this connection to include the Palomas G roup at Uxmal,
where the pyramid and its court are separated
from a lower forecourt by a barrier platform to the
north, very much in the m anner o f Tikal, Monte
Albn, and La Venta.
W hen we consider courtyards surrounded not
alone by platforms, but also by true buildings, the
same difference between open-cornered designs
and closed-cornered structures reappears, but it is
transposed to a later era. For instance, the building
courtyard at Tikal,4 now called G roup F, has open
corners, like the N unnery at Uxmal. These build
ings both may be regarded as works of the Classic
era, prior to the MexicanMaya period at Chichn
Itz.
U nique among Classic Maya examples of
closed-cornered quadrangle construction is the
Palace building o f Palenque, w here the doubled
vaults turned both corners in the northern range.
O f the crner piers here nothing survives, and the
exact date of the construction is still uncertain, but
the design differs radically from that o f all other
Maya courtyard buildings in having a colonnaded
exterior crner.
At M onte Alban (fig. 111-47) two examples of
closed-cornered quadrangle buildings confront
each other across the Southern end o f the main
plaza. The courtyard building atop the Danzantes
Mound has eight chambers. M ound S bears a
square building with possibly fourteen chambers.
Only the foundation walls are determined. A
model in stone (M arquina 1951, p. 347) probably
reports the original appearance of these build
ings, which are placed m ore like temples than like
dwellings, although their plans closely resemble
the many dwelling groups scattered about the
northern shoulders of M onte Alban, such as the
house group which rises above Tom b 105. To the
developmental interpretation o f these house
group plans we shall return in a moment. For the
time being, their superficial resemblance to the
M editerranean atrium house may be noted.
Mitla (fig. 111-48) is another site where a shift
from open corners to closed ones may be assumed.
The Southern and w estern groups are open quadrangles o f Classic date (M onte Alban III). The
other edifices all belong to a later period, probably

Polygenesis and D iffusion

Post-Classic and pre-Toltec. These later buildings
form three groups, each composed o f three quadrangles. From north to south, they are the Church
Group, the G roup o f the Columns, and the A r
royo Group. T heir chronology has never been established beyond doubt, but if we distinguish
loose from tight organization, two stylistic phases
emerge. Courts E and F have widely opened corners. In Courts A, B, and C and G, H , and I the
corners are closed w ithout being turned. In C ourt
D the corners are closed by interlocking chambers. The exterior effect suggests a suburban villa,
jealous of privacy, turning closed walls upon the
visitor, not seeking a coherent space with the
neighboring edifices.
Similar to these closely linked courts are the
buildings discovered by N oguera at Xochicalco
and called the Palace (fig. 111-49). As in C ourt D at
Mitla, the corners are closed by chambers which,
however, never turn the crner, so that no room
has a bent or angled axis.5
The newly excavated Palacio de las Mariposas at
Teotihuacan, near the southw est crner of the
Moon Pyramid, was announced to newspapers in
July 1962 by its discoverer, Jorge Acosta. This
square courtyard, surrounded by twelve square
piers, seems to belong to the type o f those at
Chichn and Tula, but its em placem ent and the
sculptural forms strongly suggest a date prior to
a . d . 700, when Teotihuacan was abandoned. U n
til further information is available, the Mariposas
Court appears to be the earliest known example of
the peristyle type, datable for the time being be
tween the M iddle and Late Classic periods, approximately a . d . 600700.
Surely later than these Classic period edifices of
Monte Alban, Teotihuacan, and Mitla are the per
istyle Toltec buildings of Chichn Itza and Tula.
The excavator at Chichn, K. R uppert,6 calis the
Mercado (fig. 111-50) a gallery-patio, and he asks
if it and other similar structures at Chichn Itza
may not have served as judicial courts connected
with marketplace activities. H e fixed the date of
the M ercado type at Chichn Itza after the opening of the twelfth century.
The excavator at Tula, J. R. Acosta, discovered
a similar but smaller patio building (fig. 111-51)
forming part of the extensive palace built upon

253

a platform adjoining M ound B.7 Two other

colonnaded chambers, Rooms 2 and 3, have rec
tangular impluvium bases like those o f Structures
3-B-3 and 3-B-8 at the M ercado and at Chichn
Itza.8 All these courts resem ble ceremonial enclosures more than houses.
Their proximity to im portant public spaces and
temple buildings makes it unlikely that they were
used as dwellings, either at Chichn or at Tula.9
The question nevertheless remains: how to explain the superficial resemblances between Mediterranean atrium houses and Mesoamerican court
yard buildings. T hat they are only formal convergences rather than functional similarides will
emerge from the following considerations.

II
The M editerranean atrium house was unknown in
ancient America. The distinguishing trait in the
Romn ho u se10 is that the adjoining chambers
continuously surround an axial, central court. The
court is like a cavity in the body of the dwelling.
The house as a whole was a body, with an envelope
surrounding the cavity. It sheltered household
fires, allowing the smoke to escape overhead,
while the pitched roofs shed rainwater into the
impluvium, or square courtyard basin. N othing
suggests that the Romn house arse as a coalition
or unin o f houses.11 It seems always to have been
unitary, from M esopotamian examples onward,
without separate com ponent traditions. It was
never a resolution or unification o f previously disconnected parts.
O ur Mexican and Maya courtyards, on the other
hand, em erge from a different tradition of domestic architecture. Instead o f the archetype of the
house-body, as in the Romn dwelling, we encounter another process o f formation to which we
may give the ame of the house-cluster. It is clear
ly evident at Teotihuacan, in the newly excavated
Tzacualli-period settlem ent of Oztoyahualco,12
where Plaza 1 has a radiocarbon date about the
time o f Christ. Three house m ounds form a court
yard with open corners. The south side is defined
by a low platform spanning the court. Similar dispositions seem to have been normal at Tikal,13 and
they governed, as we might expect, the arrange-

254

A ncient Am erica

ment of temple groups from an early date at

Uaxactn.14
Later on at Teotihuacn the big house clusters at
Tlamimilolpa and Xolalpan span perhaps five centuries, and in them both we can read the same
process o f clustering among separate units around
a common courtyard, whose originally open corners gradually were blocked with secondary constructions.15 This process is most clearly docum ented at the A tetelco house ruin (fig. 111-52),
where colonnaded vestibules (roofed?) led out
from the northern courtyard corners. O ther stages
in the process o f the filling o f the corners are documented in the Viking G roup and in the 1917
excavations.16
Linn estimated the Tlamimilolpa ruin as having
over three hundred rooms. Their arrangem ent is
less regular than at Xolalpan, which belongs to a
later period, but at both groups, the principie of
accretion reflects a fundamental clustering upon
three faces o f a rectangular precinct, as at Oztoyahualco. In this connection the house-group ruins
at M onte Albn are clearly house-clusters, symmetrically surrounding small sunken quadrangle
courts, and m ore open on the eastern entrance
face.
Perhaps the clearest o f all docum ents for the
Precolumbian house-cluster in Mesoamerica are
the sixteenth-century drawings reporting the form
of Aztec palace compounds. O ne of these, in the
agave-paper Mapa de Quinatzn (fig. 111-53), portrays a palace o f Texcoco during the reign of
Nezahualpilli at the cise of the fifteenth century.
The date of the actual drawing is 15424 6 .17 The
sense o f an open-cornered courtyard surrounded
by platforms, each with its row o f chambers, is
accurately portrayed w ithout European perspective conventions. It is a house-cluster arranged
around a public place, and nothing in it suggests
the organic unity o f the atrium plan o f the M edi
terranean world.
Codex Mendoza, which was painted on European paper after 154218 contains a perspective
rendering o f M ontezum as palace on two levels in
the new European style of drawing, but it too re
cords the house-cluster o f Precolumbian antiq
uity, uncontaminated by European ideas of the
M editerranean atrium house.

This review o f Mesoamerican courtyard types

displays a long seriation o f which the linkages are
still unclear. Y et the main m ovem ent from opencornered to closed-cornered courtyards is unmistakable first in respect to the platforms, and later
repeated in the compositions o f the buildings
themselves. T he first evolution perhaps began in
Southern Veracruz, and it was probably complete
by the period o f Teotihuacn II (Miccaotli phase)
before about a . d . 300. The second stage, in the
closing o f the corners o f courts formed by build
ings, was probably not achieved until the building
o f the Palace at Palenque long after a . d . 500.
The peristyle court o f M editerranean architecture
appears in ancient America first at Teotihuacan,
before a . d . 700, and again much later on, after
a . d . 1200, at Chichn Itza and Tula without any
historical connection. In regard to house groups,
the evidence suggests that courtyard clusters are
the tradidonal mode, and that resemblances to the
atrium-type house of other world areas are both
fragmentary and fortuitous.
In conclusin, the case is a useful and important
one, for it allows us to observe many stages in the
process o f convergence, whereby two traditions,
initially distinct, different, and unrelated, eventually arrived at approxim ate resemblance. The
diffusionist argument, on the other hand, is like
assuming a cise blood relationship between per
sons who look alike, although born many centuries apart, o f different races, and on different
continents.

NOTES
1 The most complete statement of the new diffusionist arguments is the group of essays entitled Asia and North
America: Transpacific Contacts, comp. Marian W. Smith,
Society for American Archaeology, Memoirs, no. 9 (Salt
Lake City, 1953). For the argument based upon motifs
appearing in art, see the essay by Gordon Ekholm entitled
A Possible Focus of Asiatic Influence in the Late Classic
Cultures of Mesoamerica, ibid., pp. 7289- To be added
to his bibliography are the major works by C. Hentze,
Rituels, croyances . . . de la Chine antique et de lAmertque
(Antwerp, 1936); Miguel Covarrubias, The Eagle, thejag
uarand the Serpent (New York, 1934);Harold S. Gladwin,
Excavations at Snaketown (Globe, 1937). The most com

plete diffusionist statement by a modern art historian is

the essay by R. Wittkower, Eagle and Serpent: A Study in

Polygenesis and D iffusion

2
3

5
6

10

the Migration o f Symbols, Journal of the Warburg and

CourtauldInstitutes, II (1 9 3 8 -3 9 ), 293325. See also the
three articles by R. H eine-G eldern: Representations of
the Asiatic Tiger in . . . Chavn . . " (I, 32126); Ch
nese Influences on the Pottery o f Mxico . . " (I, 207
10); and Chnese Influences in Mxico . . . the Tajn
Style . . " (I, 195206), in X X XI I I International Congress ofAmericanists, San Jos, Costa Rica, 1958, Proceedings
(San Jos, Costa Rica, 1959). My argum ents appear in Art
and Architecture of Ancient America (Baltimore, 1962), pp.
llff.
G. Kubler, T he D esign o f Space in Maya A rchitecture,
Miscellanea Paul Rivet (Mxico, 1958), II, 51531.
A m ore or less level and m ore or less square or rec
tangular area fairly set apart by platform and/or building
walls on two or m ore adjacent sides (L. Satterthwaite,
Piedras Negras Archaeology: Architecture, part 1. no. 1
{Philadelphia: University M useum , 1943], p. 17.
Linton Satterthwaite, in litt., writes that its veneer masonry is a safe indicator o f "Late Classic! To date only the
survey and related field notes have been accomplished by
the Tikal project.
I. M arquina, Arquitectura prehispnica (Mxico, 1951),
lm. 44.
K. Ruppert, "The M ercado, Chichn Itz, Yucatn, Car
negie Institution of Washington, Publication 546: Contribu
tions to American Anthropology and History, V III, no. 43
(1943), 2 2 3 -6 0 .
J. Acosta, La tercera tem porada de exploraciones ar
queolgicas en Tula, Hidalgo, 1942, Revista Mexicana de
Estudios Antropolgicos, VI, no. 3 (1944).
O f the o ther examples, only the C hultun-G roup structure
qualifies as a clearly defined courtyard type; the others (6E-3, 5-B-19, and 2-D-6) all are colonnaded chambers,
while 5-B-17 has two galleries but no colonnaded patio.
A nother peristyle court like those of Tula and Chichn is
at La Q uem ada in Zacatecas, with twelve cylindrical col
umns o f stone spalls. See Marquina, Arquitectura pre
hispnica, lm. 72 and fot. 97.
H. D rerup, Bildraum und Realraum in der rmischen
A rchitektur, Rmiscbe Mitteilungen, LXVI (1959), 147
74; and A. K. Lake, T he Origin of the Romn H ouse,

11

12
13

14

15

16
17

18

255

American Journal of Archaeology, XLI (1937), 509-601.

T he oldest known example o f courtyard houses of this
type is in Southern M esopotam ia, dated 2800 B.c. by radiocarbon. P. Delougaz, The Temple Oval at Chafajah,
O riental Institute Publications, LIII (Chicago, 1940), pl.
V.
T he idea that the tablinum was originally a hut and the
atrium a fenced-in courtyard in front o f the hut has been
discarded by Romn archaeologists in favor o f the idea of
an organic whole which developed independently from a
simple origin (Lake, Origin o f the Rom n H ouse, p.
599).
R. Milln, T he Beginnings o f T eotihuacn, American
Antiquity, X X V I (1960), 1 -1 0 .
R. F. Carr and J. E. Hazard, Map o f the Ruins o f Tikal,
Tikal Report, no. 11, M useum M onographs, T he U niver
sity M useum (Philadelphia, 1961).
O. G. Ricketson, Jr., and E. B. Ricketson, Uaxactun,
Guatemala, Group E, 19261931, Carnegie Institution of
Washington, Publication 477 (W ashington, 1937), es
pecially figure 22.
S. Linn, Mexican Highland Cultures: Archaeological Re
searches at Teotihuacn, Calpulalpan, and Chalchicomula in
193435 (Stockholm, 1942), pp. 9 1 -1 2 3 ; also Archae
ological Researches at Teotihuacn (Stockholm, 1934), pp.
4049. R. Milln and J. A. Bennyhoff, A Long A rchitec
tural Sequence at T eotihuacn, American Antiquity,
X X V II (1961), 561. L. Sjourn, Un palacio en la ciudad
de los dioses (Mxico, 1959).
I. Marquina, Arquitectura prehispnica, lam. 25.
D. R obertson, Mexican Manuscript Painting of the Early
Colonial Period: The Metropolitan Schools (New Haven,
1959), p. 139.
Ibid., pl. 26 and pp. 95 ff.

In reply to an article by B. M undkur appearing in Current

Anthropology this year, I mentioned Isomorphy Before Diffu
sion, which will appear in Notas Mesoamericanas (Mxico; in
press in Spanish). T he argum ent is that visual resemblances
m erit com parative study without assuming the diffusion of
isolated traits taken o ut of context. g k ]

III.5
Precolumbian Mural Painting

It is fairly clear today that fgural elements are

older in Maya wall painting than in vase painting.
Both these are older than the illustrations in
manuscripts. W hether the same sequence wall
painting first, followed by fgural vase painting,
and concluded by illustrated books also holds
for highland Mxico is still impossible to prove.
The truth is that we know far less about wall paint
ing the presum ptive source o f painted conventions than about pottery painting, which has
received the m ost attention because of the abundance of the remains.
The question I would like to examine is w hether
clearly marked stages can be found within the development o f mural painting, stages like those
which have been postulated for pottery painting.
In fact, among the many kinds o f ancient Am eri
can art found by archaeologists in the past century,
mural paintings were the last to be discovered. In
the Maya country, rain and vegetation destroyed
them, and in Mxico, the od walls were broken
and were consumed by fire and covered by drifting
earth. W hen H enri Beuchat published his Manuel
darchologie amricaine in 1912, he m eant it to be a
defmitive synthse o f all that was known. W ithout
reproducing them, he knew only o f the Maya
muris discovered at Chacmultun by E. H.
Thompson in 1904. H e also m entioned the wall
paintings in the Tem ple o f the Tigers at Chichn
Itz, which he reproduced only in a small Sundaysupplem ent sample of that rich series of scenes.
Finally, he illustrated a painted frieze in a dubious
drawing from the walls o f Chan-Chan on the north
coast o f P er .1
In 1943, about thirty years later, Pal Kelemen
had little more to say. H e relegated muris to a

short chapter near the end o f his survey, and he

illustrated only ten sites, including two in the
southwestern U nited States but none in Per.2
By 1946 the repertory compiled by Salvador
Toscano,3 the Mexican art historian, listed twentytwo Mexican and Maya buildings with wall paint
ings. But in that same year, Giles Healey dis
covered a building entirely decorated with Maya
muris at Bonampak in the State of Chiapas. This
event transformed all prior notions about Maya
wall painting. Bonampak was immediately recognized as a masterpiece of world art, and the study
of these extraordinary scenes is still in progress.4
In 1958, U N ESC O published an lbum on Pre
columbian paintings in Mxico. They are divided
into four geographic and cultural groups: the Teo
tihuacan style, the Toltec-Aztec style, the Classical-Maya school, and the Maya-Toltec style. The
classing is ethnic, by tribe or people. T here is no
mention of the position o f the style o f Monte Alban, although two o f the tomb paintings there are
splendidly illustrated. All the plates are of actual
frescoes, newly photographed.5
The examples now known all display clearly the
whole system o f ancient American pictorial conventions. W hether in Mxico or in Maya country,
the system always depends upon outline drawing,
colored in fat, local tones, without gradations of
light and shade and w ithout the illusion o f rounded bodies achieved by gradated color.
Many m ore paintings have been discovered
since 1946, especially at Teotihuacan, and their
appearance suggests the outlines of an interpreta
tion which considers form and content simultaneously, under a classing by pictorial systems
instead o f by subject or tribe.
256

Precolum bian Mural Painting

THE MISSING EARLY FRESCOES

Still lacking are any integral remains of the art in its
earliest stages. In the Valley o f Mxico, the oldest
large mural is probably a destroyed fresco found in
a lower level construction at the Tem ple of Agriculture at Teotihuacan. It exists today only in a
copy (fig. 111-54). The fresco probably was painted
in the period called Teotihuacan II, during a 500year span between 250 b . c . and a . d . 250. Unfortunately, we have no idea o f the preparatory stages
leading to this astonishing design, which I will discuss below in m ore detail, or have other exam
ples o f this type been recovered. Among Maya
muris, that in Structure B X III at Uaxactun (fig.
111-55) likewise is our earliest regional example,
and it records an advanced m om ent in the history
of Maya wall painting. The pottery and architec
ture found with the mural place its date between
the fourth and the sixth centuries a . d . depending
upon the choice o f correlation with Christian time.
Likewise at M onte Alban in Oaxaca two frescoed
tombs survive (fig. 111-56). The earlier one was
painted about a . d . 500. Thus we have to begin
everywhere with mature or even late examples of
painting. For the earlier stages, the prior ceramic
history sometimes suggests the presence o f a lost
art of monum ental painting, but it would be much
less justified to attem pt to reconstruct lost frescoes from vase paintings in America than in Periclean Greece. T here the attem pt is warranted by
texts describing lost paintings. Y et we now know
numbers o f muris in America, probably more
than in classical G reece, and it seems justified to
try to discuss them as a group.

PROBLEMS OF D A TIN G
It is now apparent that Teotihuacan in the Valley
of Mxico was the metrpolis for highland Mx
ico, radiating upon the early Maya settlem ents and
upon those o f Oaxaca in Southern Mxico. The
simplest divisin o f its long history is by Pre-Classic, Classic, and Post-Classic periods, spanning
about 1,500 years, from 600 b . c . until about a . d .
900. The early period embraces 600200 b . c .; the

257

middle or Classic period endured from about 200

B.c: until the seventh century a . d ., when the great
site was burned and largely abandoned. A PostClassic era followed the seventh century. At T eo
tihuacn proper, the muris appear at the cere
monial center itself, as well as in the outlying suburbs surroundingTeotihuacn, and they all belong
to the Classic period, 200 b . c . a . d . 700.
At M onte Albn the third period corresponds
to Teotihuacn II and III, without usable subdivisions. This duration is insufficiently defined both
at the beginning and at the end. In Maya territory,
the dating o f the frescoes at Bonampak and at
Chichn Itz is equally uncertain. A possible error
of several centuries still beclouds their position in
time, yet there is agreem ent that Bonampak was
painted between a . d . 550 and 810. The frescoes
of Chichn Itz in Yucatn were painted between
a . d . 1000 and about 1260, but their exact se
quence in that period is still uncertain. The de
stroyed muris o f Santa Rita are the least certain of
all: the accepted date has long been late fourteenth
or early fifteenth centuries, although a case can be
made for their being as early as the tenth century.
H enee, wherever we look, uncertainties of dating
are present, and it is very diffcult to be sure which
carne first, and which affected the others, so that it
is best to consider the types o f mural painting
without trying to fix their dates m ore exactly than
the tools o f chronology now allow. Perhaps the
pattern of distribution among the types we discussed earlier will allow us to define m ore clearly
than others have the question o f those regional
peculiarities in painting, which em erge when for
mal types are considered.
O thers, like Salvador Toscano6 in 1944, preferring to go by subjeets, have classed these paintings
as decorative, mythological, and historical. But
this classing by subject m atter immediately blinded students to the formal organization of the
frescoes and therefore to their functional significance and to their historical derivations.
Several distinct types of pictorial system emerge
from a study o f these muris. The main ones are:
friezes, processions, scenes, narrative registers,
and illustrations. Each o f these terms needs some
definition.

258

A ncient Am erica

THE FRIEZE
A frieze is a band of decoration composed o f repeatingelem ents, occupying an architectural field,
such as a cornice or a door frame (figs. 111-78 and
III-l 11 are friezelike). It is probably the oldest
type o f wall painting. It surely occurred long be
fore the invention o f pictorial scenes, although we
still lack any intact examples o f great antiquity in
America.
The function o f the frieze is to strike the atten
tion by repeating the same shape, and it serves the
same purpose as ritual and liturgy. O ften, when its
elements display bilateral symmetry (fig. III-l 11),
the effect is static, but when the elements are
asymmetrical (fig. 111-78), their repetition will
take the eye in the direction suggested by each
form. In friezes, this directional property is unorganized, in not being exploited beyond the simplest suggestion o f passage from one part of a
space to another.
Painted friezes as defined here seem rare everywhere excepting at Teotihuacn, where they are
very common. Elsewhere friezes are Iikely to be
made o f stone elements composed for striking
chiaroscuro effects, as at Tajn, Mitla, or Uxmal. In
Maya territories, painting was reserved for protected interiors and for the enhancem ent of m onu
mental sculpture, as at Piedras Negras, where the
reliefs were colored green and red. For the pre
sent, Teotihuacn seems to have been an originating center o f Precolumbian wall painting. Continuing excavations will probably show that this
apparent concentration is merely an accident of
preservation.

THE PROCESSION
A procession can be described as a directionally
organized figural band. It contains a variety of ele
ments or variations upon an element, instead of
the redundant m onotony of the frieze proper. The
procession usually has profile figures shown in
motion from both sides toward a center, thus containing varied actions o f a high degree o f unity
within a banded space. Friezelike files o f priests,
warriors, or animals, who converge upon a shrine

or image with attributes or offerings are common

forms.
Sacrificial processions are common at Teotihua
cn (fig. III-l 17 is part o f a procession), and they
recur in the tombs of M onte Albn (fig. 111-56).
These are both of Classic date and prior to the
seventh century a . d . The two pictorial traditions
are closely related. T he same type, with Indian-file
warriors and priests converging from both sides,
appears frequently at Chichn Itz during the
Toltec Maya period and at Tula, northwest of
Mxico City. A t both places, the type may be
dated in the twelfth and thirteenth centuries. It
probably had a connection with the ball-game
courts which were so im portant in Post-Classic
civilization throughout Mesoamerica. H enee processional compositions o f this kind seem related to
the ascendancy o f highland Mexican peoples in
Mesoamerican archaeological history.
Painted reliefs commonly appear among the
Toltec Maya processional compositions. Painting
alone is both evanescent and fugitive, but it can be
made m ore durable by applying it to relief sculp
ture in a m ore m onum ental expression. But in that
case it becomes cosmetic, for the artist has turned
his attention away from painting as the primary
illusion and towards sculptural problems such as
raised line or the intaglio field, with new relations
to changing sources o f light. The technical conditions are different, and the forms therefore differ.
T H E SCENE
U nlike friezes or processions, scenes describe
places with figures in varied actions. A scene may
describe one place or several within the same
frame, but some indication o f place is essential. It
does not accompany the spectator, as the frieze
does, but it arrests him, by drawing his attention
from the setting to the action portrayed.
W here there is a difference of function there is
often Iikely to be a difference o f forms. The func
tion of the frieze is to command attention by re
peating the same shape.
A scene marshals the spectators attention altogether differently. The function of the scene is
to absorb the onlooker into its variegated system.

Precolum bian Mural Painting

If a frieze resembles ritual and liturgy, a scene
resembles dramatic art. As ritual is typologically
older than drama, so is the frieze older than the
scene, without ever being displaced or replaced by
the newer function.
H enee friezes and scenes may be regarded as
different form-classes within ancient American
painting. T he frieze usually appears subordinated
to buildings, to vessel forms, and to textiles. But
the scene fixes attention m ore firmly upon its own
character as coherent illusion, and it requires a
setting undisturbed by other fields o f interest. The
mural found at a lower level o f Teotihuacan, later
built over so that the od walls survived with their
paintings beneath the new buildings, may be the
earliest fully developed scene we know
(fig. 111-54). It is known today only in a copy made
in 1886.7 The original has nearly disappeared
through weathering. The scene was painted no
later than the Early Classic period, that is, about
a .d . 300, as we know from the presence o f an
image of a tripod cylinder vessel of that period. In
the scene, a perspective system similar to that of
Egyptian Od Kingdom painting is in use. The
main elem ents are clearly represented. Statues
limit the field, and in front o f them are funerary
pyres, w here mummy bundles are being burned.
Between the pyres there are personages who
make bread and bring offerings, and in the bottom
register wavy forms represent water.8 A specific
place and m om ent are represented.
A nother large scene painted at about the same
time was discovered in 1942 in a house ruin of
Tepantitla, a suburb of Teotihuacan.9 The mural
consists of upper and lower registers. The upper
half (figs. 111-74, 111-75) displays the figure o f the
raingod Tlaloc, flanked by censer-bearing priests
from whose mouths come florid speech scrolls.
These signify their prayerful songs to the god of
fertility and germ ination, who stands upon a wavy
sea filled with marine creatures.
The lower half (fig. 111-73), according to Alfon
so Caso, shows the souls o f the dead rejoicing in
their eternal home. At the center is a mountain
upon a lake. In the lake are swimmers, and from
the lake spring two rivers flowing in opposite directions. Each stream is studded with eye-shaped

259

forms, perhaps representing the bubbles o f live

water. The right-hand stream flows to a lake bordered by trees and flowers. M ost o f the figures
sing and play games among butterflies. At the right
a singer is shown who perhaps had just entered
Paradise through the water o f the lake, bearing a
branch in his upraised hand. The animation of
these figures may represent a latergeneration than
the crem ation scene. They recall the celebrated
portrait figurines o f the Middle Classic period at
Teotihuacan, while those o f the cremation scene
evoke the figurines o f the Early Classic period, or
Teotihuacan II. To present knowledge, however,
the painters of Teotihuacan did not compose
scenes on several adjoining walls. This kind of pic
torial extensin was perhaps a Classic Maya
invention.
Room 2 at Bonampak (figs. 111-57, 111-58),
painted a century before the end of the Initial Se
ries Period, has one o f the most coherent mural
scenes in the history o f world painting. This battlescene is a triptych, covering three walls. It de
scribes a raid by Maya warriors upon a tribe of
stringy-haired, dark-skinned folk who live in
dense vegetation. Facing the triptych, on the entrance wall, is the arraignm ent of the captives. The
two scenes probably show related instants in the
capture and public display of the same enemy at
this border-station on the southwestern Maya
frontier. T he triptych is filled with the noise of
battle, while the judgm ent is hushed and static.
H ere all the figures are quiet, to hear the order
pronounced by the prince who is ciad in a jaguar
skin, pointing with his left hand.
D uring the Toltec domination of the Maya peopie at Chichn Itza, from about a . d . 1000 until
1260, this Maya tradition o f scenic composition
was continued in the frescoes which adorned the
Temple of the W arriors and the upper Tem ple of
the Jaguars. But these differ from the art o f B o
nampak in being small-figure designs. The sea
coast village and the raid with its procession of
captives, o f probable twelfth-century date, in the
Temple o f the W arriors (fig. 111-59), recall certain
conventions o f the south Mexican manuscripts,
usually identified as Mixtee, like Codex Nuttall,
although their freedom and animation might also

260

A ncient Am erica

bear comparison with the Paradise fresco o f T eoti

huacan. The Chichen Itza frescoes are today in
fragmentary condition, and they can be studied
only in watercolor reproductions made for the
Carnegie Institution of Washington.
A bout a century later, another generation of
painters at Chichen Itza left a much m ore rhythmic
system of crowded fgural action. The battle scene
from the upper Tem ple of the Jaguars is one panel
among seven, portraying a siege scene, two differ
ent battles, and a paradise scene. The best preserved (fig. 111-60) is a battle scene with about 120
figures grouped in eleven rows and eleven columns, making a rhythmic panoram a o f spearmen,
whose attitudes are in sequence, like the frames o
a cinematic or stroboscopic action exposure. O ur
illustration, redrawn by Jean Charlot in the 1920s,
shows the m otions in question.

THE NARRATIVE REGISTER

A nother kind of wall painting, at least as od as the
Early Classic period, before a . d . 300, is the narrative register. It consists o f a banded succession
of scenes, sharing a common ground line. Whether or not the scenes are divided into separate
panels, the register leads the spectator through a
series o f narrative events. Its relationship to the
procession and to the scene is obvious: with processions, the register shares a linear order of fig
ures, and with scenic painting it shares the detailed
characterization o f the setting. In other words,
some registered narratives approach friezes, while
others are scenic, always relative to the degree o f
figural variety developed within the register. The
storytelling register can be either ritual or dramatic, according to its repetidousness or its variety.
The oldest known Maya mural is a narrative reg
ister (fig. 111-55). It carne to light in 1937 at Uaxac
tun on the wall o f a small unvaulted room. T en and
one-half feet long by three feet high, it is composed in two registers. At least twenty-six human
figures appear, painted in five colors on a brownish-pink ground. Their gestures have an animadon
lacking in sculpture o f the same Early Classic peri
od. For instance, the posture o f the dancing dwarf
in the upper register appears in sculpture only
about a century later. There are three scenes: a

standing conference on the left; three seated per

sons inside a house; and some dancers on the right,
with a drum m er and a dozen or m ore spectators.
The figures are in three scales large, mdium,
and small corresponding perhaps to rank.
At Bonampak (fig. 111-64), w here the muris
were painted a century before the end o f the Initial
Series Period, only the central cham ber contains
the large-scene muris discussed above. The end
rooms of the building are decorated with narrative
registers. Room 1 (figs. 111-61,111-62, and 111-64)
shows robing scenes for im portant persons. At the
level of the bench, files o f musicians and atten
dants face away from the door to converge upon
the central personages at the center o f the back
wall. O ver the doorway is a register with seated
servants. O n the rear wall an attendant displays a
child to the white-robed courtiers. Every detail
conveys a sense of preparation and rehearsal. The
whole room concerns the private life o f the court,
revolving about the family o f the prince. Costumes, gestures, and expressions display a refined
ritual o f social behavior, governed by luxurious
tastes and by a rigid code o f manners.
In Room 3 (fig. 111-63) the terraces o f a pyra
midal platform extend over three walls. The pyra
mid has light stages rising between other plat
forms. T he terraced steps o f the pyramid swarm
with winged human figures wearing feather headdresses and big wings suspended at their hips. On
the end walls the scene is watched by a prince upon
his dais, accompanied by family and servants and
by a dwarf carried in a litter by a cluster of costumed men. T he general impression is of music,
dancing, and pageantry, witnessed by important
persons and centering upon a display o f dancers.
O ne whole act is drawing to an end; another will
soon begin, to judge from the acrobats and tumblers approaching from the left o f the end wall.
The question of the order o f the painting of
these three rooms is still an open one. Pros
kouriakoff and Thom pson suggested that Rooms
1, 2, and 3 were painted in that order, but Tozzer
preferred to place Room 2 first, followed by
Rooms 1 and 3. I w rote in 1962 that the register
compositions o f Room 1 ought to be the earliest,
and the scenes later, as works o f different periods
and persons, in the order 1, 3, 2. I have now

Precolum bian Mural Painting

changed my mind, in the light o f these comparisons between muris from different regions,
because there is no proof now possible that the
painters at Bonampak were inventing new pic
torial solutions during the time the rooms were
painted. It is m ore Iikely that different pictorial
systems, corresponding to differences in subject
matter, were employed during a brief time, say a
year, when the rooms were all being painted.
Closely similar to the muris o f Bonampak are
paintings on the walls o f a room at Mul-Chic, near
Uxmal, 8.4 m. x 2.2 m. high, showing a battle with
massacre and capture o f prisoners, a priestly p ro
cession, and preparations for the sacrifice of pris
oners. These scenes merge into one another along
the length o f the south wall, moving from right to
left and continuing on the east wall. The narrative
sequence clearly approaches a register composi
tion of the same class as the end rooms o f Bonam
pak. O n the strength o f resemblances with dated
costume traits o f the Puuc regin, Romn Pia
Chan10 ascribes them to the closing centuries of
the Initial Series Period. H e sees the battle as occurring between traditional Puuc Maya people and
Toltec intruders. T he style o f painting is for him a
Puuc style at the end o f the Initial Series Period,
and therefore not too rem te in time from Bo
nampak. The pictorial system at Mul-Chic reduces
every narrative elem ent to a procession.

ILLUSTRATIONS
The end o f the pictorial road seems generally to be
book illustration, when the picture takes its orders
from a w ritten text. Ancient Maya manuscripts
have solid pages o f text perforated at intervals by
text figures which Ilstrate the meaning o f the
glyphs. Mexican manuscripts o f Precolumbian
date are genealogical, as Codex Bodley, or ritual,
as Codex Borgia. In both cases the pictures follow
apredeterm ined historical or calendrical text, rep
resented by dates w ritten with conventional signs.
None o f these is older than a . d . 1000, and it is
clear that many pictorial conventions are shared
alike by wall painters and manuscript illustrators.
But if we accept this late dating of the Precolum
bian manuscripts, their conventions m ust have
been derived from those o f the muralists.

261

Strangely hybrid wall paintings nearly five feet

high once adorned thirty-five feet of the walls o f a
platform at Santa Rita in British H onduras (fig.
111-65), published by Thomas G ann in 1900.11
The glyph forms are Maya. The architectural profiles recall Puuc Maya comparisons. The figural
style, however, shows Mexican connections, perhaps closer to Mixtee sources than to Toltec ones.
The contorted bodies, the angular costume panels,
the rectilinear units, the high color in seven tones,
and the enrichm ent by oversized tassels, garlands,
feathers, and jewelry all these give a strong suggestion o f non-Maya models perhaps from the
Mixteca in Oaxaca, like those o f the type of Codex
Borgia and the muris of Mitla.
Farther up the east coast at Tulum (fig. 111-66)
are fragments o f frescoed walls on buildings erected during the eleventh or twelfth centuries.
These frescoes resem ble Codex Peresianus (fig.
111-67), one o f the three Precolumbian Maya
manuscripts, especially in the divisin of the wall
by registers and com partm ents like those o f the
pleated, screenfold manuscripts.
Muris o f fifteenth-century date offer an extraordinary dem onstration of the relation between
the manuscripts and the muris. A mural from
Tizatlan near Tlaxcala (fig. 111-68) and the Codex
Borgia manuscript (fig. 111-69) include almost
identical representations o f the god Tezcatlipoca
or his priest. Unfortunately, the remarkable iden
tity gives no clue, internal or external, as to the
direction o f the relation, unless it is that wall paint
ing and pottery painting both took leads at this
time from book art, as we can see in other exam
ples, like the book-sized frescoes on the lintels of
Mitla (fig. 111-70). The latter closely correspond to
Burgoas description o f the way in which screen
fold manuscripts were hung as wall decorations
around the rooms o f the houses before the Conquest. But with these examples we are moving
away from wall paintings as such to derivations
from other arts, which happen to be painted on
walls.
C O N C L U S IO N
I have tried in this paper to dem nstrate the variety of types which coexisted in mural painting at all

262

A ncient Am erica

points of its history during about 1,500 years. The

dated examples forbid us to treat them as a developmental sequence. Instead they are found together at the same m om ent, suggesting in a way
that each was suited to a different kind o f meaning.
Sacrifices and battles, which cali for many figures,
often were shown as scenes in a deep-space perspective. Processions, rituals, and narratives were
more likely to be fitted into banded registers. The
case of Bonampak proves that both m odes existed
simultaneously.
Furtherm ore, the instances of muris based
upon book illustrations show that that problem is
only a special case o f the more general question of
compositional schemes. We did not examine all
possible schemes, but it is easy to see how the
same schemes of composition reappear in other
techniques, such as relief sculpture, book illustration, metalwork, and pottery decoration. Y et if
mural painting was for a long time the theater
where formal innovations and developments first
were devised and presented, then it is to mural
painting rather than to ceramic history that we
should look for the shaping events in the history of
ancient M esoamerican art.
Ironically, the reverse has happened: the ar
chaeological histories of Mexican and Maya ce
ramic types are far m ore detailed and voluminous
than the history o f the wall painting, which probably governed the pictorial decoration of pottery. It
will be interesting in the long run to observe how
the disciplines o f the social sciences, with their
stress upon statistical events, upon pots m ore than
art, will eventually accommodate to the qualitative
fact of the wall paintings studied here.
For the present, the archaeologists have given
us a detailed history o f pottery types, but we still
lack an inclusive history o f ancient American
painting, within which the history o f painted p o t
tery is only one chapter. W hen that history can be
written, if it ever can be w ritten, the history o f the
muris should be among its chief subjects.

NOTES
1 H. Beuchat, M anueldarchologieamricaine: Amriqueprhistorique C h ilisations disparues (Paris, 1912).
2 P. Kelemen, Medieval American Art: A Survey in Two Vol
umes (New Y ork, 1943).
3 S. Toscano, Arte precolombino de Mxico y Amrica Central
(Mxico, 1944), pp. 5 5 2 -7 5 .
4 K. R uppert, J. E. S. Thom pson, and T. Proskouriakoff,
Bonampak. Chiapas, Mxico, Carnegie Institution of
W ashington, Publication 602 (W ashington, 1955). Also
A. Villagra Caleti, Las pinturas de Bonam pak, Cuader
nos Americanos, VI (1947), 15168.
5 I. Bernal, Mxico: Pinturas prehispnicas, Coleccin
U N ESCO de A rte M undial (Paris, 1958).
6 Toscano, Arte precolombino, p. 325: (1) decorativos, (2)
mitolgicos, and (3) histricos o descriptivos. T he latter
includes historical scenes and religious ceremonies.
7 Leopoldo Batres, N ouvelles fouilles Totihuacan, Revue dethnographie, V (1886), 478; Les fouilles operes
T otihuacan, in Congrs international des Amricanistes,
Quebec, 1 9 0 6 ,A cta (Q uebec, 1908), I l,2 7 7 -8 2 ;a n d Teoti
huacn o la ciudad sagrada de los Toltecas (Mxico, 1906).
8 E. Seler, Gesammelte Abhandlungen zur amerikanischen
Sprach- und Altertumskunde (Graz, 1961), V, 411; M.
Gamio, La Poblacin del Valle de Teotihuacan (Mxico,
1922), II; G. C. Vaillant, The Aztecs of Mxico (New York,
1942), pl. 24. Seler interpreted the lateral figures as descending dem ons o f darkness; Gamio as crem ation scenes;
and Vaillant as water-goddesses. All were plausible and
none is com plete.
9 See A. Villagra Caleti, Trabajos realizados en Teotihua
cn, 1952, Anales del Instituto Nacional de Antropologa e
Historia, Mxico, VI (1954), 6 9 -7 8 ; A. Caso, El paraso
terrenal en T eotihuacn, Cuadernos Americanos, I, no. 6
(1942), 12736; and L. Sjourn, Un palacio en la ciudad
de los dioses (Mxico, 1959).
10 R. Pia Chan, Algunas consideraciones sobre las pin
turas de M ul-Chic, Y ucatn, Estudios de cultura Maya, IV
(1964), 7 7 -7 8 .
11 Thomas Gann, M ounds in N o rth ern H onduras, 19th
Annual Report, Bureau of American Ethnology (1897-981,
part 2 (W ashington, 1900), pp. 66592.
[Princeton University Press will soon publish The Muris of
Bonampak, a dissertation at Yale by Mary Ellen Miller in

1981.g k ]

III.6
The Iconography of
the Art of Teotihuacn

The detailed meaning o f the art of Teotihuacn

remains a mystery. There are no texts coeval with
its forms, and no image has been clearly or unequivocally identified in the terms intended by its
maker. Some students, like Laurette Sjourn,
have tried to find a unitary or magic key to interpretation in cultural symbolism ,1 and others,
like Hasso von Winning, have approached the art
by examining motifs for detailed comparative
study.2 This paper will attem pt to define iconographic relationships rather than search for a
single key to unlock all the meanings of Teotihua
cn iconography, and rather than nibble at single
themes. I wish to try an intermediare solution by
attempting to describe the whole configuration
and its internal relationships.
I
To begin, let us question one assumption that usually governs the reading o f Teotihuacn iconogra
phy: that the faithful representation o f biological
species and daily objects was the primary concern
of the artists o f Teotihuacn,3 in spite o f the preponderance in that art o f com pound forms which
have no counterpart in visible reality.
We can proceed by listing those few representations which are integral replicas of natural forms.
The human figure appears most frequently from
Late Tlamimilolpa (IIA III, a . d . 3 0 0 4 0 0 ) on,4
sometimes w ithout costume, as in the Paradise
mural (fig. 111-7 3) o f Tepantitla,5 but m ore com
monly in costume, as priest (fig. III-l 17) or war
rior (figs. 111-83 and 111-84), and often representing or im personating a deity (fig. 111-87), by bear

ing or wearing iconographical attributes such as

butterfly wings, a conch (figs. 111-79 and 111-80),
or an animal helmet. The helm et may be an eagles
beak framing the face or a com pound animal-head
mask (fig. III-9 1)- In pottery figurines these elaborations increase with time, becoming most conspicuous in the M etepec (formerly Amantla)
phase of Teotihuacan IV (fig. III-107). Serpents,
birds, and jaguars are rarely shown naturally; they
are symbolically com pounded. Coyote, dog, frog,
and eagle are present and there are several kinds of
shells, as well as starfish and plant forms. Y et the
delineation of these species rarely shows the natu
ral form alone; it is usually combined with other
delineations o f species whose relationship is symbolic or ritual. Examples are the serpent-birds,
double-bodied jaguars, and feathered jaguar-serpents leaping in the waves of the muris at the
Tem ple of Agriculture (figs. 111-95 and 111-96).
Thus the iconography of Teotihuacan advances in
symbolic complication m ore than in naturalistic
description. Even the delineation of the human
figure is subjected to symbolic com pounding of
this kind.
In addition to at least forty-five images o f life
forms, the repertory at Teotihuacan includes over
fifty glyph-signs (see table III-l on pages 2 6 5 -6 7 )
appearing singly and in combination. M ost of
these signs and images reappear in Veracruz, at
M onte Alban, and at Xochicalco. Some, like the
goggled rain figure and the seashell forms, appear
at Early Classic sites in Guatemala, like Tikal, Kaminaljuyu, and Lake Amatitlan, but their occurrence is not noted in the table. {Kaminaljuyu is
added in this edition e d .] Teotihuacan shares
263

264

A ncient Am erica

some of these forms only with M onte Alban, o th

ers only with Veracruz, and still others only with
Xochicalco. A few signs and figures occur in all
four regions: a jaguar-man (fig. 111-85), an od man
(fig. 111-90), crossed bands or nets (fig. III-l 17).
The year-sign or trapeze-and-ray, a m otif of
interlacing iines common on Teotihuacan headdresses (fig. 111-84), appears throughout M eso
america (Caso 195859; 1962). Among com
pound forms only the animal-helmet, composed
of a jaguar muzzle with serpent fangs and round
bird- or butterfly-eyes (fig. 111-91), recurs at
M onte Alban, Xochicalco, and Chichn Itza.
There are a few signs and images, both single
and compound, that are peculiar to Teotihuacan
alone. Single elements which do not appear elsewhere in Mxico (except in the Maya area) are
quetzal (fig. I I I - l09), water-drops (fig. 111-86),
isolated human eyes (fig. 111-73), and tripled
mountains (fig. III-102). Compound forms include the feathered jaguar (fig. 111-92) and a serpent-bodied owl (fig. I I I - l09).

II
The entire repertory o f pictorial expression at
Teotihuacan supports the view that painters and
sculptors were seeking forms o f logographic clarity and simplicity. They were less interested in
recording appearances than in combining and
compounding associative meanings in a quest for
viable forms o f writing. H enee I shall stress the
narrow range o f pictorial motifs at Teotihuacan,
the rarity of integral organic representations
(about one in four), and the numerical im por
tance of compound signs and symbols.
N o one has yet attem pted a general soludon by
constructing a linguistic model that would corre
spond m ore or less accurately to all the systematic
relationships without seeking detailed interpretations of any single form. The present linguistic
approach pretends to no high degree of accuracy
but only to a probabilistic assessment of whether
or not the various kinds o f reoresentations can be
related to one another.
The linguistic model requires that each form be
examined for its grammatical function, w hether
noun, adjective, or verb. It appears from this study

that among some one hundred signs and images

(see Table III-1) at Teotihuacan, a majority is used
as nominal expressions to describe beings, substances, and concepts. N ext most numerous are
adjectival uses describing qualities and rank. Least
common are verbal statements relating to operations and actions, as when a jaguar-headed man
approaches a tem ple upon a roadway marked with
footprints (fig. 111-85) or when a spectacled raingod warrior appears at a dancing platform indicated by scattered footprints (fig. 111-83). Such
distribution o f the parts o f speech corresponds
less to any narrative or exposition than to litany
and liturgy, where the ames o f the deity and the
worshipers petitions are pronounced with an
abundance o f nominal and adjectival forms and
few verbal ones.
A litany consists o f a recital o f the ames and
qualities of the deity being worshiped, together
with the favors requested from it. The liturgy is an
order o f Service, specifying the sequence of the
parts. To each deity special forms o f address are
used, and special favors are asked of it in return for
properly perform ed rites and sacrifices. The litany
is grammatically poor but it is rich in metaphors
and titles.
At Teotihuacan the mural scenes are arranged in
connecting rooms as at Zacuala or on ascending
terraces (Sejoune 1959). This ordering strongly
suggests liturgical sequence. W ithin each mural
composition a principal them e or figure is evident,
enriched by associated figures and by meaningful
frames suggesting a recital o f the powers o f the
deity, together with petitions to be granted by the
god. We can assume that the images of Teotihua
can desgnate complex liturgical comparisons,
where powers, forces, and presences are evoked in
metaphors or images.
Several verbs are clearly shown, like praying,
dancing (fig. 111-83), traveling (fig. II1-85), and
sacrificing (fig. 111-54). Any further interpretation
depends upon how we read the nouns and adjectives, and how we assess their importance or rank.
The word-pictures can be divided into simple and
compound forms, as well as frontal and profile
images. Their composition obeys rhythms of alter
naro n and of fourfold and fivefold divisin. The
context or level of discourse is indicated by frames

Table UI-1*
M onte
Alban
H uman
1. Raingod, goggles only

2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.

Raingod, as warrior
Water female
Od male
Fat male
Shrouded head (Xipe?)
Cleft head
Skeletal figure
Club-footed figure
Bearded male
Warrior or hunter
Female containing figurines
Ballplayers
Swimmers
[Hunchback]
[H eads]
Other Life
15. Serpent
16. Jaguar
17. Eagle
18. Quetzal
19. D og or coyote
20. Owl
21. Piste
22. Vulture
23. Butterfly
24. Starfish
25. Squid
26. Conch
27. Pecten
28. Flower
29- Biznaga
30. M onkey
31. Maguey
32. Frog
33. Turtle
34. Corn
[Cacao pods]
[Cactus fruit (heart?)]
Compound Life Figures
35. Raingod jaguar
36. Raingod reptile

Classic
Veracruz

Xochicalco

Others

KJ =
Kaminaljuyu]

[Occurs]

[M iccaotli]

[M iccaotli]
X
X
X

X
X

X
X

[Early Xolalpan]
[M etepec]
[Early Xolalpan]
[KJ]

[Late Xolalpan]

[Late Tlamimilolpa]
X

[Late Tlamimilolpa]
[Early Xolalpan]
[Late Xolalpan]

X
X

[M iccaotli]
[M etepec (birds)]
[M etepec (birds)]
[KJ]
[KJ]

Kaminaljuyu

[Late Xolalpan]

X
Copan [KJ]
[KJ]
[KJ]

[Late Tlamimilolpa]
[Tezoyuca ?]
[Late Xolalpan]

X
[KJ]
[Early Xolalpan]
X

X
X
(continued)

265

Table II1-1* (Continued)

M onte
Alban

37. Raingodvegetation
38. Raingodstarfish
39. Feathered serpent
40. Feathered jaguar
41. Eagle-headed serpent
42. Coyote-headed serpent
43. Serpent-bodied owleagle
44. Jaguar-serpent-bird
45. Jaguar (muzzle)-serpent (eye)
[Tlaloc effigy]
Signs
46. Stepped fret
47. Crossed bands
48. Quincunx
49. Cross
50. Diamond
51. Trilob[al drop]
52. Treble scroll
53- Scroll (Tajn)
54. Trapeze-ray [folded triangle]
55. Saw-tooth ray
56. Hanging drops [tres gotas de
sangre]
57. Water-eye, elongated
58. Eye, round
59- Half-star
60. Teeth
61. Four-way hatching
62. Water-tongue
63. Footprints
64. Mountain, pointed
65. Mountain, rounded
66. Comb-and-bar
67. Curl RE
68. Mouth RE
69. Speech scroll
70. Interlocking scroll
71. Temple
72. Nose-pendant butterly
73. Nose-pendant raingod
74. Cobweb
75. Tri-shingle
76. Aspergillum
77. Crossroads
78. Feathers

Classic
Veracruz

X ochicalco

Others

[Occurs]

X
X

[Early Xolalpan]
Chichen Itza

[M etepec]
[Late Tzacualli]

X
X
X

[Maya]
Maya
Maya

[Patlachique]

[M etepec]

Tula,
Chichen
[KJ]

Maya

[Early Tlamimilolpa]

[KJ]

[Late Xolalpan]

X
X
X
X
X
X

X
X
X
X

[Late Tlamimilolpa]
[Late Tlamimilolpa]

X
X
X
p

[KJ]

X
X
X

[KJ]

X
X

X
X

[Tezoyuca?]
[Early Xolalpan]
[Early Xolalpan]
[Early Xolalpan]
[M iccaotli]
[Early Xolalpan]

[Early Xolalpan]
X

79. Flam es

266

[Early Tlamimilolpa]

T he Iconography o f the Art o f Teotihuacn

267

Table III-l* (Continued)

Monte
Albn
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.

Arrows
Shield
Knife blade
Chevron band
Brazier
Diadem
Labret
Ear-plug
Armadillo segments
Eggshells
Hand, isolated
Diagonal scallops
Asterisks
Paired brackets
Four-element group
Bar-dot numeral
Talud-tablero ornament
Spear-thrower (atlatl)

Classic
Veracruz

Xochicalco

Others

[Occurs]
[Early Xolalpan]
[Early Xolalpan]
[Early Xolalpan]

[net]
[beads]
[waves]
[braids]
[discs in bands]
[crenelations]
[masks]
[mantas]

[Early Xolalpan]

[Early Xolalpan]

Maya
Chichn
Tula,
Chichn

[Early Xolalpan]

[Patlachique]
[Miccaotli]
[Miccaotli]
[Late Tlamimilolpa]
[Late Tlamimilolpa]
[Early Xolalpan]
[Early Xolalpan]
[Early Xolalpan]

*The tabulations o f them es and their correspondences at other sites are based in part upon seminar reports (at Yale in 1965) by
Cynthia T im brook, M anuela Jem m a, Thom as Reese, Anthony Fehm, C atherine W ilkinson, David Summers, and G abriella
Yablonsky; (and at Harvard in 1966) by A rthur Miller, Clemency Coggins, Kate Spencer, Nicholas H ellm uth, Bruce MacDonald, Stephen T obriner, David van Z anten, and Diana H ubbard. T he author is responsible for the present nom enclature,
classing, and interpretation o f the table.
[ Additions made by G eorge K ubler in his copy o f the book are placed in brackets. They expand the list o f traits, add a fuller tally
of motifs that occur at Kaminaljuyu, and identify periods in which the motifs occur according to Florencia M llers now
somewhat dated chronology. See Florencia Mller, Secuencia ceramica de T eotihuacn, in Teotihuacn: X I Mesa Redonda
{1966}: Sociedad Mexicana de Antropologa (Mxico, 1966), I, 3144 ED.]

and borders. The entire iconographic system, fnally, resolves into five or m ore m ajor clusters of
motifs, probably corresponding to different cults.
Compositional modes may be described under
seven headings:

1. Noun and Adjective

Many forms occupy both grammatical positions.
The nominal usage occurs when substantive prop-

erties, such as jaguar-muzzle, serpent-fang, and

birds eye6 are com pounded into one cult image
(fig. 111-91). If this form, however, is worn as a
headdress, it characterizes and qualifies the wearer
as a devotee or celebrant (fig. III-l 17), and it can
be classed as adjectival. Many forms appear in
both contexts, and their rank differs according to
w hether they are principal or accessory objects.
To represent the elem ents, such as fire (Se-

268

A ncient America

journe 1964) or w ater (von W inning 1947b;

Rands 1955), there are both nominal and adjectival forms. A goggled human head is usually nom
inal, having to do with the raingod (fig. 111-72), but
a band containing eyes and meaning water is nor
mally adjectival (fig. 111-73). A brazier on the head
o f an od man is nominal for an od god
(fig. 111-90), but a strip or band with the diamond
or rhomboid shapes o f the brazier on it is usually
adjectival, as on the eye-band o f figure 111-75.
Certain forms, such as feather borders (figs.
111-89 and 111-99), fringes o f fame (fig. III-109),
radiant saw-tooth frames, and bands o f starfish or
shells (fig. III-117), are probably always adjectival,
with feathers suggesting valu, flames and rays
connoting splendor, and waves, starfish, and shells
evoking the sea.

111-95 and 111-97). Simple figures are exceptional,

being rare and occupying secondary or subord
nate positions. Com pounds are far more common
and far m ore prom inent. Simple figures usually
appear in accessory positions such as the edges of
speech scrolls (fig. 111-87), or the streams issuing
from charismatic hands (fig. 111-93). Thus com
pounds have m ore rank than simple figures, being
the carriers o f more meanings within the same outline and relating m ore closely to the supernatural
than to the sacrifices or petitions indicated by sim
ple forms.
4. Frontal and Profile

The forms are principal or accessory in rank, and

the same form may occupy different positions of
rank. In general, nominal usages are o f principal
rank, with several grades of importance. A djec
tival forms are accessory, and they appear as
needed in various clusters o f forms. Thus cult objects (fig. III-117) have higher rank than figures of
celebrants. But costumes, frames, and borders are
accessory or adjectival.
The reptiles eye (RE) glyph (von W inning
1961) in m outh and curl variants7 always seems
nominal, but it differs in rank, being set upon a
feather shield (fig. III-108) in one context, or resembling a m outh in a facial context (fig. III-102),
and even being stamped upon the forehead o f a
figurine in another context (fig. III-103). It always
seems to label the shape with which it is included,
as when it appears upon molded plaques called
adornos, which decrate cylinder vessels and braziers. The curl (RE) glyph is very rare in mural
painting, although it appears at Zacuala, where it
labels a butterfly with long legs (fig. 111-94), and at
Tepantitla, where it occurs as a cartouche on a
painted wall.

Among painted and incised representations,

frontal figures (fig. 111-74) have m ore rank than
profile figures. The evidence for this is that frontal
figures are larger and more elabrate than profile
figures in general. In addition, frontal figures are
frequently flanked by profile figures. Frontal rep
resentations probably describe cult images. The
main examples are goggled rain figures (fig. III72), female donor figures (fig. 111-74), butterflies
(fig. III-115), compounds of jaguar-serpent-bird
(fig. 111-99), shrouded (Xipe) heads (fig. 111-104),
and curl (RE) glyph faces (fig. III-102). Isolated
diadems are also frontal (fig. 111-86). Profile humans are usually shown as celebrants. Profile animals, such as serpents, jaguars, eagles, and various
compounds, normally appear as accessory figures.
An apparent exception is the frontal human
shown in the upper right quadrant of the mural at
Tepantitla (fig. II1-76), which resembles the clay
figurines of dancing humans in Period III. The
frontal figure at Tepantitla is distinctly subord
nate and vernacular. Its type relates to the realist
fragments found at Tetitla (Wicke 1956; Villagra
Caleti 1954, pp. 6 1 -7 2 , pl. I).8
The hypothesis that many frontal figures are
cult images or supernatural beings allows us to
suppose that the same figures, when shown in profile (fig. III-7 1 and III-l 17), represent human cel
ebrants, or priests, or impersonators.

3. Simple and Compound

5. Alternation and Polymorphism

Among nominal forms, simple figures show one

species only within one contour; compound fig
ures conjoin the traits o f different species (figs.

Nominal forms are often shown in dual alternating

repetition or in fourfold and fivefold aspects. The
repetition o f dual alternating forms appears com-

2. R an k

T h e Iconography o f the Art o f T eotihuacn

monly on cylindrical vessel walls and in mural
frames. An example is the talus frame o f alternaring diadems and starfish at the Palacio de los Jag
uares (fig. 111-86). The alternation may corre
spond to many meanings. The main possibilities
are that one form precedes, implies, causes, or
governs the other, if the forms are meant to be
understood in some successive relation. If, however, a simultaneous relation is meant, the forms may
be contrasting or similar, extensive or inclusive in
relationship.
Spectacled raingod and flayed-god frontal fig
ures alternate on the same vessel (fig. III-104) in a
possible opposition o f wet and dry, but the alter
nation may also signify acom plem entarity o f plant
and animal life. The com m onest alternations dis
play curl glyphs with mountains, or with butterflies, or flowers, or trapeze-rays (fig. III-101).
Flowers and butterflies also sometimes alternate.
Many o f these alternations have lower borders of
lightning or fame signs; such alternations belong
together with the brazier assemblies discussed
below.
The curl (RE) glyph has attracted more discussion than any other sign in the lexicn.9 The term
reptiles eye is B eyers, but the earliest in
terpretation o f it as an eye was made by Seler. Caso
read it as a calendrical sign for Quetzalcoatl at
Xochicalco(fig. III-100). Von W inning believed it
to stand for fertility and abundance. I read it even
m ore specifically as an earth sign because it usuallv
appears under bands o f lightning and rain, in the
base position on the cartouche. O n braziers it is
repeated in four different colors to display directional character (fig. 111-98), as one might expect
o f an earth sign. In adornos resembling human
faces it always occupies the position o f the m outh
(fig. III-102). Indeed, K idder saw it as related to
speech and prayer.10 These readings fertility
(von Winning), m outh images (Kidder), and earth
symbolism are only different modulations o f the
same theme. N one o f these three precludes the
others.
A molded orangeware bowl excavated by Linn
at Las Colinas (Linn 1942, p. 68) is the outstanding example (fig. III-l 13) o f a fourfold o r fivefold
divisin. Sherds from the same mold appear at the
Yayahuala suburb during Teotihuacn III (S

269

journ 1963, pp. 43 ff., figs 8 -1 0 ). As a super

natural, spectacled figure on the base, the raingod
presides over four separate aspects, which embody both nominal and adjectival forms, as well as
principal and accessory representations. Four emblems or cult images (a serpent, a quetzal, a rain
diadem, and a coyote) are attended by priests with
copal bags, who pour libations and give voice. At
the rim hang various drops, and the priests tread
upon water lilies like those in the Tem ple o f Agriculture muris. Linn suggested that they symbolize the different kinds of water, as recorded by
Sahagun, and that they refer to the four cardinal
directions. The circular order and the emblems
themselves strongly indcate a directional sense on
visual evidence alone, w ithout help from texts
written a millennium later. The Las Colinas bowl
probably summarizes and brings together several
cults of which the muris give us only fragments of
scattered evidence.

6. Frames and Borders

The frames surrounding mural panels (figs. 111-82
and 111-86) resem ble the borders on costume ele
ments and on headdresses (fig. III-l 17) because
both contain adjectival forms which qualify what is
inside them. The panel frame establishes the kind
of space that is represented, and the costume
border establishes the rank and the attributes o f its
wearer, in much the same manner as the eagle(fig. 111-82), coyote- (fig. 111-87), or jaguar- (fig.
III-l 17) helmets do.
Twining serpent bodies bearing different markings and sometimes term inating in jaguar, bird, or
coyote heads (fig. 111-79) are a common kind of
mural frame. The serpent bodies may signify that
the scene is m eant to represent supernatural
events in a celestial dimensin. The different com
pound animal forms may relate to world quarters
or to eras o f mythological time. Analogous frames
suggesting celestial space appear in the sky-signs
crowning the muris at Bonampak, in the tomb
muris o f M onte Alban, and in the scrolls sur
rounding the ball-court reliefs at El Tajn.
At the Palacio de los Jaguares, there are profile
figures o f jaguars blowing conchs (fig. 111-86) rep
resented on the four walls. T he scene may allude
to a liturgical prayer for water in a directional con-

A ncient Am erica

270

text. The surrounding frame contains an interlace

of flowered serpents, as well as diadems alternating with raingod-starfish. These adjectives proba
bly enrich and specify the main them e with clauses
or phrases o f liturgical character, qualifying place
and time. O th er walls are reticulated by interlacing
bands (fig. 111-78), which surround many replicas
of the same profile human and specify the liturgy
or ritual that is represented. Simple nets may refer
to fishing and bird-snaring; serpent-nets may have
a metaphorical meaning o f divine or supernatural
order.
In costumes and headdresses, the borders often
have adjectival ornam ents o f which the meaning is
fairly clear. Feathered ruffs and edges (fig. 111-81)
suggest precious quality.
Saw-tooth rays
(fig. 111-85) and eyes (fig. 111-73) represent brilliance. Borders o f fame (fig. III-109) relate at
least to fire. Half-stars, interpreted as sectional
conch shells, connote water, perhaps specifying
the saltwater o f the oceans.
7. Ico n o g ra p h ic C lusters
The art o f Teotihuacan as we know it today con
tains few themes. These, grouped as humans, ani
mals, signs, and com pounds, total only about one
hundred distinct forms. Among them the outlines
of several clusters or groupings can easily be discerned. They are the subject o f part III.

III
A. The raingod cluster is the most common, with
five or six variants in the representation of the
deity, under repule (Fig. 111-73), jaguar (fig.
III-104), starfish (fig. 111-86), flower (fig. 111-73),
and warrior aspects (fig. 111-84). Associated with
goggled (Tlaloc) figures are feathered serpent (fig.
111-82), quetzal, butterfly (fig. III-l 14), jaguar,
and dog or coyote forms. These representations
are common in indoor muris, w here the same
combination will recur identically in many repetitions on the walls o f one room (fig. 111-83), as if it
were a special prayer for beneficent water.
On pottery vessels (fig. I I I - l04), abbreviated
forms of the raingod complex are common, without cise replication. Variety in the clustering may

have been deliberately sought, within permitted

limits.
B. A butterfly complex is common on braziers
(fig. 111-98) and cylindrical tripod vessels (fig.
I I I - l06). The curl (RE) glyph, flower forms,
mountains, and raingod heads are usually associated (figs. III-101, I I I - l08, III-l 14, and III-l 16).
The brazier form suggests that this cluster relates
to funeral rites, and that the butterfly may be an
image for the soul. The curl (RE) glyph has an
accessory, adjectival role on the braziers (fig.
111-98), w here it appears on thin molded terracotta
plaques called mantas, often painted in different
colors suggesting directional symbolism. O n cylinder tripods, the funerary symbolism of the bra
ziers is abbreviated and reduced to alternadons of
butterfly with curl glyph or butterfly with flower
(fig. III-l 14). The mountain shapes11 and crossroads forms in this cluster may function as place
ames (fig. III-l 16). In any case, there are very
few such place signs. It is reasonable to think that
some refer to an afterlife and to the residence of
souls, as Caso has supposed for the mural of Tepantitla (fig. 111-73), which he calis the mural of
Terrestrial Paradise (Caso 1942).
C. A third m ajor complex is smaller in scope,
having fewer associated forms than the raingod or
the butterfly clusters. The figure o f an owl predominates (fig. III-l 12), with arrows and a shield
as emblems of warlike meaning (von Winning
1948). At the Quetzalpapalotl courtyard (Acosta
1964) frontal representations o f an owl with skeletal features occupy the west side.12 Facing them
on the three remaining sides there are figures of
profile quetzal birds which are clearly accessory
(fig. III-109). T he quetzal piers have flames on
their bases and at the heads, unlike the owl piers,
which have green feather bands. Owl and quetzal
surely have connected meanings, probably pertaining to war and a named dynasty. Elsewhere, as
on M etepec or Phase IV figurines, quetzal is part
of the elabrate costume o f personages who are
possibly historical, such as members of a lineage
marked by the quetzal em blem (fig. III-107). At
the Quetzalpapalotl court, this lineage and its devotion to a warrior cult symbolized by the owl are
probably the subject o f these reliefs.13 In another
context at Tetitla, the owl is allied to rain (figs.

T he Iconography o f the Art o f T eotihuacan

III-110 and I I I - l l l ) , in a com pound which may
relate war and relief from drought: if nature cannot provide, aggression will, under the raingods
aegis.
D.
A compositional scheme common at T eoti
huacan shows a cult object flanked by worshipers
or impersonators. An example is the radiant circle
containing a net, m ounted upon a pedestal, with
priests on both sides, in the mural o f the Casa de
los Barrios at Teopancaxco (fig. III-117). The
priests wear feathered jaguar headdresses and carry incense bags. Flowered scrolls of song rise from
their lips, and libations pour forth from their
hands.14 Surely the radiant rim signifies brilliance,
while the netted center, with its jar-shaped interlace sign may signify a cult, to which other ref
erences appear in the netted figures o f jaguarheaded men and profile jaguars in the muris of
Tetitla (fig. 111-85) and Atetelco. Thus both the
disc and the priests have strong jaguar associations.15
This symmetrical scheme is inverted in the vanished mural o f the Tem ple o f Agriculture (fig.
111-54), w here two cult images flank a group of
worshipers whose offerings are addressed to the
sides rather than the center o f the scene.16 Further
variants upon the them e of the worship rendered
to an image appear in several rooms at Zacuala,
where priests are shown as in a litany, performing
the rites suitable to the room. Thus a goggled fig
ure bearing a corn plant im personates the rain deity (fig. 111-71) in several replicas on the walls of
one room. His copal bag has the shape o f the serpent-tail rattle, and he bears ears o f corn on his
back. N o cult image survives, but we may suppose
that a stone or wood figure occupied the room.
At Tepantitla the paradise mural is capped by
such a scene, showing the frontal figure o f a rain
image, with w ater pouring from its hands, flanked
by priests and standing upon scrolled waves filled
with sea creatures and shells (figs. II1-74 and
III-75). Possibly the com pound frontal figure,
showing a goggled face with bifid serpent tongue,
framed by a jaguar muzzle surm ounted by an owl
diadem over diamond-shaped eyes, is another
such cult image, com pounding the attributes of
water, air, earth, and fire. T he facial glyph of figure
III-102 may be an abbreviation o f this image.

271

E. The foreheads o f the jointed figurines studied by von W inning (1958, pp. 78, figs. 2 7 -3 3 )
bear signs, stamped from a mold, called the fourelem ent group, standing for rain and ground
water, lightning, and fire (fig. III-105). These figu
rines, branded or stamped as devotees, probably
were related to a cult image. It may have repre
sented these paired elements o f water and fire by
rain and butterfly forms in com pound or dual fig
ures. A nother reflection o f such a cult appears in
the cylindrical tripod vessels decorated with but
terfly or curl (RE) glyphs alternating with rain
symbols (fig. III-108). The braziers (fig. 111-98),
with their butterfly and fire symbols, also may reflect this cult.

IV
This model o f the iconography of Teotihuacan
may not be reliable in all its details, but its main
outlines as here presented are consistent with the
visual evidence o f the pottery, muris, and sculp
ture. As to the forms which Teotihuacan shares
with other regions (see table), we m ust beware of
disjunctive situations w here form and meaning
separate and rejoin in different combinations.
This caution runs counter to the practice, standard
since before the time o f Eduard Seler, o f assuming
that similar forms in different periods and places
of Mesoamerica must carry similar meaning.
There has been an assumption that strong continuities connect Teotihuacan with Aztec art and
with early colonial records,17 despite an eighthundred-year interval between the abandonm ent
o f Teotihuacan and the discovery o f America. The
easiest and most seductive historical patterns are
those which assume simple continuity o f happening. U pon reflection, however, it is apparent that
only biological and ecological occurrences are
continuous, while events in the domain of sym
bolic experience show a much greater instability
and are m ore susceptible to transformation. It is
axiomatic that history is m ore discontinuous than
biology. It is also self-evident that long historical
periods are less continuous than short ones.
Such historical discontinuities are measured by
the separation of form from meaning, as Erwin
Panofsky dem onstrated in 1960. In a review of his
book I paraphrased its message as follows:

272

A ncient Am erica

Having left Antiquity unburied, the Middle Ages

alternately galvanized and exorcised its corpse in
the Carolingian renovatio, in the Christocentric
spirit of Ottonian art, in the renascence of the
twelfth century and the reaction after the Black
Death. Finally, the Renaissance stood weeping at
its grave and tried to resurrect its soul. And in one
fatally auspicious moment it succeeded with the
mutational changes carried both in Italy and in
northern Europe by the men born about 1400.18
Panofsky was writing about what he called the
law o f disjunction, which pertains to the separation o f form and significance in the art o f the high
and later M iddle Ages:
Wherever in the high and later Middle Ages a
work of art borrows its form from a classical
model, this form is almost invariably invested with
a non-classical, normally Christian significance;
wherever in the high and later Middle Ages a work
of art borrows its theme from classical poetry, legend, history, or mythology, this theme is quite in
variably presented in a non-classical, normally contemporary form.19
In other words, with successive cultures spanning
a duration on the order o f magnitude o f about one
thousand years in the same regin, such as w estern
Europe or M esoamerica, we may expect to ob
serve disjunctions o f form and meaning m ore
often than marked continuity in their association.
As O rpheus and the G ood Shepherd displayed
different meanings in similar representations, so
may we expect representations o f the feathered
serpent and Quetzalcatl to display meanings at
least as different across m ore than one thousand
years in M xico.20 It is for this reason that I have
avoided the use o f Aztec or N huatl ames, preferring rain figure to Tlaloc, fire figure to
H uehuetotl, and flayed figure to X ipe Totee.
Mesoamerican cultures contem porary with T e
otihuacn contain forms similar to those o f T eoti
huacn, but they appear in greatly different contexts. Until each o f these other systems is put
through the same sort of configuradonal analysis
we have conducted with Teotihuacn, we are
blinded to their proper connection. Beside the
pictorial language o f Teotihuacn, Maya vase
painting and Classic Veracruz sculpture seem notably more secular, lacking the reverent m onotony
of the litanies and prayers o f Teotihuacn. The

relation to M onte Alban is obviously cise, but the

direction o f influence is still undeterm ined. At
M onte Alban, the public sculpture and the tomb
furnishings differ greatly. The tom b pottery and
the muris closely resem ble those o f Teotihuacn,
to the point o f sharing many forms, but there is
nothing at Teotihuacn even distantly like the
danzante reliefs or the conquest stelae or the marriage slabs. These them es are closer to Maya dynastic art than they are to those o f the Valley of
Mxico.
The art o f Teotihuacn can be distinguished
from other regional styles o f ancient America by
its strongly marked liturgical character. The gen
eral form o f a painted communication in the style
o f Teotihuacn displays the liturgical actions that
embody the cult, which was itself represented by
appropriate images. The liturgy may be portrayed
in specific actions by the celebrants, as in many
mural paintings, or it may be evoked by symbolic
figures and signs, as on decorated pottery. There is
no reason to believe that different sectors of
meaning are figured in muris and on pottery,
however much m ore compact and stenographic
the signs on pottery may be.
A t Teotihuacn, every mural o r decorated vessel is a prayer exalting the elem ents o f nature, unlike El Tajn, w here the pictorial system centers
upon the ballgame, or M onte Albn, where public
sculpture com m em orates conquests and great captains, as do Late Classic Maya art and Post-Classic
art throughout Mesoamerica. T he real affinity of
the configuration at Teotihuacn is with Early
Classic Maya art in its liturgical aspeets, as at Tikal
or Kaminaljuyu.21 In its tendeney toward com
pound figures, Teotihuacn iconography resembles that of Olmec art m ore closely than any Late
Classic expression.
Teotihuacn is perhaps the terminal expression
of an od theocratic system represented also by
late Olmec art and by Early Classic Maya art. The
few traits it shares with Late Classic Maya dynastic
art belong mainly to Teotihuacn IV. Personal
glorification, the cult o f war, and the appearance of
dynastic lineages are the traits o f the new, emerging society o f the centuries after a . d . 500. The late
Quetzalpapalotl court reliefs (fig. III-109) and the
final M etepec-stage figurines (figs. III-107 and
III-l 12) are alone at Teotihuacn in portraying

T h e Iconography o f the Art o f T eotihuacn

this new dynastic society. All earlier Teotihuacn
representations reflect an aging theocratic system
organized on the basis o f animistic worship, which
probably survived only in the funeral customs and
imagery o f later periods after the collapse and
abandonm ent o f Teotihuacn.

GoodmanMartnezThompson
Correlation

GoodmanMartnezThompson
Correlation

Phase ames

700

Metepec

600

Xolalpan

500
400

Phase Numbers

Late
Early
Late

Teotihuacn IV
Teotihuacn IIIA
Teotihuacn III

Teotihuacn 11A

1
r
\ Classic
Teotihuacn IIA-III l Period

Tlamimilolpa
300

Early

Ktontinued)

Phase Numbers

Miccaotli
Late

Teotihuacn 11
Teotihuacn IA

Tzacualli
A.D.
B.C.

500

1 An example o f Laurette S journs unitary prem ise and

conclusin is the following interpretation: T he City o f
the G ods constitutes a hymn to the glory o f the advent o f
Man, that creature capable o f defying the limits o f nature
by his self-conversion into luminous energy (1965, p.
156). S journs writings on Teotihuacn reflect the European m ovem ent called Strukturanalyse in classical ar
chaeology, which is devoted to discussing the central psychic m etaphors o f cultural history, represented in Europe
by Friedrich Matz and G uido von Kaschnitz-W einberg,
and in America by the writings o f Mircea Eliade. Examples
o f Sjourns psycho-historical m ethod are El universo de
Quetzalcatl, Burning Water: Thought and Religin in An
cient Mxico, and Los sacrificios humanos: religin o po
ltica. H er m ore recent works, Arqueologa de Teotihuacn:
la cermica and El lenguaje de las formas en Teotihuacn,
contain useful drawings and photographs o f vessels and
figurines respectiveiy.
2 Von W innings studies (1947b, 1958, 1961) are essential
to any examination o f Teotihuacn iconography.
3 Sjourn 1965. In this article Sjourn seeks to show that
the owl reliefs on the west range o f the Q uetzalpapalotl
court are really frontal representations o f quetzal birds.
H er thesis has been challenged by A rthur Miller in a paper
to appear in the volum e covering the X I Mesa Redonda
meetings held in Mxico City during August 1966 [M iller
1966],
4 T he chronological table below, excerpted from that presented by the Instituto N acional de A ntropologa e H isto
ria at the X I Mesa redonda, is followed here throughout. It
brings into concordance the two systems o f phase ames
and phase num bers in current use. Paddock (1966, p. vi)
gives a developm ental table showing the elaboration o f
the chronology for Teotihuacn studies since Vaillants
work (1938) and modifying the system now in use by the
Instituto.

Phase ames

200
100

100

NOTES

273

Early

Teotihuacn 1

Patlachique (Chimalhuacan)

Late
1
\ Pre-

Classic
Period

Pruto-Teotihuacn 1

5 Caso (1942) has interpreted th e en tire scene in relation to

sixteenth-century texts (such as Sahagns ethnographical
encydopedia o f native Mxico) as an Aztec Tlalocan.
6 Sjourn (1966b) has observed the cise resemblance of
this jaguar-serpent-bird com pound with the Venus im
ages in the relief sculpture o f the Toltec Maya period at
Chichn Itz, in support o f her argum ent that Q uetzalc
atl was the principal deity and cult at Teotihuacn.
Q uetzalcatl as the planet Venus, attested by Nhuatl
sources only, is her textual link.
7 H ere I distinguish the m outh form from the curl form, as
in figs. III-101 and III-102.
8 T hese fragments resem ble Maya mural painting in delineation and in costume.
9 T he best discussion o f the literature is von W inning 1961.
10 K idder et al. (1946, pp. 2 2 1 -2 2 ) com pared the pairs of
small protuberances within the cartouche to those o f the
speech scrolls.
11 These images were studied in my seminar at H arvard Uni
versity in 1966 and at Teotihuacn by Stephen Tobriner,
as well as in another section o f the same seminar at Yale
University in 1965 by David Summers, who presented his
work at the Frick Symposium in 1966. T o briners paper
has been subm itted for publication in the X I Mesa Redon
da [T obriner 1972}.
12 These figures were studied at the Harvard seminar in
1966 by A rthur Miller, whose paper was subm itted for
publication in the X I Mesa Redonda [M iller 1966].
13 This supposition still requires p ro o f but it conforms with
recent discoveries o f com parable ritual, historical, and genealogical material in Mixtee manuscripts (Caso 1949)
and in Maya inscriptions (Prokouriakoff 1960, 1963;
Berlin 1959; Kelley 1962a, 1965).
14 Seler (1915, p. 417, pl. X I) thought the interlaced threads
represented the m oon rather than the sun, as proposed by
Peafiel (1900, p. 82). Caso (195859, p. 58) proposed it
as an em blem o f the day-count.
15 Jaguar figures are presumably related to an earth cult both
in older O lm ec and in later Toltec and Aztec worship.
Sjourn (1960, p. 68), however, prefers to relate them to
the Toltec and Aztec w arrior societies as T iger Knights,
contrasted with Eagle Knights, citing the sixteenth-century historian o f Tlaxcala, D iego M uoz Camargo.
16 Beyer (in G amio 1922, p. 286) regards the lateral forms as
altars bearing burning material.
17 This point o f view em erged clearly as the official Mexican

274

A ncient Am erica

position at the X I Mesa Redonda, with spirited defenses by

Caso and Jim nez M oreno. In effect it has been the working prem ise o f Mesoamerican studies since the tim e of
Seler that sixteenth-century sources clarify the m ost rem ote historical events. Seler (1915, pp. 4 0 5 -5 8 5 ) discussed the whole iconographical configuraron in Aztec
terms. T he same m ethod was applied again by Armillas
thirty years later (1945, pp. 3541). Armillas identified
eight aspects o f the Aztec raingod (Tlaloc), as jaguar,
feathered serpent, owl, quetzal, butterfly, flower, radiant
eye, and trapeze-ray; a firegod (H uehuetotl); a vegeta
tion god (Xipe); and an unidentified fat god. Caso (1958
59) later attem pted to prove that Aztec calendrical observances were prefigured at Teotihuacn by a few signs for
num bered years and days. Sjourn (1956, 1958, 1962)
sought to show that Aztec human sacrifice was a corrupt
degradation for political ends o f the spiritual transcendence cultivated by their forebears at Teotihuacn. Thus
she rem ained, even with this argum ent, inside the orthodox habit o f explaining Teotihuacn by the use o f Aztec
beliefs. At the X I Mesa Redonda, Jim nez M oreno sought
to prove that the presence o f coyote images in the art o f
Teotihuacn was evidence that the inhabitants spoke
Nhuatl, since the coyote is not indigenous to the valley

and its representation would presumably have accompanied N ahua immigrants from the north. But by the
same token, quetzal, which is not a valley bird, could be
taken as p ro o f o f foreign speech am ong early dwellers in
the valley.
18 K ubler 1961, p. 34.
19 Panofsky 1960, p. 84.
20 Examples: M onum ent 19, La V enta, supposedly prior to
400 B.C ., and the serpent columns o f T ula (Hidalgo), ca.

a .d . 1 2 0 0 .
21 W. Coe 1965; K idder.Jennings, and Shook 1946. During
the Harvard seminar in 1966, Nicholas H ellm uth (67
Harvard CoIIege) prepared a detailed report assessing the
presum ed Teotihuacn presence at Kaminaljuyu and
Tikal respectively. H e found it lacking in Kaminaljuyu
sculpture, but he concluded that the art o f Tikal on the
whole was less affected by Teotihuacn than was the high
land site.
[Hasso von W innings work on Teotihuacn iconography is in
press and it will bring together all that is now known. The
book will appear at the Instituto de Investigaciones Estticas,
Universidad Nacional A utnom a de Mxico, this year or
next. g k J

III.7
Iconographic Aspects of
Architecturai Profiles at Teotihuacn
and in Mesoamerica

The iconography o f architecture has been under

intense study for several decades both here and in
Europe (Lehmann 1945; Bandmann 1951; Smith
1956). T he methods o f that study can also be made
to yield useful results when they are applied to the
architecture of ancient Mesoamerica. It is now apparent that no building is w ithout some conventional meaning conveyed by its spatial order as
well as by its decorative themes. It is also apparent
that such meanings can be recovered even from
the designs o f peoples who left no written records
and whose societies vanished long ago. An exam
ple is A ndr Leroi-G ourhans study o f the intended meanings that were set out in the distribution of
the Paleolithic cave paintings o f southw estern Eu
rope. In caves like Lascaux, the chambers o f the
cave are marked out by the animal paintings into
ritual divisions corresponding to a liturgical and
processional order o f ceremony as in a sanctuary
(1965, chap. 6).
The main divisin at Teotihuacn among fgural
decorations, upon an architecturai scale, is be
tween those on the interior walls, with their
painted dadoes, and those on the exterior, usually
applied to the enfram ed tableros and their sloping
bases. The inner walls bear emblems, processional
figures, and narrative scenes. They are generally
enframed at the sides and across the top by wide
borders containing figures that reflect or echo the
sense o f the larger figures in the enclosed field.
The repertory on the interior walls is far richer
than the surviving repertory on the tableros, and it
is as though the tableros bore only the m ost com-

pact designations for the rituals housed within. If

so, the tableros may be compared to sign paintings
on shops and business concerns in the urban scene
today: they desgnate different ritual activities by
using logographic symbols not unlike those in
modern advertisements or on road signs portraying an action in pictures rather than words.
At Teotihuacn the ancient architecture built
from 300 b . c . to about a . d . 700 included many
forms that are suitable for iconographical study.
They occupy a period spanning about ten cen
turies, and they reappear at distant places in
Guatemala and Yucatn, signifying at least some
continuity o f meaning both in time and in space.
The most distinctive and durable physiognomic
trait o f this M esoamerican architecture is the ter
race profile. It was used to articlate pyramidal
platforms, and is often called by its Spanish ame
even in English writing the talud-and-tablero
profile, talud meaning talus and tablero signifying
the cantilevered vertical panel rising above the
slanting talus.
A m ore detailed analysis o f the form distinguishes several parts: the frame, which is like a
molding around a picture; the panel, like the plae
of the canvas in a painting; the ledge of cantilevered
slates; the shelf o f the projection of the frame; the
face o f the frame; and the skirt, or talus, like a
pedestal or base upon which the frame is m ounted.
The construction o f these terrace profiles (fig.
111-118) at Teotihuacn has been described by
Jorge Acosta (1964, fig. 14). A t the tableros bordering the square plaza on the north end o f the
275

276

A ncient Am erica

main avenue, the m ethod used about a . d . 500

began with the talus. U pon it an overhanging ledge
of slates, called lajas, was laid to carry the tablero.
The tablero itself began as an angled foundation at
the lower shelf projection. At the rear this frame
angled up to form the next terrace slope. O n the
lower shelf the tablero was com pleted in three
stages. First, the frame was added on the cantilevered shelf o f lajas. Second, the panel was built
within the angled frame foundation. Third, the up
per shelf was built on a second cantilever. The
stucco sheathing was applied in three stages: on
the talus before the laying of the lower cantilever;
on the panel and lower frame before laying the
upper cantilever; and on the upper frame and ad
joining talus prior to beginning the second-story
frame and panel. The overhangs protected the
stucco, and the stucco was laid in sheets calculated
to shed water and to cover the cleavage planes of
the construction.
A further problem with the tablero, beyond
protection for the panel and the talus from weathering, was its stability. T he projecting frame, rest
ing upon ledges of slate, acted as a cantilevered
weight destined sooner or later to collapse. Early
frames are wider, deeper, and thicker than late
frames. Early frames are made o f massive masonry
blocks, as at the early Quetzalcatl platform in
the Citadel, in contrast to the thin, shallow frames
of the later centuries, resting upon slate ledges.
These ledges were a limiting factor. The overhang o f the frame could not exceed the available
lengths and breaking strength o f the slate cantilevers. Thus the frame was unlikely to project
more than about 30 centimeters. In fact its sheltering function diminished geometrically with linear
increases o f size. The result is that few panels ex
ceed the height o f a m an.1
As to the visual effect, the overhanging frame
upon a small talus produced a shadow when the
angle o f the sun was high. This shadow gives an
effect of levitation, as o f the massive tablero frame
resting upon a cushion of darkness, especially evident in the small courtyard dwellings o f the last
periods.
The main difference between public, or religious, and secular, or private, constructions was
probably one o f size. Large platforms were public;

small ones were for dwellings or for household

shrines. But within the household, a difference of
proportion preserved the difference between divinity and the people o f the compound. The shrine
rose high upon its slanting base, while the surrounding megaronlike chambers occupied lower
pedestals. These domestic tablero profiles may reflect an encroachm ent by civil authorities upon
cults of natural forces.
It has long been accepted that the buried and
filled-in lower-level structures at Teotihuacan belong to the earlier history o f the city in the Mic
caotli period. This was the great urban renewal
period in the third century a . d ., associated with
the overall grid and the laying out of the main
north-south avenue.2 Perhaps two events are indicated: an early real estte subdivisin about a .d .
200 and a later rash of construction about a .d .
500. The gap defines the boundary between early
and late. The buried platforms of the earlier age,
however, show the same kinds o f proportional
schemes, in the reladon o f the tablero to the talus,
as do the surface platforms built over them upon
the reticulated grid o f blocks 57 m eters square.
These were imposed about a . d . 200 in the Mic
caotli phase, or Teotihuacan II (Ren Milln
1960, p. 7). The proportional schemes vary be
tween short tableros on tall bases and high tableros
on short bases. Very tall bases were used in the
early building history of the site, and the shortest
bases appear in the late periods, as shown in the
accompanying table, w here the height o f the base
is taken as the unit, and the tablero height is given
in reladon to this unit. T hree groups are defined:
ratios below 2:1, from 2:1 through 3.5:1, and 4:1
and larger.
At the Tem ple of Agriculture, the earliest plat
form has two terraces with tableros built upon
high sloping skirts. The lower talus is two-thirds as
high as its tablero and the upper talus is actually
higher than its tablero. The new platform, built to
face the avenue, has a tablero fully three times the
height of the talus. This three-to-one proportion
reappears in the facades surrounding the Moon
Plaza. It also appears in the buried platform called
the Tem ple o f the Feathered Shells underneath
this renewal at its southw est crner. Thus the 3:1
ratio appeared before renewal, and it charac-

Iconographic Aspects o f A rchitecturai Profiles

Table III-2.
Ratios between Tablero and Talus Heights
(l=talus height; E=early period to a .d . 450; L=late
period after a .d . 450*)
.93:1
1.5:1
1.5:1
1.66:1
1.9:1
1.9:1
2:1
2.3:1
3:1
3:1
3:1
3:1
3:1
3:1
3.3:1
3.5:1
3.5:1
4:1
4:1
6:1

Temple of Agriculture, first period,

upper terrace
Temple of Agriculture, first period,
lower terrace
Tetitla Altar
Zacuala Main Court, southwest side
Main avenue, Structure 16, jaguar
mural
Main avenue, Structure 15, upper ter
race

Temple of Agriculture, third period,

upper terrace
Main avenue, Structure 15, lower ter
race
Caracoles emplumados, beneath
Quetzalpapalotl
Subterrneos, lower level
Moon Plaza, Structure 5
Temple of Agriculture, third period,
lower terrace
Yayahuala, northeast crner of court
Yayahuala, altar
Subterrneos, Tajn-style tablero
Quetzalcoatl Platform, inner terraces
Quetzalcoatl Platform, outer terraces

Zacuala, court to south

Moon Plaza, Structure 15, third terrace
Atetelco, court, east building

L
L
L

E
L
L
E
L

L
E
E
L
L
L
L
E
E
L

*Rene Milln (1966a, b) thinks that the idea o f a great grid

renewal began about A.D. 200 with m ajor rebuilding occurring after 450 in the Xolalpan and M etepec periods to 750.

terized many elevations both during and after the

renewal o f that spot.3 Y et the table suggests that
the 3:1 ratio became canonical with the great re
newal, although other ratios could be applied to
special situations in renewal design. For instance,
at the mural o f the great jaguar on the east side of
the avenue, the ratio approaches 2:1 upon a ter
race height o f about 3 meters. This ratio repeats in
adjoining platforms o f smaller size. Perhaps de
signers achieved imposing axial effects along ex
tended facades by using tall bases. But in more

277

inclosed settings like the Citadel court and the

Moon Plaza, the 3:1 ratio was preferred, perhaps
to give the effect of accessibility. In the intimacy of
the residential courts, short bases in ratios o f 4:1 to
6:1 were favored.
In its com pleted appearance about a . d . 600, the
main avenue was an extraordinary composition of
differing architecturai variations. All were held to
gether in a system set by the tablero-and-talus
theme. In the regin between Platforms 15 and
16, for instance, a long section in two tableros of
2:1 ratio was interrupted by Platform 15, where
two terraces o f different ratios (2.2:1 and 5.5:1
above) displaced the larger rhythm with their
more abrupt and cise intervals.
H ere and elsewhere, as at Platform 16, the ter
race heights were not equal: the upper tablero was
raised upon a base higher than that o f the lower
one, as if to improve its visibility from below. At
Platform 18, to the east o f the Pyramid o f the
Moon, the four terraces are unsymmetrically divided by the stair. The western terraces are longer
than the eastern halves by about half a m eter (70
centimeters at the lowest terrace, 60 centimeters
on the second, 50 centim eters on the third, and 40
centimeters on the top terrace.)4
Tablero and talus are om nipresent at Teotihua
can. They domnate all parts o f the vast city as the
privileged form chosen to distinguish the facades
of temples and platforms. N o other exterior profile com petes, excepting two early versions that
will be discussed below.
The carved and painted decorations on these
enframed facade panels bring to mind modern
commercial showcases. Their exposure to millennial ruin and leaching has destroyed most o f their
ornament, but the surviving designs are like those
on pottery vessels, reflecting cult practices and
foreign styles.
The repertory o f the ideas represented is not
very large. The Pyramid o f Quetzalcoatl (fig.
111-119) is unique in having full-round and relief
sculpture with painting. These tableros and bases
are constructed o f blocks o f precisely cut stone.
Each talus bears in relief carving a feathered ser
pent set among seashells. The tablero above is dec
orated with similar feathered serpents whose
heads protrude as full-round sculpture, alternating

278

A ncient Am erica

with scaly goggled heads o f unidentified nature.

Thus talus and tablero both show feathered ser
pent forms among conch and pecten shells, sug
gesting watery associations. A nother feathered
serpent, but lacking water symbols, appears in a
large tablero painting at Zacuala (Sjourn 1966c,
fig. 9, pl. 54).5 A diadem is shown among the
feathers.
A second them e o f tablero decoration is a water
representation o f banded designs set with eye
forms, as if to suggest the brightness o f flowing
streams in horizontal layers or falling rain in ver
tical bands. These appear at Zacuala and at the
Moon Plaza. At Zacuala, in the third patio, there is
a tablero with a human m outh flanked by these
vertical eye bands. O n the main north-south avenue a large red jaguar painted with diagonal stripes
appears upon an east side panel; another panel
flanking the stair shows remains o f a yellow jaguar.
Among such rare examples o f tablero decora
tion, water associations seem to have been significant but not dominant. Feathered serpent and jag
uar are common symbols in Teotihuacn icono
graphy. Their exact meaning is not known, despite
many efforts to extraplate by analogy from Aztec
sources. The feathered serpent may unite ideas of
earth and air. The jaguar may relate to water rites.
At the Subterrneos, farther north near the
M oon Plaza, an older buried platform is painted
with roundels on the frame and interlocking
scrolls resembling those used in Tajn sculpture
(Proskouriakoff 1954).6 The signifcance o f these
interlocking scrolls is distinctly foreign at T eoti
huacn. They have been interpreted as symbolic of
the conflict o f opposing forces, being most commonly associated at Tajn with ball-game iconogra
phy (Tuggle 1966).
Because the preserved representations on ta
bleros are rare, it is profitable to look at those
pottery decorations that may portray the tablero
and talus. These are molded tripod supports, nose
ornaments on human heads, and painted images of
temples in muris and on pottery.
Tripod supports representing tableros are com
mon, and they occur late in the seriation o f vessel
feet. Crossed bands, roundels, conch shell sections, frets, S-curves, and chevrons may appear in
the tablero (fig. III-120). M ost bases carry a design

resembling three downward-pointed feathers;

others have inverted trapeze-and-ray signs (Peterson 1952, pp. 1314).7
These all may have served as logographic expressions related to scenes or designs on the cylindrical vessel wall. For example, the crossed-band
design on the tablero feet concurs with a warrior
wearing a butterfly helm et and geom etric but
terfly nose-pendant, carrying a butterfly shield device, seated among long-stemmed flowers (fig.
111-121). If the butterfly be accepted as the emblem o f the soul (K ubler 1967, p. 9),8 because of
its frequent occurrence in the decoration of burial
pottery at Teotihuacan, this scene may show
funeral rites or beliefs. A nother vessel (fig.
III-122) shows flayed-skin X ipe heads on Maltese crosses separated by drops o f water. H ere the
association o f life within the dead skin of the
flayed-head image seems to relate death and fall
ing water with crossed bands on the slab feet. This
image o f crossed bands may be an isolated or sin
gular case o f the net designs studied by Hasso von
Winning, who interpreted them as ritual devices
for bringing rain (1968, pp. 3146).
N ose ornam ents shaped as tablero-and-talus facades frequently hang from the septum on pottery
face-masks enshrined inside the templelike funerary urns o f Teotihuacan burials. These nose orna
ments also appear on painted vessels among the
ornaments o f the personages depicted. The deco
ration often resembles a geom etric butterfly,
sometimes characterized by wing markings, round
eyes, and by a recurved proboscis (Caso 1963, pp.
7797; Franco 1961). It occasionally decorates
tripod-vessel walls, alternating with other logograms like the RE-glyph, or a stylized flower, or a
warrior. In general, its contexts resemble those of
the vessel foot shaped as a tablero facade. These
associations indcate a death-and-rebirth sym
bolism for butterflies, associated with water and
flowers and earth (the RE-glyph) (Kubler 1967,
p. 9).
Images o f tem ple platforms, painted or incised
on pottery and muris, insofar as they represent
only themselves, are perhaps the most revealingof
all representations o f tablero-and-talus forms at
Teotihuacan. O ne sherd (fig. III-123), published
by Von W inning (1947a, fig. 1), shows a temple

Iconographic Aspects o f A rchitectural Profiles

platform with a stairway flanked by tableros, alter
nating with goggled rain masks. Leaves sprout
from the mouths o f the masks. The tableros and
their bases, however, are w ithout ornament. This
juxtaposition seems to connect ideas of worship
with the goggled mask.
In other words, the tablero and its base may
have had a meaning like sacred architecture.
The domain of cult and ritual was marked off from
secular building by the notched and cantilevered
profile o f the platform. The tablero may or may
not bear indications o f a specific cult. Its main purpose is to set sacred edifices apart from dwellings
and other secular buildings. If this guess is correct,
the tablero and talus are significant without additional information, which, if it had been offered,
might even be redundant. In this case, we may
suppose that the architectural profile is, in itself
and of itself, a m ajor indicator o f meaning, specifying both the function o f the building and the ethnic identity o f the builders.
The similarity between the geom etrized form of
the butterfly and the tablero profile at Teotihua
cn has already been pointed out. If butterflies
represented otherworldly life, this geometrical
equivalence between butterflies and temple plat
forms may have provided a metaphorical exten
sin of meaning for both butterflies and temples as
to the mystical promises o f religin about life after
death.
A nother poetic extensin o f the meaning of the
tablero-and-talus profile appears in the vase paint
ings found at Zacuala(fig. III-124). O n these vases
a mountain is shown with a rectangle at its summit,
resembling the cloud caps that frequently envelop
the tops of the hills rising from the floor o f the
Valley o f Mxico. The promise of rain, and the
architectural facsmile o f a m ountain implicit in a
pyramidal platform, may have been combined
metaphorically in the profile we have been discussing to suggest a cloudland where rain is divinely generated, a paradise concealed among
clouds where butterflies abound.

279

[N ote to reader: The last four pages of this paper

are deleted, as they duplicate material covered
more amply in the last article in part III. See pp.
3 5 3 -5 6 e d . ]
NOTES
1 T he largest may be the painted feline two m eters long on
the east side o f the main avenue. Clara Milln (1966, pp.
1011) ascribed it to phase III in her mural chronology
(Xolalpan period), although the survival of this painting
was due to its tem porary obliteration by a renewal-period
facade o f Xolalpan date. W hether the painting should be
ascribed to Tlamimilolpa or to Miccaotli remains uncertain.
As a buried wall, it may be earlier than Miccaotli.
2 Ren Milln ascribes the grid and the construction of the
platform o f Q uetzalcatl to the Miccaotli period as about
A.D . 200 (1966a, p. 7), and the urban renewal to the
Xolalpan period a .d . 4 5 0 -6 5 0 .
3 T he dates of the great rebuilding are still in disorder. Al
though early excavators (M arquina 1922) proposed three
periods prior to the building o f the main avenue, Armillas
placed the rebuilding along the Street o f the Dead in the
Miccaotli phase (1944, pp. 12136), though w ithout convincingproof (see Clara Milln 1962, p. 8). T he carbn-14
date for Tlamimilolpa (Linn 1956, p. 191), from beneath
the lowest floor, dates its earliest tom b A .D . 2 3 6 -6 5 , confirming the supposition that the suburban enclosures
were already then in existence. Michael Coe (1962) has
reported carbn-14 dates in Burial 10 (Teotihuacn III,
Zacuala) as a .d . 3 3 0 -8 0 , and in G rave 24 at Zacuala as a .d .
2 9 0 -8 0 .
4 I am obliged to A rthur M iller for the use o f measured
drawings he gathered at Teotihuacn when working on the
muris there.
5 Clara Milln observed in her doctoral dissertation (1962, p.
99) that this mural fragm ent is o f uncertain provenience,
since it was found four m eters distant from the east plat
form with which Sjourn associates it.
6 T he Subterrneos scrolls are o f the rectilinear or earlier
type A as discussed by Proskouriakoff (1954).
7 Petersons examples all are vessel-leg fragments. T he vessels decoration probably reflected or determ ined the
nature o f the use o f the com plete vessel.
8 T he vessel, drawn in Sjourn 1956, p. 147, is in the mu
seum at Teotihuacn.
[The reference to A rthur M illers work on the muris at T eo
tihuacn, now published, is Tbe M uralPaintingof Teotihuacn
(Washington: D um barton Oaks, 1973). g k J

III.8
Early Architecture and Sculpture
in Mesoamerica: Commentary on
a Paper by T. Proskouriakoff

Tatiana Proskouriakoffs paper (Proskouriakoff

1971) is evenly divided in two halves, the first
about architecture and the second about sculp
ture. In the latter she treats painting and writing in
some detail where relevant. I shall therefore discuss her architecturai observations first, then
sculpture, and finish with the implications o f her
paper for the study o f painting and writing.

society integrate conjointly, like a mollusk and its

shell?
Proskouriakoff (1971, p. 141) also wonders
about the priority o f the village or the ceremonial
center. This is the question raised also by Bernal
(1971), when he discusses his belief that Oaxacan
architecture is older than Olmec. I believe the dif
ferences between Oaxacan and Olmec architec
ture may have to do with climate and ecology.
Buildings were needed in tem perate upland Oax
aca, but ceremonial spaces were favored in Olmec
C o a s t a l rain f o r e s t s . O f course, preservation may
be the differential factor as to dwellings. The spa
tial order o f M onte Albn (fig. III-47) seems more
tentative, being achieved gropingly during many
centuries and reflecting the outside monumentality o f La V enta (fig. 111-45) which, as Bernal
observes, lies closer to the beginnings o f civiliza
tion.
Bernals thesis (1971), however, that Oaxacans
invented architecture while the Olmecs perfected
sculpture is not easy to accept. H e declares that
architecture hardly . . . existed amongst the
Olmecs, but he has selected materials, and structural solutions for his criteria, and he disregards
architecturai thought, which is to materials and
structure as mathematical thought is to bookkeeping. At La V enta or San Lorenzo, the main themes
of monum ental M esoamerican architecture all
were airead y clearly stated and articulated long
before 400 b . c . These I have tried to classify in
four ways.

SETTLEMENT PLANS A N D
ARCHITECTURAL TECHNOLOGY
Proskouriakoffs remarks (1971, p- 141) on economic needs suggest that early public architecture
multiplies such economic needs, rather than merely responds to them. W ith this I would agree.
W hatever the independent variables of cultura)
history may be, they are not alone the economic
life of the people. They are also other ill-defned
cultural factors o f which economic behavior is
only one among the results. Indeed it may be that
artistic activity itself is closest to being an indepen
dent variable. Because of it, much change is likely
to happen, as we note in early societies, w here the
production o f ritual and its artifacts was a changing
activity satisfying many needs, rather than being a
grudging occasional Service to a greedy pantheon
and its priests. For example, can the plan o f T eoti
huacan (fig. 111-46) be correlated with afaculty for
planning and ordering the society as a whole? If so,
which is the prior expression or independent vari
able? The town or the society? O r do town and
280

Early A rchitecture and Sculpture in M esoam erica

As monum ental form, architecture commemerates valuable experiences by distinguishing one
space from another in a durable way. The basic
modes o f monum entality are the precinct, the hut,
the cairn, and the path. The precinct marks off a
sacred area; the hut endoses it in part; the path
signis a direction; and the cairn marks a point by
elevation (see K ubler 1958). From precinct to stadium forms one typological series; from hut to
cathedral another; from path to arcade-lined boulevard another; and from cairn to pyramid still an
other. The combinations o f path, precinct, hut,
and cairn yield all the possibilities o f monum ental
architectural form, not only in terms o f solids, but
also in terms o f the spaces bathing those solids. All
these occur at La Venta.
Proskouriakoffs remarks (1971, pp. 143 ff.)
that three-sided platform-and-building clusters in
Mesoamerica lasted at Tikal for a thousand years
beginning at the very base of the N orth Acropolis,
reassure us that w hether at Tikal, Oztoyahualco,
Kaminaljuy, Uxmal, or Quinatzn, these threesided clusters underline the traditions both of
house and tem ple construction. She is reluctant to
derive Petn architecture from the Guatemalan
highlands, but I think she believes the tradition of
clustered construction to be the fundamental one
in Mesoamerica, rather than the closed-corner
quadrangles, which appear only at advanced dates
in the Classic period and correspond to new functional types, such as judicial courts, or guildhouses, or seasonal residences for an elite population.
Looking again at the aerial photos o f M onte Al
bn (see Acosta 1965, p. 816), 1 am always ast o n i s h e d by the discrepancy between the publ i s h e d plans and the reality. Many axial systems
seem to com pete in the plan published by Marquina (fig. 111-47). T he observatory [M ound J]
seems as if it had been gradually s u r r o u n d e d by
changing systems o f reoriented axes. Its original
axis is uncertain: it may once have been aligned
with the earliest facade o f the Danzantes, to which
n o t h i n g else is parallel. These two structures a l o n e
belong to Period II. The S o u t h e r n platform later
on aligns with M ound M. T he northern platform
connects better with System IV than with anything

281

else. O ne is led to ask if there is not an early an

gular two-sided enclosure resembling M onte N e
gro in M onte Albn I, evident in the southwestern
quadrant, and defined by the platforms o f the cen
tral pyramid, G roup M, Danzantes, and System
IV, which later was enlarged east and north by
leveling. The observatory thus might reflect, in
its complicated and atypical arrowhead plan, these
changes in the field of axial forces. But other reconstructions o f this most complex layout are possible. Acosta (1965, p. 818) suggests that the level
ing now visible is no earlier than Period II (400
1 0 0 b . c .), when the first muris appeared in tomb
architecture, like the early glyphic inscriptions
upon stelae. It is certain only that M ound J is eccentric to every axial forc. M ound J is like an
eider and peculiar grandparent living within an excessively ordered family and differing from it in
every perspective, but reflecting an earlier way of
existence. The western courts are like temples for
the residence o f gods, but the newer eastern range
is lined with residential platforms for a newer kind
of dynastic dweller, and it has a ball court o f East
C ourt plan, perhaps even of Post-Classic date in
Monte Albn IV, like that o f Q uiotepec. Thus
Monte Albn appears to have grown northward
and eastward from the southwest crner. The
north platform with its sunken court resembles
the Classic architecture of Teotihuacn and the
Maya lowlands. The east range recalls Classic Vera
Cruz architecture, and the residential structures
seem akin to Late Classic Maya art of the centuries
approaching a . d . 1000. As Acosta makes clear,
the southwest nucleus began in M onte Albn I,
but the leveling o f the central plaza did not occur
until Period II after 400 b . c ., when an overall
planning scheme was initiated along lines first evi
dent at La Venta.
The situation at Izapa is especially suggestive in
this respect. The platforms and plazas are oriented
upon a north-south axis turned a little toward early
summer sunrises, as at Teotihuacn or Cholula, in
a grid designated as the M esoamerican solstice
plan. In her paper Proskouriakoff (1971, p. 150)
regards this plan as a Pre-Classic layout. Shook
(1951) sees the axis as filled with four-sided
courts. A t Izapa are many sculptured monuments

282

A ncient Am erica

all aligned in relation to the platforms, usually

paired as altar-and-stela combinations. These face
south m ore often than north and east m ore often
than west as is evident in M artnezs plan in Susanna Ekholms report on M ound 30a (Ekholm
1969). The sculpture is Pre-Classic, but the setting
and the alignment are like those o f Classic lowland
Maya sites. May we not suppose that Pre-Classic
sculpture was reset by Late Classic people? In this
connection the Bilbao site is o f this solstice-plan
type and its sculpture, like that o f Izapa, stressed
narrative scenes in formal alignments. This is at
some M iddle Classic date, to use Lee Parsonss
welcome term (1971).

PROPORTIONAL STUDIES
These alignments at Izapa raise the question o f
proportional relationships, about which I regret to
note that Proskouriakoff had no space to give us
her views. Among the great gaps in Mesoamerican
studies is the study o f architectural proportioning,
which is still an untouched subject, although the
topic in Egyptian, M esopotamian, and Minoan ar
chitecture has proved extremely fruitful, especially with the work o f Badawy in Egypt and Preziosi
in the Aegean. They have shown that the dimensioning of Pre-Classic architecture conforms to
geometrical figures constructed mathematically
upon numerical relationships, such as arithmetic
and geom etric ratios and Fibonacci series, which
are all implicit in num ber itself and w ithout which
it would be difficult to achieve the scaling, placing,
and conforming of large units o f construction.
Some ratios are apparent at La Venta: the geo
metric masks are square, but the diagonal of the
square is the length o f the panel including the four
pendant ornaments in the ratio 1:V 2, which is
Serlios diagonal proportion in the architecture
of the Renaissance in Europe. It reappears at T eo
tihuacan in the still-unpublished court on the west
side o f the Street o f the Dead immediately north
of the axis of the sun and in the macrocomplex
reponed by Wallram in 1966. I paced this off and
its width is to its length as 2:3 at the foot of the
platforms surrounding the court.

FRONTIERS A N D STYLE
Proskouriakoff (1971, p. 152) notes that during
the Classic Period . . . the two dom inant styles
the lowland style of the Maya and the highland
style of Teotihuacn expanded their influences,
m eeting finally on their frontiers. This wording,
which implies a tightly packed jigsaw puzzle map
does not seem to me to fit the emerging notion of
Pre-Classic Mesoamerican geography. Jigsaw
maps appear where definite territorial interests
mark out boundaries. That the shores o f the lakes
in the Valley of Mxico were thus demarcated,
we know from sixteenth-century documents from
Texcoco like Cdice Xolotl. But the jigsaw idea
still cannot be made to apply to what we know
about Pre-Classic Mesoamerica. Im portant valleys were trading partners, like the valley o f Oax
aca and the Olmec heartland, as Flannery pointed
out in 1968 and as Parsons (1971) also noted.
These trade routes surely attracted settlements
profiting from the m ovem ent o f people and
goods. The primary motive may have been trade
doubling with religin in the form of cults attracting large numbers o f pilgrims to the sanctuaries
maintained by priestly corporations. Trade developing along these pilgrimage routes would have
served the needs both of pilgrims and priesthoods, in a pattern of marketplace devotions
which persists today in the great romeras to Esquipulas, Chalma, Guadalupe, or Chichicastenango. Thus if we were to imagine the mental ge
ography o f Pre-Classic travelers, it would resem
ble a network o f paths rather than a jigsaw map,
and it would display the nodes or crossroads
more prom inently thah the network or the
boundaries. U npopulated deserts and mountains
would be less im portant than towns and their
alignment along rivers and road. Thus a Pre-Classic map would have looked like points, and
lines connecting them, rather than a map of
areas sharing boundaries.
The correlation between this geography o f mar
ketplace devotions and the history of art is best
docum ented in phenom ena like the great pil
grimage roads o f twelfth-century Europe, when
peoples carne together as pilgrims at the edges of

Early A rchitecture and Sculpture in M esoamerica

the known world, in Santiago de Compostela or at
Le Puy, in a behavioral pattern docum ented by
Chaucers Canterbury Tales. The history of the art
of the European pilgrimage roads is well known,
and it forms a distinct period o f medieval art. A
similar tracing out o f Precolumbian pilgrimage
roads in M esoamerica is already possible, using
such traits as architecturai profiles on platform ter
races and iconographic motifs and themes.

OLMEC HEADS
Proskouriakoff does not touch upon the problem
of the seriation of the big Olmec heads, perhaps
because she did not wish to overcrowd her inventory paper with a question that requires a long
history. B ut to art historians (and we all are art
historians when our main evidence is the sculpture
itself) the seriation raises an issue of period and
workshop. W hen I proposed a seriation in The A rt
and Architecture of Ancient America, written in
1959 and published in 1962, it was based upon a
number of two-valued traits: conical heads or long
heads; yoke brows or furrowed brows; blank eyeball or carved iris; closed lips or parted lips (Kub
ler 1962, p. 67). T he heads having conical shape,
yoke brows, blank eyeballs, and closed lips form
one group (for example, La V enta head no. 1; fig.
III-207) and the long heads with furrowed brows,
carved irises, and parted lips are another (for ex
ample, San Lorenzo head no. 4; fig. III-208). Assuming that schematic conventions preceded
more lifelike ones, I placed the first group earlier
than the second.
Charles W ickes dissertation (Wicke 1965) soon
after made use of G uttm an scaling to attem pt a
seriation, which seemed to confirm mine as Tres
Zapotes/La Venta/San Lorenzo in chronological
order. Michael C oes own seriation made before
his 1966 excavations at San Lorenzo places Tres
Zapotes at the end instead o f the beginning. His
excavations convinced him, later on, that San
Lorenzo preceded La Venta, on the evidence of
carbon-14 dates for the ceramic stratigraphy in
the pits w here the San Lorenzo heads had been
buried. But there still were heads at San Lorenzo
with La V enta traits, and vice versa, so that Coes

283

seriation seemed to disregard the stylistic evi

dence, although he speculated at first that there
was a transfer o f some m onum ents . . . and presumably leaders to La V enta . . . about 800 b . c .
(M. Coe, mimeographed abstract).
Later on Michael Coe (1968) suggested that all
the heads had been carved in less than a century.
The problem o f seriation was not discussed at that
Dum barton Oaks Conference in 1968. But it is
clear that one style centers at La Venta and another
at San Lorenzo. W hat is the meaning o f these dis
tinct styles? D o they reflect period or workshop?
If all are nearly the same age, two workshop traditions m ust be present. If the two groups are of
greatly d i f f e r e n t ages, such as 1200 B . c . a n d 400
b . c ., there is room for slow development. By analogy, however, with G reek kouros figures or with
early G othic portal figures in France, a brief devel
opm ent not longer than two centuries seems likely, perhaps with migrant sculptors moving from
site to site. But as Proskouriakoff (1971, pp. 1 4142) remarks concerning San Lorenzo, We have
no data either on architecturai assemblages or on
the forms o f buildings at that tim e.
To those words I want to add my own doubts
about C oes placing o f the megaliths in the Olmec
sequence. These large stones behave differently
from sherds. Sherds are discards but megaliths
are too valuable to stay long like sherds on the
refuse heap. Megaliths are as restless as heirlooms,
and they rarely remain long in one place. They
return to use again and again, being exhumed,
transported, smashed, and m ended and reappearing w here and when the need for big sculpture
recurs. An ancient megalith incorporares tradi
tion, and it therefore invites removal while resisting destruction. Being mineral, its own historie
age cannot be known by radiocarbon: it looms like
an uncharted island upon our im perfect maps,
skewing our graphs, and drawing our theories to
destruction. Dating megaliths by the surrounding
strata is like dating a piece of sculpture by the
architecture o f the museum containing it today.
The Olmec heads, however, pose still another
question. Why are they so big? For such work to
exist there must have been an artistic tradition and
a psychological demand, as well as a suitable set-

284

A ncient Am erica

ting. Artistic traditions comprise meanings, fig

ures, and techniques. The iconographic tradition
was humanistic, in that im portant meanings were
conveyed as personifications. The figurative tradi
tion perm itted portrait likenesses and natural appearances. The technical tradition allowed subtle
effects o f fleshy texture and animated gaze. N one
of these is Iikely to spring suddenly from nowhere,
and if Coe is right in assigning the great heads to
the San Lorenzo Phase ( 1 2 0 0 b . c . - 9 0 0 b . c .), the
antecedent stages are totally unknown to us, and
we find ourselves again in what I cali Vaillants
dilemma by having to assign an early position to
late productions which we know intuitively to
have em erged from a long preparation.1

MACROTECHNICS
M ens rational observations and experiments dur
ing stone age periods often required extremely
large-scale instrum ents and theaters o f operation.
Stonehenge and Avebury in England are gigantic
observatories prefiguring small optical instru
ments o f glass and metal. Like Stonehenge are
those observation platforms in America composed
of buildings at whose calculated intervals the sun
was seen to rise on the solstices and equinoxes, as
at Teotihuacan or Uaxactun. T he star-sighting
lines stretching many miles across the south Andean Coastal deserts, which have been studied by
Paul Kosok and Maria Reiche (1947), are another
instance o f the effort to achieve accuracy by magnifying the size of the instrum ent or position taken
by the observed.
It also seems justified to speak o f a macrotechnic character in early instrum entation, a character
which reappears in craft operations w here primitive instrum ents are used to achieve work o f great
delicacy and precisin. People using stone tools
must also observe a direct and necessary relation
between the size of the instrum ent and the scale of
the work it can be made to produce. A stone handaxe cannot be made to produce m inute effects or
can the drills useful for working jewel stones like
jadeite be made to produce large sculpture economically. Indeed the effort to shape stone with
stone tools inevitably led to the realization that an
enlargement o f the work to colossal proportions

was the only way to achieve finely detailed control

over sculpture. Stone hand-tools have their charr
acteristic weights and shapes: a stone blade-edge
will cut finer detail when the size o f the work itself
is enlarged, but below large work-sizes, stone
tools cannot shape lines or modeling finer than
their own edges. It is like the problem o f trying to
draw a map with a pen coarser than the scale of
representation: the line showing a river looks
many times wider than the river should. Thus the
Olmec sculptors, envisioning and wanting finely
modeled anatomical detail, discovered that they
could achieve it in stone only by working at the
largest available scale.
The lessons thus learned m ust have been analogous to working today under great magnification:
the mind can witness the effect of every stroke in
a m anner otherwise unknowable at ordinary
scale. Colossal sculpture thus was a mode o f discovering the correspondences between organic
forms and their representation. The work perm it
ted the invention o f ways o f representing life
forms all while verifying their plausibility or
fidelity to the model, by magnifying the worksize beyond the limitations o f N eolithic tool-size.
In this perspective, colossal sculpture was analogous or equivalent to drawing in m ore recent
time. W here we experim ent by drawing our idea,
the Olmec sculptors felt their way into the un
known or uncharted domains o f exactly representative art by sculpture with stone celts and mauls
and picks, which they wielded as sensitively as a
draughtsman today handles his pencils and
brushes. Among early civilizations elsewhere, the
emergence o f painting is Iikely, as in the Mediterranean civilizations, to follow after sculpture and
to depend upon conventions of representation
established in sculpture. This was perhaps because o f the same proclivity among early figural
artists to prefer full-round work to illusion in two
dimensions. Illusionism was technically more difficult, depending for its appearance upon a prior
mastery o f full-round replication before the devices o f illusionism could be approached, first in
low-relief sculpture, and later in paintings and
drawing. This simplistic versin o f the earliest
history o f art has an od history in art theory, but
this versin is suggested by the archaeological re

Early A rchitecture and Sculpture in M esoamerica

cord for Mesoamerica, as Michael Coe and Igna
cio Bernal have made clear in their writings, and
as the colossal heads themselves lead us to believe. Bernals phrase about Classic Maya
painted stone is a happy find.
O ur most penetrating recent insights into the
nature of all artistic process were given by Ernst
Gombrich in A rt and IIlusin (1960), when he
showed how images arise from coded schematic
conventions m ore than from direct observation.
In providing a psychological model for this pro
cess, whereby works o f art are generated out o f
other works o f art rather than by the artists returning to nature, G om brich was extending the
argument first presented by H enri Focillon in The
Life of Forms in A rt (1934), which was later popularized by A ndr Malraux in The Psychology of A rt
(1949-50).
This argum ent proposes that the making o f im
ages depends upon schematic conventions. These
are codes for stating relations which will make im
ages look like nature. Such schemata are: placing
forms, coloring them, marking their texture, and
devising clues to expresin. T he schemes do not
simplify or abstract so much as classify and artic
late. They are relational coding conventions, classifying experience in a game like that in which the
players agree to describe everything in a two-term
language with no oth er words than ping and
pong. It is G om brichs contribution, based upon
the psychology o f form -perception, to have seen
that finding such a code must precede the making
of images. In a next stage, the making o f images
precedes their matching against nature. Finally,
matching precedes correction. This model of operations explains why art has a history. Indeed all
artistic discoveries are equivalences enabling us to
see reality as an image and vice versa.
Gombrich goes on to establish that the form of
an image is affected by its function. G reek images
serve a narrative function; Renaissance forms convey a structure o f space; Chnese landscape paint
ing serves the aims o f poetic invocation. M odern
art in turn proceeds by finding, making, and
matching equivalences for the inner psychic world
in a coded relational convention like those o f the
other m ajor historical configurations.
N o part o f American antiquity was considered

285

by Gom brich in this context, but Mesoamerican

visual traditions fit easily into the schematic con
ventions o f his analysis. H e regarded nature as
the object o f European mimesis or imitation, but
in ancient America, there is no systematic effort in
this direction, because the idea o f nature was
differently present, being conveyed m ore by a
wide range of cult practices and animistic beliefs
than by natural philosophy.
Because our knowledge of ancient American
thought about the natural world is mainly restricted to artifacts which suggest ritual use for cult
purposes, we may substitute cult for nature as
the object o f imitation (H vidtfeldt 1958). The
Precolumbian artists aim was to find coded con
ventions corresponding to the profusion o f cults in
which American myths were embodied and to
make images that would be recognized as portraying cult practices. T he relational code and the
making o f equivalent images, followed by match
ing and correction, thereupon yielded, as they did
elsewhere in the world, a history o f art.
Ever since G reco-Rom an art, as in Pompeiian
muris, the key inventions in the European rela
tional code have been foreshortening, tonal modeling, texture by highlights, and physiognomic
clues to expression. But few o f these inventions
appeared in ancient American art:
1. In place o f the perspective convention o f fore
shortening, however, we can substitute an
other sort o f drawing. H ere each delineation is
spread out for the greatest ideational clarity, in
forms often approaching the orthogonal projection o f architectural drawing, as in Mexican
manuscript and mural traditions.
2. Instead o f tonal modeling, we can consider
color symbolism in ancient America. These fat
tones w ithout shaped modeling describe ideas
rather than appearances. They mark cardinal
directions and symbolize elements, such as
earth, fire, air, and water, as well as portions of
calendar and places and titles. As in heraldry,
ancient American color is not descriptive but
prescriptive: color changes as to meaning in
stead o f changing as to appearance.
3. The suggestion o f material texture by highlighted accents derives from tonal modeling,

286

A ncient Am erica

and it is therefore absent from the American

relational code. If the nature o f materials cannot be described by conventions in line or
color, ideas such as ripeness, richness, or glossiness can be suggested by other means. For in
stance, an ideograph signifying preciousness
and representing jeweled ornam ent may be infixed or affixed as an explanation in a glyphic
composition which approaches the pictorial
forms o f writing.
4. The European coding o f expression by physiognomic clues has many ancient American
parallels. Transcendent rapture can be sug
gested by prayerful posture, as in the Classic
Maya wooden kneeling figure in the Museum
o f Primitive Art. Erotic excitem ent appears in
some Jaina figurines, w here theatrical posturing and menacing gestures were also possible.
Maya wall painting at Bonampak resembles
Egyptian New Empire mural art as to ex
pressive variety, symbolic color, and linear indications o f texture. Classic Maya ruler portraits in monum ental sculpture distinguish
young and od in several age groups, but the
sculptors resources for indicating emotions
were underdeveloped, being channeled into allegorical forms and associated glyphlike attributes.

EARLY WRITING
Several points about early writing might be made
in these contexts. W riting can properly be regarded as a m ethod for miniaturizing or compacting more bulky Communications, such as a sculp
ture or painting. In this context writing seems to
separare two functions. W here figural art both
represented and communicated, perform ing two
services simultaneously, phonetic writing does not
imitare by images. It merely communicates, by
nonrepresentative signs.
But in M esoamerica it is not yet clear w hether
writing was either phonetic or logographic or
both. The advantage o f logographic writing, as we
know from European roadsigns for motorists, is
that pictures are instantly legible in every language, whereas phonetic signs require translation.
If Maya writing, as Thom pson and Proskouriakoff

suppose, was less phonetic than logographic, then

Olmec w ritten signs, which still are undeciphered,
can be supposed to have been mainly logographic,
for use by pilgrims and foreigners as well as by
local adepts. This, I take it, is the sense of Prems
com pounded ideograms (Prem 1971).
Proskouriakoff (1971, p. 152) assumes the exis
tence of manuscript painting as early as Olmec
time, unless I mistake her meaning. I think that
she finds her support in the similarity between
early sculptural conventions and the manuscripts
we know, none dating from before a d 1000 at
the earliest. It is well known that classical M editer
ranean manuscript painting appeared only in late
Imperial Romn books, recapitulating long prior
sequences o f sculpture and mural representation
(Weitzmann 1947). The case is m entioned, not to
suggest a necessary law-of-necessity, but only to
mark the generic lateness with which book illustration is usually associated among art histor
ians.
Like Prem (1971, p. 112), Proskouriakoff (1971,
p. 147) assigns a greater role among peoples to
visual imagery than to writing and mathematics. I
would agree, with the modification that visual im
agery in Mesoamerica was adapted very early to
serve as writing, and that what is called writing
never was divested o f all its attributes as visual
imagery.
The case o f Classic Maya inscriptions is most
revealing here. Thanks to Proskouriakoff we have
gained since 1960 an entirely new historical approach to the inscriptions, but it is now evident
that about half the glyph blocks, or less, in most
texts pertain to calendrical matters. The remaining
blocks give ames and titles. Some verbs eight
or ten action glyphs have been identified, and
another handful o f place-names. The gist o f these
historical clauses seems always to pertain to local
individual rulers or priests whose lineage, titles,
offices, and services in war o r peace are enumerated. But the calendrical arm ature o f the Initial
Series date, Supplem entary Series, and Distance
N um bers, as well as occasional cyclical counts like
the 819-day augural count, actually take up most
of the room. The interstices, which sometimes admit only one or two glyph blocks, are not big
enough to contain vast am ounts o f information. As

. .

Early A rchitecture and Sculpture in M esoam erica

Thom pson (1965, p. 657) put it: tenses and pronouns seem to be absent.
A nother inference to be drawn from the histor
ical signs concerns the intended vales of the
graphemes. Like the calendrical portions, these
historical sections convey little or no phonetic val
u. They are m ore logographic than we used to
believe. Entire thoughts are conveyed by one
glyph block, composed of many graphem es, as in
modern advertising logos (shortened from logograms), in which one simple but pregnant schema
conveys the nature and use o f something being
advertised. In the Maya glyphs phonetic and syllabic vales2 are not excluded, but their occurrences seem less systematic than one would hope
for in an alphabetic writing like the one Knorozov
has proposed.
It may be that Maya writing records few Maya
sounds, and that its function was m ore to provide a
visual language to be understood w ithout reference to any single language or dialect. Thus the
bar-dot numeris are intelligible across all barriers
of speech, like the period glyphs and the lunar
expressions, all of w hich get meaning from their
positions in the long Initial Series clause. In this
respect Maya inscriptions are less like writing as
we know it than like a pictorial coding for histor
ical Information. Thus an emblem is usually term i
nal in a clause opening after a Secondary Series
date. Action glyphs can precede ames, but ames
occur w ithout such prefatory remarks. A nother
glyph-cluster states that the subject captured a
prisoner; others have to do with sacrifices and
bloodletting; with birth, inauguris, anniversaries
and death. O f this code, the rare full-figure glyphs
convey historical material, as on M onolith B at
Quirigua. There, the appellative of Two-Legged
Sky and the em blem o f Quirigua are given in full
figure variants at the end o f the full-figure clause.
This code may have originated with schematic pic
torial expressions which were keyed into a se

287

quence. If so, it is possible that the full-figure

glyphs like those at Tikal (Stela 1, Bz2) or Yaxchiln of Early Classic date reflect m ore o f the earliest preliminaries of Maya historical coding than
previously believed.
Thom pson (1965, p. 636) was the first to com
pare Maya noncalendrical writing to historical
coding in Oaxaca. Such records appear on the
Zaachila slabs as well as in genealogical m anu
scripts. Their pictorial character is much m ore primary than in the Maya coding, but the relationship
between the date and the event to which it pertains
is like that in Maya inscriptions. In Oaxaca, the
record is m ore concerned with events and less
obsessed with dates, but the intended meaning
emerges from the position o f the pictures in the
sequence, and the relationships of the figures are
regularly coded to a limited num ber of types. Barthel (1968) has recently com m ented on these par
allels between Mixtee genealogies and Classic
Maya stone reliefs.
To conclude: em ergence is like the actor Corn
ing on in the prologue to the play. But unless he
can say something o f valu to the audience they
may walk out on him. H ere is the question once
again of the valu systems about which G. Willey
(1971) spoke in this, the grandest of games.
NOTES
1 [See the following article in this part for a later and m ore
extensive analysis o f these problem s. ED .]
2 T hom pson (1965, pp. 939, 645, and 65253) shows that
shifting o f affixes arges against syllabic decipherm ent;
that particles w ere expressed as affixes {al, il, te)\ and that
Yucatec rebus puns alone received pictorial treatm ent. O n
the oth er hand, he regards pictorial glyphs as insignificant
(p. 656). M ore recently, T hom pson has developed the idea
o f m etaphorgram s.
[The idea of macrotechnics (pp. 28485) in relation to
stone-age periods, having had no acceptance and no disproof,
may still be regarded as viable. g k J

III.9
The Styles of the
Olmec Colossal Heads
[In memory of L. J. Savage (1917-1971)]

Investigators who have made efforts to serate the

colossal Olmec heads have until now assumed no
m ore than that their sequence would be a onetrack alignment w here each head, regardless o f its
location, occupied a different temporal position in
the relative sequence among the heads.1
In these attempts, little or nothing was said
about differences o f sculptural tradition among
the heads or about the interplay o f artistic influ
ences through Olmec space and time. It is the purpose o f this paper to begin to identify and trace out
such patterns o f interchange among different
workshops.
Since 1967, furtherm ore, many new conditions
require a reexamination of the possibility of tem
poral seriation. These conditions are the fol
io wing:
1. Reassessment o f the excavations at San
Lorenzo (196668).
2. Discovery o f m ore heads in stratigraphic asso
ciation at San Lorenzo (1970).
3. Rem easurem ent o f the radiocarbon dates for
La V enta (1967).
4. The super-head discovered at Cobata (1970).
5. G uttm an scaling as nondirectional order.
6. M ultidimensional mapping o f geographic and
stylistic dissimilarities.

1. EXCAVATIONS AT SAN LORENZO,

1966-1968
We still do not know the correct duration o f the
styles of the colossal Olmec heads or the order in
288

which they w ere carved, although exact knowl

edge o f other m atters was augm ented greatly in
the excavations at San Lorenzo Tenochtitlan, directed by Michael C oe.2
The declared purpose o f these excavations was
to establish the stratigraphic record, that is, to
associate the sculptures with a cultural phase
which could be dated by normal m ethods, such as
cultural comparison and radiocarbon analysis.3
Coe has carefully pointed out that the published
radiocarbon dates do not directly measure the age
of the colossal heads.4 The key charcoal samples
were taken from stratified hearths in the riverbank
at Tenochtitlan, 3.5 km. distant from San Lorenzo.
These six hearth dates carne from deposits nearly 6
m. deep, associated with whole and broken vessels
of the San Lorenzo Phase (1 2 0 0 -9 0 0 b . c . ) and the
Villa Alta Phase ( a . d . 8 0 0 -9 0 0 ).
Among the remaining eleven radiocarbon sam
ples only three from the San Lorenzo site are admissible as relating to stone m onum ents.5 Monum ent 21, a stone with relief o f prowling feline or
canine, was found overlying sample Y-1907,
which measured as a . d . 330 120, and Coe rejected it as too young, being from complex stratigraphy at the juncture of three gullies.6
The other two radiocarbon samples relate to
M onum ent 30, a fragment of a stela with werejaguar-dragon relief. Sample Y-1911, measured
at 1140 B.c. 80, is from a zone predating the
final setting o f the m onum ent. Sample Y-1933
(1310 b . c . 120), from the excavations of M onu
m ent 30, is identified as an unnamed, pre-San
Lorenzo Phase.7

T he Styles o f the Olm ec Colossal Heads

In effect, only sample Y-1911 is immediately
relevant to any Olmec stone relief. It carne from a
hearth under the setting o f the m onum ent in
Zone D at the top of Floor 7, and it tells us nothing
about the date o f the setting o f the m onum ent
(which is a fragment), or about the date of its fragmentation, or about the date of its carving.
M onum ent 30, like the other buried stones at
San Lorenzo, was buried in a stratified site after
mutilation. But the carbon-dated strata pertain to
the burial pit rather than to its contents. These pit
strata are like the walls of a museum: the building
date of the museum does not reveal the date o f the
objects it contains. These objects may be older or
more recent than their container.
In any case, the analogy of the museum is a
useful one because o f the probability that San
Lorenzo became at some time a storage area for
the conservation by burial of valuable carved re
cords o f the past. It is also possible, of course, that
they were buried to rem ove them from public
view for ideological reasons.
M onum ent 34, excavated in 1967, yielded what
Coe has called the clinching evidence.8 The excavator reports that the kneeling figure was
placed upon a red gravel floor and then covered up
with a special fill.9 T he burial operation required
thousands o f cubic m eters . . . brought in basket loads on the backs o f m en. 10 This material
was previously described as a special fill containing a great deal o f limestone and fragments of
white bentonitic rock brought up in basket loads
from the deepest ravines. 11 Elsewhere Coe adds
that the burial fill he has dated as 900 b . c . was in
part derived from ceremonial m ounds leveled at
this tim e. 12
The question arises as to the date of the fill in
the burials. Coe says that in every case the m onu
ment . . . was completely covered by, and encased in, a tough San Lorenzo [se. Phase?} fill
containing sherds, chunks o f bentonite, gravel,
and other rocks. 13 But if the fill came from the
deepest ravines and from dismantled cere
monial m ounds, that buried m useum at San
Lorenzo was built with materials salvaged from
other settings.
To return to M onum ent 34, the excavator says
that some large sherds o f San Lorenzo Phase date
were placed between the legs o f the figure be

289

fore being covered up in the p it.14 The filled pit

was then covered by a shallow layer of late San
Lorenzo refuse (Zone B), topped by a thin stratum
of Villa Alta date (Zone A ). Coe therefore regards Z one A, dated a . d . 800900, as leaving no
question about the placement o f these monuments in the San Lorenzo Phase before 900
b . c . 15 Y et he also leaves open the possibility that
the contents o f the m useum could be either ear
lier or later than the m useum itself. At least we
may so believe until the full excavation reports
appear.
2. N EW H E A D S A T SA N L O R E N Z O
(SL 8, SL 9)
The stratification found by Brggemann and Hers
at San Lorenzo brings into question that dating of
all the heads as carved late in San Lorenzo Phase
III prior to 900 b . c . 16 T heir excavation showed
that some heads at San Lorenzo not only were
never deeply buried, but were made o f a different
stone, presumably carved after the burial of the
mutilated stones, and were lying in the topm ost
half-meter o f the humus layer at San Lorenzo.
This is the case with M onum ent 53 (San
Lorenzo head no. 8, hereafter cited as SL 8),
ound, like SL 2, lying face up at a depth of only 50
cm. below the surface in the top humus layer. The
stone o f SL 8 is the same microstructural variety of
andesite as M onum ents 51, 17 (SL 6), 18, 55, 57
and 2 0 .17 T he position o f SL 8 shows that the head
has been long visible near the surface, and that its
damaged condition owes m ore to weathering and
flaking than to deliberate Pre-Classic mutilation.
M onum ent 61 (SL 9) was found unmutilated
and buried 4.15 m. down. The buried heads and
the heads near the surface tell us of different uses
or the heads. Some were concealed, but others
remained visible, like SL 2 and SL 9, being moved
presumably from the San Lorenzo Phase zone into
much later surface strata. Brggemann and H ers
conclude that these groups (1, superior, including SL 6 and SL 7; and 2, inferior, including SL 9)
were o f different andesites from different quarries, and that they correspond to different phases
of sculptural activity at San Lorenzo, each using
material from different sources.18 They also say
that their lower, middle, and upper levels corre-

290

A ncient Am erica

spond to different varieties o f the andesite, which

distinguishes San Lorenzo from the basalt of La
Venta. They identify the red gravel floor as the
red cuaternary form ation. The San Lorenzo
Phase fill for them is a geological stratum o f
compact clay, which in turn underlies a brownish fill (relleno cafetoso). The top layer, which
they cali humitic, is the Villa Alta Phase. They also
stress the periodic burial o f m onum ents as a curious custom peculiar to San Lorenzo.

3. REMEASURED RADIOCARBON
DATES FOR LA VENTA
After recounting and adjusting the original sam
ples from La Venta, Berger, Graham, and H eizer
dated its duration anew as from 1190 90 b . c . to
about 600 b . c ., making it coeval with San Lorenzo
and of longer duration, rather than following after
the abandonm ent o f San Lorenzo.19 The authors
believe20 that the colossal heads from San Lorenzo
and La V enta are o f the same age near 600 b . c . 21
This chronological pairing o f San Lorenzo with
La V enta brings new relevance to the dissimilarities mapped by Miyawaki that will be discussed in
section 6.22 As we will see, the Berkeley data of
Clewlow et al.23 support the separation of San
Lorenzo, Tres Zapotes, and La V enta as being distinct and w ithout overlap, while the Yale data24
support the coeval existence o f the three sites in an
interchange o f stylistic traits. N either interpreta
tion, however, can be made to supply any tem
poral sequence w ithout further inputs, which have
been lacking until recently.

4. THE HEAD AT COBATA

The largest o f all known heads (3.30 m. high) was
found in 1970 facing south near the surface on the
flank o f C erro El Vigia at the Rancho de Cobata.25
Pieces o f Late Classic pottery appeared at its base
to a depth o f 70 cm. A fine-orange pate containing
an obsidian blade carne into view during excavadon at the left of the mouth. The association
proves only a possible resetting in Late Classic
time. R. K rotser, who was present when the head
was found, has been quoted in a letter to the author from F. Clancy as saying that the head was in

a wash, almost a barranca, and that the pottery

evidence can not be trusted for dating.
The head was probably made from one of the
huge, round, smooth-faced boulders described
by Williams and H eizer as having been weathered spheroidally on the profusely littered up
per slopes of El Vigia, cise to the summit.26
These were rolled or dragged down slope be
fore carving. The same basalt was used for Tres
Zapotes head no. 2 (hereafter T Z 2, but also called
Cerro N estepe, or NS) and the Tres Zapotes head
no. 1 (hereafter T Z l) .27

a. The Macrotechnic Scenario

The only heads weighing twenty tons or m ore are
the Cobata head, which is the largest, followed by
San Lorenzo head no. 1 (25.3 short tons), La Venta
head no. 1 (24 tons; fig. III-207), and San Lorenzo
head no. 2 (20 tons). Tolstoy suggests that weights
and sizes of colossal heads display coherent
trends and may be useful as chronological
guides.28 But for this writer, thefact that each site
(SL, LV) other than Cobata also has one very large
head suggests that its temporal position is near the
beginning o f the series at the respective sites, ac
cording to an argum ent that proceeds as follows.29
Lithic technology, being limited to nonmetallic
instrum ents, required extremely large-scale theaters of operadon, with a m acrotechnic instrum entadon conditioned by the large instruments
and the scale o f the work they could best produce.
The tradition o f shaping stone with stone tools
may inevitably have led to the realization that an
enlargem ent of the work to colossal propordons
was the only way to achieve finely detailed control
over sculpture. Below large work-sizes, stone
tools cannot be made to shape lines or modeling
finer than their own edges. It is like the difficulty
o f drawing a map with a pen coarser than the scale
of representation: the line showing a river is many
times wider than the river itself. Olmec sculptors,
working at the largest available scale, thus learned
lessons analogous to those o f working today under
great magnification. T he mind can witness the ef
fect o f every stroke in a manner otherwise unknowable at ordinary scale. Once achieved, and
codified as a m ode o f representation, the colossal
scale could be reduced progressively, until the
making o f miniatures became possible, rendering

T h e Styles o f the O lm ec Colossal Heads

the colossal proportion unnecessary, and allowing
the production o f innum erable diminutive rep
licas and variations in clay and jades.
b. C o ffee-b ean Eyes
This, however, is not to say that diminutive productions in clay did not precede the colossal stone
heads. O n the contrary, the experim entation at
colossal scale in stone may well have begun with
archaic conventions such as the coffee-bean
eyes we see on the Cobata head and on LV 1 (fig.
III-207), as well as flattened features and spheroid
or conical head shapes, drawn from temporally
primitive conventions o f anatomical modeling in
clay figurines. Those o f the San Lorenzo Phase,
according to Coe, resem ble Las Bocas clay figures
from M orelos in the central Mexican highlands, as
to hollow and solid white-ware figures, with
eyes, always indicated by shallow troughs, but
lacking the punching so frequent in La V enta
figurines.30
c. Colossal H ead Sizes
Tolstoys view31 is identical with that o f Clewlow
et al.32 as to intrasite subgroupings, but his tem
poral sequence differs from theirs. Tolstoy agrees
with Wicke in supposing that LV 1, LV 4, T Z 1,
and TZ 2 (also called NS) are the m ore recent
end o f the series, and he thinks increasing head
dimensions are better guides to chronology than
the ratios between dimensions used by Clewlow
and his colleagues.33 Thus he reaches the position
that SL 2 and SL 4 (fig. 111-208) are earlier than SL
1 and SL 5. H e further affirms W ickes thesis that
LV 3, LV 2, and LV 4 are in chronological order
from east to west at the north end o f Complex A.
These laconic clues allow the following order as
approximating Tolstoys tem poral seriation:
most recent LV 1 (fig. III-207), LV 4, T Z 1,
TZ 2
less recent SL 1, SL 5
older SL 2, SL 4 (fig. III-208)
This departs from W ickes only as to the ordering
at San Lorenzo:34
Wicke: SL 1, SL 3, SL 4, SL 5, SL 2
Tolstoy: SL 2, SL 4, SL 1, SL 5
(Clewlow et al. are not compared because they
eschew temporal ordering.)

291

A nother grouping by three clusters resembles

the multidimensional mapping that will be discussed below. Tolstoy maps the clustered likenesses as follows,
SL 3, SL 6, SL 4;
SL 5, LV 2, LV 3;
SL 2, SL 1, LV 1, LV 4
in a link diagram spaced proportionally to likenesses. N o temporal sequence is suggested by
Tolstoy for this diagram.
5. G U T T M A N SC A L IN G A N D
S E R IA T IO N
Having prepared a G uttm an scalogram, Wicke at
first accepted the scale from left to right as an
expression o f sequential developm ent through
time involving the stylistic factor.35 Later on36 he
reversed his position, to follow the new Coe seria
tion,3 as SLLV * TZ. In abandoning the origi
nal sequence to TZ LV SL, as first proposed
by K ubler in 1959,38 Wicke correctly observed, as
had Graves, Graves, and K obrin,39 that scaling
cannot tell us that one end is later and the other
earlier, w ithout external evidence for the direction of reading.
W ickes diagram shows his revised sequence for
ten heads in two columns o f five heads each.40 TZ
and LV form the right-hand column, and SL heads
are on the left. The arrows mark his seriation as SL
1, SL 3, SL 4, SL 5, SL 2, followed by LV 3, LV 2,
LV 4, LV 1, closing with T Z 2.
At present writing (April 1975), the main positions on tem poral sequence are outlined in the
table below. Both Tolstoy and Fuente adhere to
Coes second sequence.41

Oldest
Kubler (1962)
Wicke (1965)
I Coe (before 1968)42
II Coe (after 1968)43

TZ
TZ
LV
SL

Intermedate
Kubler
Wicke
I Coe
II Coe

LV
LV
SL
LV

292

A ncient Am erica
M o st R e c e n t

Kubler
Wicke
I Coe
II Coe

SL
SL
TZ
TZ

6. M U L T ID IM E N S IO N A L M A P P IN G A N D
S T Y L IS T IC D IST A N C E S
Miyawaki, on K ublers request, prepared under L.
J. Savages direction a multidimensional scaling at
the Yale C om puter C enter, in order to map the
dissimilarities among the heads.44 H e used the
three available sets o f data: that proposed in 1939
by Kubler; that revised and augmented by Kubler
in 1970;45 and the data published by Clewlow et
al. at Berkeley.46
The main difference between the data from
Berkeley and that from Yale is that the latter are
restricted to anatomical and morphological traits,
while the Berkeley traits include ornaments, costumes, geological sources, weights, dimensions,
and defacements.
K ublers assumption, in both his data lists, was
that body parts and stone shapes would be more
revealing of chronological position than other
characteristics which he expected would reveal
only geographical and stylistic differences. In ef-

feet, the Berkeley investigators were unable to

find any classification other than clustering the
twelve heads in six groups by three sites, 'with
absolutely no temporal implications, as A (La
Venta), B (San Lorenzo), and C (Tres Zapotes).47
Miyawaki prepared eleven maps showing the
dissimilarities between the heads as geometrical
distances in relation to others, calculated according to five different functions (two were necessary
for K ublers revised data). Each o f the eleven maps
is characterized by a num ber called the stress,
which measures the im perfection o f the best possible portrayal (a perfect fit having a stress of zero).
For two dimensions (selected from among computations in one, three, and four dimensions), the
Berkeley data in the fifth function showed stress
of 0.167, while K ublers first data48 showed more,
with 0.187. But K ublers revised data were down
to 0.090, which is the lowest stress (or best por
trayal) among the eleven maps in two dimensions.
Maps 3 (Berkeley data; and 11 (Yale data) accordingly portray the differences among twelve
heads in different ways. The Berkeley data sep
rate the three sites perfectly w ithout overlapping

oLV4
O

LV3

TZ1

Key TZ ~ Tros Zapotes

NS _ Cerro Nestepe
LV " U) Venta

NS1

LV1

SL6

# S L1

LV2

SL2

SLS
SL6

TZ1

O
LV

SLS

O
LV1

SL4

0
SU

SL2

SL5

SL4

T he Styles o f the Olm ec Colossal Heads

or scatter; the territory o f each is unequivocally
marked. The Yale data do not separate the sites,
but they do separate the heads by styles of sculp
ture in undefined temporal direction.
Map 11 (Yale data) is based on thirty-six traits
combined as a variable head shape taking vales 1,
2, and 3; and a m outh shape taking vales 1, 2, 3,
and 4. Miyawaki cautions that the portrayal must
not be taken too seriously. The procedure reflects
simply the geom etrical structure of the prepared
data, and the results depend strongly on the judgments made in preparing the data.49
Map 3 (Berkeley data) separates La V enta heads
from those o f San Lorenzo along a diagonal from
lower left to upper right. Tres Zapotes heads occupy the top left crner. T he least distances sepa
rate the heads from the same sites (intrasite dis
tances); the greatest distances separate Tres Za
potes heads from those o f San Lorenzo (intersite
distances).
Map 11 (Yale data) shows less stress than Map 3
in respect to the 1 correlation coeffcient, or
m easurem ent o f association on a relative basis,
meaning that it is a better portrayal of its data. The
sites are not separated on Map 11, but the stylistic
groups are portrayed at the expense of geographic
distinctions. Styles are differentiated rather than
places. T he differences portrayed are only those of
anatomy and shape.
The greatest distance separates SL 4 (fig.
III-208) from LV 4, and it is best seen in the profile views as the distance between a spheroidal
boulder (LV 4) and a slablike ovoid (SL 4).
The least distance separates SL 1 from SL 5: in
frontal view both are oblongs with rounded ends,
depicting supple flesh and correct anatomy.
A stylistic area of cise resemblance is por
trayed among LV 2, SL 2 and T Z 1. In these heads
the frontal aspect is spheroidal and individual in
expression. The distances in this area are admittedly implicit in K ublers data, but the Computer
mapping in two dimensions makes unexpected resemblances visible w here they escaped notice
before.
Also interesting is the proximity o f T Z 2 (or
NS) to SL 1, SL 5, and SL 6, especially as to resemblances in conical profile, sloping more at the
face than at the back. A nother such grouping ap

293

pears among T Z 1, SL2, SL 3,and SL 4, where face

and back are closer to being parallel planes.
The most instructive distancing, however, is the
wide separation among La V enta heads, contrasting with the cise intervals among most San Lor
enzo heads. These variations may correspond (1)
to wider differences among La V enta sculptors, or
(2) to longer intervals of time between their surviving works, or again (3) to the presumed disappearance of many heads. At San Lorenzo more
heads survive to fill the record, but work there
may also have been produced more rapidly, by
fewer workshops, and in closer temporal texture
of tradition.
For the art historian at least, several major surprises em erge from these seemingly irreconcilable
approaches by social scientists and by humanists.50
1. M ultidimensional scaling and its many relatives51 can be compared to a statistical camera
by recording as transformations into numbers the
exact configurations o f the data, whatever their
origin and bias.
2. The different portrayals o f the same subject
from different angles are not inconsistent but
complementary. Seemingly contradictory, they
are mutually enlightening, as here, in respect to
place and style.
3. Such scaling by dissimilarities is useful in revealing unexpected likenesses or contrasts and oppositions or concordances. N o t only are spatial
relations m arked out, but the velocities and densities are adum brated in the fabric of time ve
locities and densities that ordinarily are apprehended only by archival study o f artisans lives
and of their guilds and patrons.
4. Qualitative analysis is an arduous operation of
known difficulty, but once achieved, its statistical
representation and portrayal are simple, if laborious, procedures.

SUMMARY
By assembling various recent discoveries, methods, and theories in the study of Olmec colossalhead sculpture, it has been possible here to suggest the outlines o f a new approach to its stylistic
history.
The first requirem ent was to establish the possi-

294

A ncien t Am erica

ble simultaneity o f those sites w here the heads are

found. Reviewing the evidence for temporal se
quence at San Lorenzo and for the parallelism of
the histories o f the three principal sites, it is suggested that they all had roughly coeval beginnings
in the production of colossal heads.
The largest o f them all at Cobata may be earliest,
on the strength o f comparison as to conventions
for the representation o f eyes shared by the C o
bata head with Las Bocas figurines and on the assumption that large sculpture perm itted great con
trol, at a very early date, over anatomical rendering by sculptors who thereafter were able to
reduce the size o f their work.
Tolstoys proposal that ranked weights and sizes
of heads might be useful as a guide to chronology.
It was adapted here by reversing the temporal direction o f his table, which depends on a scaling
technique without clear time direction.
In the absence o f adequate evidence, the pro
posal o f still another one-track alignment for all
the heads is avoided here. Instead, multidimensional scaling offers a m ore ampie possibility for
exact knowledge. This model assumes the sites to
be coeval and interrelated by traveling crews of
sculptors, and to be participating in a simple net
work of very early artistic influences. These assumptions are supported by com plementary Com
puter mappings o f high fidelity to the data, as
separately formulated at Berkeley and at Yale.
The findings are initial ones o f m odest scope,
but we may hope for m ore by increasing the descriptive com pleteness o f the data when still more
heads are brought to light. It is admitted that the
present descriptions are still incomplete.
In detail, SL 1 and SL 5 are cise in style, like LV
2, SL 2, and T Z 1 in another period. Farthest apart
are LV 4 and SL 4 (fig. III-208). Larger groups,
defined by resemblance, appear among LV 2, SL 2
and T Z 1, as well as among T Z 2, SL 1, SL5, and SL
6, and among T Z 1, SL 2, SL 3, and SL 4, these
groupings all varying as to the geom etric and ex
pressive data. Finally, the gaps among the known
heads suggest differences o f school o r o f survival
or of local abundance before depletion.

CONCLUSION
The two maps display in effect the different aims

o f an anthropological archaeology (Berkeley) and

o f a humanistic history o f art (Yale). N either aim is
wrong; neither aim precludes the other; neither
aim is com plete. Both aims can be interpreted as
com plem entary52 rather than as contradictory
statements. The anthropologist seeks quantitative
completeness and objectivity; the art historian addresses qualitative problem s o f form and ex
pression. Each needs the other more than either
realizes.
NOTES
1 G. K ubler, The A rt and Architecture of Ancient America
(Baltim ore, 1962); C. R. W icke, Olmec: An Early A rt Style
of Pre-Columbian Mxico (Tucson, 1971); C. W. Clewlow,
Jr., R. A. Cowan, J. F. O Connell, and C. Benem ann, The
Colossal Heads of the Olmec Culture, Contributions o f the
University o f California Archaeological Research Facility,
IV (Berkeley, 1967); M. D. Coe, T he Archaeological
Sequence at San Lorenzo Tenochtitln, Veracruz, Mx
ico, Contributions of the University of California Archae
ological Research Facility. V III (1970), 2134; P. Tolstoy,
Review o f Olmec: An Early A rt Style of Pre-Columbian
Mxico by Charles R. W icke," American Journal of Archae
ology, LX XV I (1972), 4 5 7 -5 9 ; C. W. Clewlow, Jr., A Stylistic and Chronological Study of Olmec Monumental Sculp
ture, C ontributions o f the University o f California Ar
chaeological Research Facility, X IX (Berkeley, 1974); B.
de la Fuente, La cabeza colosal de Cobata, Anales del
Instituto de Investigaciones Estticas, Mxico, X II, no. 43
(1974), 4 9 -5 6 .
2 M. D. Coe, Preliminary R eport on the First Seasons
W ork at San Lorenzo Tenochtitln, V eracruz, mimeographed (N ew H aven: Yale University, 1966); La
segunda tem porada en San Lorenzo Tenochtitln, Ve
racruz, Boletn del Instituto Nacional de Antropologa e His
toria, Mxico, X X V III (1967), 110; San Lorenzo and
the O lm ec Civilization, in Dumbarton Oaks Conference on
the Olmec, ed. E. P. Benson (W ashington, 1968), pp. 4 1 78; Archaeological Sequence; and M. D. Coe, R. A.
Diehl, and M. Stuiver, O lm ec Civilization, Veracruz,
Mxico: Dating o f the San Lorenzo Phase, Science, CLV
(1967), 1 3 9 9 -1401.
3 Coe, Diehl, and Stuiver, D ating, p. 1939.
4 Coe, O lm ec Civilization, pp. 6869.
5 Coe, Archaeological Sequence, p. 21.
6 Coe, O lmec Civilization, p. 68.
7 Ibid., p. 69.
8 Ibid., p. 48.
9 Coe, Segunda tem porada, p. 21.
10 Ibid., p. 25.
11 Ibid., p. 23.
12 Coe, Archaeological Sequence, p. 28.

T he Styles o f the Olm ec Colossal Heads

13
14
15
16

17
18
19

20
21

22
23
24
25
26

27
28
29

30
31
32
33
34

Coe, O lm ec Civilization, p. 52.

Coe, Segunda tem porada, p. 21.
Coe, O lm ec Civilization, p. 52.
J. Brggemann and M.-A. H ers, Exploraciones ar
queolgicas en San Lorenzo Tenochtitln, Boletn del In
stituto Nacional de Antropologa e Historia. Mxico.
X X X IX (1970), 18 -2 3 .
Ibid.
Ibid.
R. Berger, J. A. G raham , and R. F. H eizer, A R econ
sid eraron o f the La V enta Site, Contributions of the U ni
versity of California Archaeological Research Facility, III
(1967), 5.
Ibid., p. 7.
P. D rucker, R. F. H eizer, and R. J. Squier, Excavations at
La Venta, Tabasco, Smithsonian Institution, Bureau o f
American Ethnology, Bulletin no. 170 (W ashington,
1959).
T. Miyawaki, M ultidimensional Scaling o f the Olmec
H eads, MS (New H aven: Yale University, 1970).
Clewlow, Cowan, O Connell, and Benem ann, Colossal
Heads.
Kubler, Ancient America.
Fuente, C obata, p. 124.
H. Williams and R. F. H eizer, Sources o f Rocks Used in
Olmec M onum ents, Contributions of the University of
California Archaeological Research Facility, 1 (1965), 4.
Ibid., p. 15.
Tolstoy, Review, p. 459.
G. Kubler, in consultation with T. Miyawaki, unpublished
notes (New H aven, 1970). Also, Early A rchitecture and
Sculpture in Mesoamerica: Com m entary on a Paper by T.
Proskouriakoff, in Observations on the Emergence of C ivi
lization in Mesoamerica, Wenner-Gren Foundation for A n
thropological Research Symposium. Burg Wartenstein, 1970,
ed. R. F. H eizer and J. A. G raham (Berkeley, 1971), pp.
157-58.
Coe, O lmec Civilization, p. 46.
Tolstoy, "Review, p. 458.
Clewlow, Cowan, O Connell, and Benem ann, Colossal
Heads, p. 58.
Tolstoy, Review, p. 458.
Wicke, Olmec, p. 70.

295

35 Wicke, Olmec: An Early A rt Style o f Pre-Columbian

M xico (Dissertation: University o f A rizona at Tucson,
1965), p. 133.
36 Wicke, O lm ec, pp. 12526.
37 Coe, Diehl, and Stuiver, D ating, p. 1400.
38 K ubler, Ancient America, p. 67.
39 T. D. Graves, N. B. Graves, and M. J. Kobrin, Historical
Inferences from G uttm an Scales: T he R eturn o f AgeArea Magic? Current Anthropology, X (1969), 338.
40 Wicke, Olmec, p. 70.
41 Tolstoy, Review; and Fuente, Cobata.
42 Coe to Bernal, cited in I. Bernal, The Olmec World, trans.
D. H eyden and F. Horcasitas (Berkeley, 1969), p. 57.
43 Coe, Diehl, and Stuiver, Dating, p. 1400.
44 Miyawaki, M ultidimensional Scaling.
45 K ubler, in consultation with Miyawaki, unpublished
notes (N ew H aven, 1970).
46 Clewlow, Cowan, O Connell, and Benem ann, Colossal
Heads.
47 Ibid., p. 58.
48 K ubler, Ancient America.
49 Miyawaki, M ultidimensional Scaling, p. 2.
50 C. A. Jackson, M ultidimensional Scaling and the Seria
tion o f the Twelve O lm ec Colossal H eads, MS (Berke
ley: University o f California, 1971).
51 J. B. Kruskal, M ultidimensional Scaling by Optimizing
G oodness to Fit to a N onm etric H ypothesis, Psychometrika, X X IX (1964), 11529; L. G uttm an, A Basis for
Scaling Q ualitative D a t a American Sociological Review, IX
(1944), 13950; R. L. Carneiro, Scale Analysis as an
Instrum ent for the Study o f Cultural Evolution, Southwestern Journal of Anthropology, X V III (1962), 14969;
W. H. G oodenough, Some Applications o f G uttm an
Scale Analysis to Ethnography and C ulture Theory,"
Southwestern Journal of Anthropology, X IX (1963), 235
39; G raves, Graves, and K obrin, "H istorical Inferences,
pp. 3 1 7 -3 8 .
52 G. H olton, T he Roots o f Com plem entarity, Daedalus,
X C IX (19^0), 1018.
[[The difficulty here, as T. Miyawaki often said, was the lack of
any markers for the direction o f the sequence in any of the
sets. T he placing o f the Cobata head is still in question. gk J

III. 10
Vicos Idea of America

VICO S SOURCES ON ANCIENT

AMERICA
Precolumbian America is not usually evoked by
Giambattista Vicos ame, probably because he
m entioned the N ew W orld no m ore often than in
twenty-six passages o f the La scienza nuova seconda
in 1730. And he referred only ten times and
m ore vaguely to the w estern hem isphere in the
Principj di una scienza nuova o f 1725, so that we
may suppose that his m ore detailed knowledge
dates from the five-year interval between the first
N ew Science of 1725 and the second New Science
of 1730.1
Even in 1730, the readings about America cited
by Vico are not very many. H e acknowledged
Joest Lipss commentary on Tacitus,2 which con
tained remarks on ancient America that were
im portant in suggesting to him the parallels Lips
had noted between ancient Americans and the
G erm n tribes o f the time o f Tacitus. Gonzalo
Fernndez de O viedo furnished him with information about the Caribbean peoples.3 The first chapter of the ffth book of O viedos Historia natural
treats the religious images, songs, and dances wth
which the Indians ritually reenacted their traditional history, and it was the only chapter that
would have been relevant to Vicos purposes. An
Italian translation by Ramusio had appeared in
Della navigationi o f 1555.4 Vico also knew but
denied the thesis o f A ntoine Arnauld that the peo
ples in the Antilles lived without knowledge of
G od.5
On Mxico and Per, Vico knew the work of
the Jesuit chronicler Jos de Acosta, whose Histo
ria natural y moral de las Indias o f 15 90 appeared in
Italian at Venice in 1596.6 And for the peoples of

New France, or Caada, Vico drew on another

Jesuit, Joseph-Fran^ois Lafitau,7 whose Moeurs des
sauvages ameriquains appeared in time to supply
Vico with the thesis that its peoples were cut off
from all other nations. It was Lafitaus view, still
acceptable today, that America was settled from
Asia via the Bering Strait and separated from its
origins by the early destruction o f that land bridge.
A nother o f Vicos literary m entors was Marc
Lescarbot de V ervins,8 who w rote as an eyewitness to the eating o f human flesh; and still an
other was Jean Chassognon de M onistrol, who
provided Vico with his knowledge of the Patagonian giants.9 O n the Indians o f eastern N orth
America, Vico knew ofT hom as H ariots work on
Virginia,10 which was available to him in a Latin
versin by Theodore de B ry,11 and he knew of
Richard W hitbournes Discourse on Newfoundland, a continuation o f the same series appearing
in 1628.12 But the latter apparently contains nothing relevant to Vicos interests.
These sources exhaust the American references
he could rem em ber; other references by him to
America are incomplete, although they seem to
come from the sources just m entioned rather than
from oral tradition or unpublished erudition.

ANCIENT AMERICA IN THE NEW

SCIENCE
The twenty-six references to America in Vicos
second N ew Science o f 1730 are distributed in all
parts o f the w o rk .13 Two appear in book I, Establishment o f Principies; four in book II, Poetic
W isdom; five in book III, Discovery o f the True
H om er; fourteen in book IV, The Course of
N ations; and one in book V, The Recurrence of
296

V ico s Idea o f Am erica

Human Things in the Resurgence o f the N ations.
The Conclusin lacks any reference to America,
and it is not m entioned in the Idea o f the W ork
at the beginning, w here the allegorical engraving is
explained. The num erous references o f 1730 contrast with those o f the first New Science o f 1725,
which contained only one allusion to America,
taken from Lafitau,14 and bearing on hieroglyphic
writing. Thus Vicos interest in America greatly
expanded during the writing of the second New
Science (after O ctober 1725 and before O ctober
1729),15 and it affected the scope o f the examples
and dem onstrations o f every part of the work excepting the Idea and the Conclusin. The idea
of America is part o f Vicos intendon. H e described the second N ew Science as composed
and ordered by a single spirit and by consecutive
order in a com plete rewriting that preserved
from the first N ew Science only three passages
from book III.16

V IC O S A M E R IC A N T H E M E S
The twenty-six new references to America in 1730
(see note 13) allude to five different themes corresponding roughly to the five books. The first
theme is the subject o f the Patagonian giants,
which first appears in book I and reappears in
books II, III, and V. The second them e bears on
American beliefs in God and the immortality of
the soul, and it is restricted to book II. The third
theme deais with American hieroglyphs in books
II and IV. The fourth them e concerns ancient
American societies in book IV. The fifth theme
appears only once in book V, and it affirms the
corsi and ricorsi in ancient American tribal history.
The whole system o f these references and allusions reinforces two o f Vicos assumptions: first,
that until two centuries ago, America remained
hidden from the rest of the w orld (par. 517), and
second, that the American peoples before the discovery, though unknown to the Od W orld, fol
lowed the natural course o f the lives of na
tions . . . according to the natural succession of
human things (par. 1020). In this way Vico estabIished American arguments for independent proof
of the principies o f the N ew Science.

297

T H E M E I. T H E P A T A G O N IA N S
Vicos initial m ention o f America (in par. 89) was
to show the unreality of the geography o f H om er,
to whom the Phoenician island o f Calypso, called
Ogyia, was so distant from G reece that even the
winged god M ercury had difficulty reaching it, as
if it were as far from G reece . . . as America is
from our world. Vico thus established both the
fictional nature of H om eric distances as well as the
superlative real rem oteness of America.
Soon thereafter Vico introduced the first o f his
American themes with the mention of the monstrous and fierce creatures which travelers claim to
have seen at the foot o f America, in the country of
the so-called Patagones [= Big Feet} (par. 170),
who are discussed in five different passages (see
also Vico 1725, pars. 101, 211).
The Patagonian giants, Vico believed, were related to the giants scattered over the rest o f the
earth after the Flood (pars. 3 6 9 -7 0 ), in whom
monstrous savagery and unbridled bestial freedom were tamed by the fear o f some divinity
(par. 388). The Patagonians were of enormous
build, like the ancient Germans o f Tacitus, whose
size was owing to savage education as children
(par. 170).
Vico implied that the Patagonians, like the an
cient Germans, took the ame of mortal gods to
distinguish themselves from the immortal gods.
They were the race of the giants, who were chil
dren of mixed marriages between the sons of God
and the daughters of men, and who grew up in
neglect w ithout cleanliness or fear (pars. 36973).
Their great stature carne also from absorbing nitrous salts into their bodies . . . quite without that
fear of gods, fathers, and teachers which chills and
benumbs even the m ost exuberant in childhood.
The imagined link among the Germans of Tacitus,
the G oths of Procopius, and the Patagonians near
the Strait o f Magellan (par. 369) was im portant to
Vico as a link between sacred and profane histo
ries of the w orld.17
H enee the heroes of the next age were awkward and wild because of their recent gigande
origin, being very limited in understanding but
endowed with the vastest imaginadons and the
most violent passions, resembling not only such

298

A ncient Am erica

Hom eric heroes as Achilles and M enelaus, but

also the stubborn peasants o f Vicos own eighteenth-century N eapolitan countryside (par. 708).

THEME II. BELIEFS IN GOD A N D THE

IMMORTALITY OF THE SOUL
The context in Vicos system for this them e rests
upon the axiom that uniform ideas, born among
peoples unknown to each other, must have a com
mon ground o f tru th (par. 144). From this, Vico
observes that all nations, barbarous as well as civilized, though separately founded because rem te
from each other in time and space, keep these
three customs: all have some religin, all contract
solemn marriages, and all bury their dead (par.
333). H e concludes that these three eternal and
universal customs are the three first principies
of this Science. H e then asserts the existence of
only four primary religions (par. 334):
1 and 2. Among the H ebrew s and Christians,
the belief in the divinity o f an
infinite free m ind.
3.
Among gentiles [= pagans], the belief
in the divinity o f a plurality of gods
each imagined as composed o f body
and of free m ind.
4.
Among M ohammedans, the belief in
the divinity o f one god, an infinite
free mind in an infinite body.
The peoples o f ancient America, Vico believed,
observed the third primary religin that of the
gentile pagans in the ancient world. H e absolutely
rejected the assertion o f certain travelers that the
peoples of Brazil, South Africa [Vicos Cafra],
and other nations o f the N ew World live in society
without any knowledge of God, as A ntoine Arnauld had believed to be the case of the inhabitants
of the islands called Antilles (see note 5).
Finally, as to the treatm ent of the dead, Vico
cited Acosta, H ariot, and W hitbourne (notes 6,
10, 11, and 12) in showing that in ancient America
the gentile nations concurred that souls . . . do
not die with their bodies but are im mortal (par.
337), and this among the peoples of Mxico and
Per, as well as in Virginia and N ew England.

The sixteenth-century Jesuit historian Jos de

Acosta was Vicos principal source on ancient
American religin (note 6). In book V, Acosta
distinguished the following kinds o f worship
(idolatry) among the Peruvians:
A. Natural objects:
1. general: sun, moon, fire, earth, the ele
m ents (chap. 4)
2. particular or specific objects: rivers, fountains, trees, mountains (chap. 5)
B. H um an inventions or fictions:
3. artifacts o f wood, stone, or gold (chap. 9)
4. the bodies o f dead persons (chaps. 68).
Like Vico much later, Acosta classified these
modes o f worship by citing their parallels in classi
cal antiquity with the worship o f artifacts in wood
or stone or gold o f pura arte e invencin portraying M ercury and Pallas (chap. 2).

THEME III. THE AMERICAN

HIEROGLYPHS IN THE PRIME VAL AGE
OF THE GODS
Because it was an axiom o f Vicos that hiero
glyphic speech is a speech by physical things, as
in the new India [= Mxico] (par. 226), he was
led to refute the false opinion . . . that the hieroglyphs were invented by philosophers to conceal
in them their mysteries of lofty esoteric wisdom
(par. 435). In this Vico followed Bacon,18 and he
anticipated the recent historical thesis on an
cient Maya writing first set forth in 1960 by T.
Proskouriakoff.19
H enee the hieroglyphs were about the gods and
about family histories. As divine poetry they
were the first wisdom o f the gentile world, as
now confirmed by the American Indians who cali
gods all the things that surpass their small understanding, sharing this habit with the ancient
Germans dwelling about the Arctic O cean (par.
375). This repeated linking o f American Indians
and ancient Germ ans was suggested to Vico by the
commentary o f Lips on Tacitus.20 Elsewhere Vico
(par. 437; and 1725, par. 255) added theG reeks to
these peoples who make a god o f everything that
exceeds their limited understanding, in poetry

V ico s Idea o f Am erica

resembling the play o f children (par. 185). This
interpretation reappears in m odern anthropological theories o f art as play.
As family histories Vico compared the hieroglyphs of Mxico to the grave stelae o f classical
antiquity, called cippi, as in J. Laet and G. F.
Gem elli.21 Gemelli had been cited by Vico in the
first N ew Science,22 whereas in 1730 the com
parison (par. 486) was extended to ancient G ermans (par. 470), following Lips. H e further noted
that, as in Rome, the ame o f an American family
extended to all its m em bers and associates (par.
1033).
The topic was made into an historical principie
in par. 841: Barbarous peoples, cut off from all
the other nations o f the world, as were the G ermans and the American Indians, have been found
to preserve in verses the beginnings of their histo
ry. This principie appears in book III (The Discovery o f the True H o m er), where Vico explained H om er as an ideal hero of the G reek
peoples, and that the G reek peoples were themselves H om er, all this in defense o f the idea that
poetic wisdom is the first wisdom o f the gentile
world (par. 375).
T H E M E IV . A M E R IC A N IN D IA N
S O C IE T IE S
Vicos purpose was o f course not to provide full
descriptions o f exotic societies but only to extract
from his reading those notices that might support
the arguments of the N ew Science. H is treatm ent
of Indian society accordingly stressed two topics;
the forms o f worship o f the gods and the meanings
of various social structures.
Ancient G erm ans and American Indians here
alone are contrasted instead o f compared: Peruvian [jc. Mexican?} open-air tem ples to their
sublime god were hills up which one climbed
on either o f two sides by very long stairways; and
all their magnificence consists in their height, and
in taking heaven for god as o f the laws o fjo v e
(par. 47980; 482). The Germans, however, worshipped merely in clearings leveled in the midst
of the forest.
O ther American forms o f worship such as hu
man sacrifice and cannibalism were signifcant for

299

the mortality o f Vicos divine age (par. 517).

The context is Phoenician and Carthaginian, as
well as Germ n and American. Vico concluded
that the age was not innocent, but one o f fanatic
superstition restrained by terror of an imagined
divinity (par. 518).
As to social structure, Vico was concerned
about slave-owning classes in relation to the early
barbarism of patriarchal clans (par. 557). H ere his
American information carne from Acosta,23 who
said that the wealth o f the Incas was in their vassais, who were slaves laboring as the king desired.
Vico extended the argument to two further top
ics the selling o f children and debtors into slavery among Romans and American Indians, al
though he cited no sources for these American
examples o f such heroic natural law (pars. 658,
582).
Several other external traits o f American social
behavior fitted nicely into the second New Sci
ence. Thus Vico noted that in the W est Indies
(that is, Mxico) only the nobles are adorned with
feathers, which was incorporated to heroic emblems o f rank and property in the returned bar
barism o f the M iddle Ages (par. 488; also 1725,
par. 276). Vicos conjoining o f medieval and M ex
ican costumes among the ricorsi is perhaps acci
dental here. A nother such emblem was the long
hair o f nobility in the heroic age, for which Vico
offered no reference in the American case, al
though he may have m eant scalping when used in
punishm ent both among the Persians and the
American Indians (par. 538).
An allusion to royal scepters o f dried snakeskin
among the kings o f the American Indians (par.
542) is used as an example o f the M ental Dictionary, whereby Vico m eant his reductions o f all
languages to unities o f ideas expressed by differ
ent words in each (par. 445; first outlined in the
New Science of 1725).
T H E M E V. A M E R IC A N C O R Sl A N D
R IC O R S I
The penultim ate paragraph o f book V (par. 1095)
was originally composed at greater length than in
the final versin o f 1744,24 which is only three
lines long: Lastly, Crossing the ocean, in the New

300

A ncient America

World the American Indians would now be following this course o f things if they had not been
discovered by the Europeans. Thus the N ew Sci
ence begins and ends with the assumption that
America was hidden from the rest o f the world, in
anticipation o f the tw entieth-century orthodoxy
among Americanists, b estpresented by A. V. Kidder.25 His views were widely shared by Am eri
canists on both sides of the Atlantic until the revival of diffusionist studies inaugurated by C.
H entze and later continued by R. von HeineG eldern and G. Ekholm .26
The second N ew Science o f 1730, however,
continued for another dozen lines, suppressed in
1744. This passage was on the views o f J.-F.
Lafitau,27 the Jesuit missionary who proposed the
ancient Asiatic origin o f the Indians o f America
and their migration via the Bering Strait. But the
passage was later om itted because, as Vico said in
1730, he had not yet seen the book, and the thesis
was too difficult to prove.28 Furtherm ore, we can
see that it was in conflict with the principies o f the
New Science. Vico did not here explain again his
com mitment to the idea that America was hidden
from the world until the Discovery, but this con
flict was surely the reason for deleting Lafitaus
wonderful, prescient, unrefuted guess, which today still is the core o f orthodox Americanist belief.
Between 1730 and 1744, we see that Vico had
become willing to sacrifice Lafitaus independent
American proofs, as if to favor his own idea o f the
unfolding o f natural law in the course and recurrence of things in the history o f nations.

NOTES
1 Giovanni Battista Vico, La scienza nuova prima, ed. Fausto
Nicolini, Scrittori dItalia (Bari: Laterza, 1931), p. 355,
ndice. See also Principj d i una scienza nuova (Naples,
1725), and La scienza nuova seconda, giusta l edizione del
1744, con le varianti delledizione del 1730 e d i due redazioni
intermedie inedite, ed. Fausto Nicolini, 2 vols. (Bari, 1928).
The study by Sergio Landucci, I filosofi e i selvaggi,
15801780 (Bari: Laterza, 1972), contains a chapter
(Lo p erad el tem po, pp. 273332) on Vico and his rela
tionships to the sociology o f primitive peoples and to the
philosophers o f his epoch. H e does not analyze Vicos
thoughts on America other than in relation to early sociological writings.
2 Justus Lipsius, Ivsti Lipsi . . . Opera omnia postremvm ab
ipso avcta et recensita, 4 vols. (Antwerp, 1637), V.

3 G onzalo Fernndez de O viedo y Valds, Historia natural

y general de las Indias (Salamanca, 1547).
4 Giovanni Battista Ramusio, Delle navigationi et viaggi raccolte . . . , 3 vols. (Venice, 1606), III, 53a.
5 Antoine Arnauld, Oeuvres philosophiques . . . (Paris and
Lausanne: S. dArnay, 1 7 7 5 -8 1 ), X X I, 111; X X X I, 274.
6 Jos de Acosta, S. J., Historia natural y moral dlas Indias
[1 5 9 0 ],ed . E. O G orm an (Mxico: Fondo de Cultura Eco
nmica, 1940).
7 Joseph-Franqois Lafitau, Moeurs des sauvages ameriquains
compares aux moeurs des premiers temps, 2 vols. (Paris,
1724).
8 M. Lescarbot de Vervins, Histoire de la Nouvelle Frunce
(Paris, 1618).
9 J. Chassognon de M onistrol, De gigantibus eorumque reliquis . . . (Basel, 1580).
10 Thomas H ariot, A B riefe and T ru Report of the New Found
Land of Virginia (London, 1588).
11 T. de Bry, Historiae Americae Scriptores (Frankfurt, 1590).
12 Fausto Nicolini, Commento storico alia seconda Scienza
nuova, 2 vols. (Roma: Banco di Napoli, 1 9 4 9 -5 0 ), II, 115.
13 Vico, La Scienza nuova seconda giusta l'edizione del 1744.
14 Nicolini, Commento, I, 173.
15 G. B. V ico,T h e Autobiography of Giam battistaV ico, trans.
and ed. M. H. Fisch and T. G. Bergin (Ithaca, 1944), pp.
1 1 -1 9 , 169.
16 Vico, Autobiography, p. 192.
17 H enry P. Adams, The Life and Writings of Giambattista
Vico (London: Alien and Unwin, 1935), pp. 15960.
18 F. Bacon, De dignitate, VI, 1; cited by F. Nicolini, Com
mento, I, 171.
19 T. Proskouriakoff, Historical Implications o f a Pattern of
Dates at Piedras N egras, G uatem ala, American A ntiq
uity, X X V (1960), 4 5 4 -7 5 .
20 Nicolini, Commento, I, 133.
21 Giovanni Francesco G em elli Careri, Giro del Mondo
(Napoli: G iuseppe Roselli, 1699); Joannes de Laet, Novus
orbis (Lige, 1633), p. 241.
22 Vico, Principj [1725], II, 29; Nicolini, Commento. I, 173.
23 Acosta, Historia [1590], VI, chap. 15.
24 Vico, La scienza nuova seconda, II, 153, 266.
25 D irector, Divisin o f H istorical Research, Carnegie Institution o f Washington.
26 C. H e n tz e ,Mythesetsymboleslunaires (Antwerp, 1932);G.
Ekholm and R. H eine-G eldern, Significant Parallels in
the Symbolic Arts o f Southern Asia and Middle A m er
ica, in T he Civilizations of Ancient America: Selected Papers
of the X X IX th International Congress of Americanists, ed.
SolTax (Chicago, 1951), I, 299309; and G. Ekholm, A
Possible Focus o f Asiatic Influence in the Late Classic
Cultures o f M esoamerica, Society for American Archae
ology, Memoirs, no. 9 (1953), 7289.
27 Lafitau, Moeurs.
28 Vico, La scienza nuova seconda, ed. Nicolini, II, paragraph
1401, p. 266.

111.11
Mythological Ancestries in
Classic Maya Inscriptions"

Long-distance calculations here mean those daycounts represented with num bers greater than the
fifth order o f magnitude. W ithin these larger num
bers, historical Maya time is confined to the period
marked by Cycles 8, 9, and 10. Any date preceding
or following that span, which lasted for about
twelve centuries prior to A.D. 900, may be regarded as either mythological and legendary or
prophetic and astrological.
A few Initial Series inscriptions (hereafter cited
as IS) m ention some dates as having occurred prior
to known Maya history. O thers reach millions of
years into time, perhaps in both directions. The
meaning of these long-distance calculations is
poorly understood. Long (1923, p. 66) remarked
on their arithmetic. M orley (193738, vol. IV, p.
273) supposed them to be either mythological or,
more likely, astronom ical. Thom pson (1960, pp.
31 4 -1 6 ) discussed them as a Maya attem pt to
show, perhaps, that infinity has no starting
point.
Since the discoveries by Proskouriakoff in
1960, we now, however, can see that such longterm statements are normally east in the mold of
historical records. As such they resemble the usu
al Initial Series inscriptions which com memorate
dynastic rulers.

Tikal opens with a Calendar Round date (hereafter

cited as CR) followed by four explanatory blocks
including a skeletal head and a monkey skull. Immediately after comes the historie date 9 . 3 . 1 1 . 2 . 6
(or 19), but it is set into a 6 - d 1 as a trem endous
span, written 1 . 1 2 . 1 9 - 9 . 3 . 1 1 . 2 . 1 3 , that is, sixtyfour million tuns, a num ber of the ninth order
(Thompson 1 9 5 0 , p. 3 1 5 ) . The head o f an od man
appears at d 5 and d 7 . It may be the subject of the
clause, which follows a Tikal place-name emblem
at c 4 .1 The final clause contains a head wearing a
serrated eye-patch (g 8), accompanied by titles and
an emblem (g10). This interweaving o f chronological hyperbole and annalistic history appears to
be a constituent tradition in Classic inscriptions,
expressive of legendary antiquity behind present
events.
Stela C at Quirigua (fig. III-127), carved after
a . d . 7 5 0 (Proskouriakoff 1 9 5 0 , p. 1 9 4 , gives
9 . 1 8 . 0 . 0 . 0 2 katuns), may present a legendary
account o f the mythological past, together with a
brief parallel history o f the dynastic rules since the
opening katun o f baktun 9. The east side of Stela C
records events from the fourth millennium b .c . at
the beginning o f recorded Maya time. The IS
marks 1 3 . 0 . 0 . 0 . 0 4 Ahau 8 Cumhu ( 3 1 2 0 b .c . or
a . d . 2 0 1 5 ) . The glyphs at b 8 and b 1 0 are the only
heads which are recognizable as the ames o f p er
sons. T he head at b10 has fangs. All other heads
are either affixes or linked to affixes, like the
winged cauac signs, signifying the end o f a tun
count (Thom pson 1 9 5 0 , p. 4 9 ) at a 7 , a 8, a 1 0 ,
b 1 1, a 12, and a 1 3 .
Only b 8 suggests a legendary or mythological
human, with a protruding lower jaw, the missing

1. MYTHOLOGICAL ANCESTRIES OF
HISTORIC RULERS
Dates expressed in orders higher than the fifth
may of course refer to events in the recorded span
of historical Classic Maya time. Thus Stela 10 (figs.
III-125 and III-126) on the N o rth Acropolis at
301

302

A ncient Am erica

teeth o f od age, an aquiline nose, and a feather

lying horizontally upon the cheek. The inscription
ends at b 15 with the glyphs 6 ben-ich sky (Thom p
son 1 9 6 2 , glyph no. VI. 1 6 8 : 5 6 1 , hereafter cited as
T + glyph no.) instead o f the usual emblem. b 15
signifies possibly that the locus o f the mythical
reign recorded here was the sixth zenith step of
the layered Maya heaven (Thom pson 1 9 7 0 , p.
1 9 5 ) . This glyph also appears four times at Palen
que in the context o f mythologically ancient dates
(Berln 1 9 6 3 , p. 9 2 ) .
The west side o f Stela C is unmistakably histor
ical, opening with an IS inscription o f 9.1.0.0.0 6
Ahau 13 Yaxkin, and the em blem for Quirigua at
d 8 . A title and a ame appear at c 8 and clO
(T203.592), fish-spiderweb, followed by a Dis
tance N um ber (hereafter cited as D N ) marking
345 tuns leading to the Late Classic reign o f TwoLegged Sky, whose ame appears at d 1 3 (Kelley
1 9 6 2 a).
O n Stela F in the front or south row o f the north
group at Quirigua (fig. III- 1 2 8 ) , Two-Legged
Skys life is the continuing subject on the west
face. Clause 1 ames him at a 7 ; clause 2 mentions
his relationship to a ruler at Copan (Kelley 1 9 6 2 b )
at a12-a13. Clause 1 on the east face discusses a
moon-sign lady named at c 9 ; another lady is
named in clause 2 at e l 3 . This clause may end at
d 1 6 , where the end o f kincbiltun 1 3 (Thompson
1 9 5 0 , fig. 2 6 , p. 3 1 5 ) is m entioned, as a nonhistorical statem ent occurring at the cise o f allusions to even larger periods. It is not clear w hether
past or future time is meant. The kincbiltun interval is 3 , 2 0 0 , 0 0 0 tuns. The final blocks ( d 1 7 - d 1 9 )
are eroded.
At Quirigua, on the west side of Stela D (fig.
III- 1 2 9 ) just east o f Stela C, Two-Legged Sky is
the main subject (Kelley 1 9 6 2 a ; K ubler 1 9 6 9 ) .
Clause 1 ( a 1 - a 2 1 ) twice mentions his ame at a 1 8
and a 2 1 . Clause 2 turns the calendar back to an
earlier event in his life, possibly his unin with or
descent from a m em ber o f the moon-sign family
(mentioned at b 2 2 and b 2 6 ) . O n the east face (fig.
III- 1 3 0 ) , clause 1 ( e l - e l 8 a) discusses this person,
whose ame appears at d 1 7 . Clause 2 , opening at
D l 8 a, concerns Two-Legged Sky. Clause 3
( c 2 0 - d 2 1 ) may allude to the ladys most ancient or
divine lineage by mentioning thirteen kincbiltuns

( 1 3 X 3 , 2 0 0 , 0 0 0 tuns). The birth o f ag irl may be

reported in d 2 0 . The fourth clause ( c 2 2 - d 2 2 ) repeats the opening date in a dedicatory-date posi
tion and closes with the ames o f Two-Legged Sky
and his lady.
At Copn, Stela N (fig. I I I - l 3 1 ) opens with a
glorification o f the ruler ( a 1 - b 8 ) . Clause 2 opens
at b 9 with a D N introductory glyph bringing on
the ancient num ber 1 . 2 . 5 . 1 2 . 9 . 0 . 0 . 0 and the CR
o f 1 Ahau 8 Zac (Thom pson 1 9 5 0 , p. 3 1 4 ) . The
eroded closing block is illegible, but it probably
contained part o f the explanation.
Thus these long Distance N um bers, embedded
among otherwise sober annals o f dynastic reigns,
may correspond to a rhetorical convention for
stating the extrem e antiquity of ruling lines,
adorned also with glyphs naming mythological
predecessors and protectors.

2.
DYNASTIC TEMPLES
COMMEMORATING MYTHOLOGICAL
ANCESTORS

At Palenque, the three small buildings east o f the

Palace known as the Tem ples o f the Sun, the
Cross, and the Foliated Cross contain connected
inscriptions conveying a long message about the
ancestry o f a ruler whose deeds are celebrated on
other m onum ents as well. The Tem ple of the
Cross is in every way the dom inant m onum ent in
the group o f three. Facing south, it has the longest
stair, the largest platform, the most intricately
chambered plan, and the longest main tablet in
scription (fig. I I I - l 3 2 ) , with 2 7 9 blocks as against
1 3 3 at the Tem ple o f the Foliated Cross and 1 2 5 at
the Tem ple o f the Sun. Its opening Initial Series
( 1 2 . 1 9 . 1 3 . 4 . 0 ) is the oldest date preserved at Pal
enque, marking some event near the end of the
great cycle (pictun) preceding our own, while the
other two tablets begin at the opening o f the current great cycle (which will end early in the coming
millennium after a . d . 2 0 0 0 ) . At the Tem ple of
the Cross, m oreover, the clauses are extremely
brief, about thirty in num ber, as against fewer but
longer clauses on the other two tablets. The long
est clauses are the nine o f the Tem ple o f the Foli
ated Cross (fig. III- 1 3 3 ) , which divide 1 3 3 blocks.
At the Tem ple o f the Sun (fig. I I I - l 3 4 ) , fourteen

M ythological A ncestries in Classic Maya Inscriptions

clauses divide 125 blocks. Considering the three
tablets as a connected group of inscriptions, we
find a staccato summary at the Tem ple o f the
Cross, and m ore recent and extended commentaries at the other two, with longer clauses at the
Temple of the Foliated Cross than at the Tem ple
of the Sun.
Earlier interpretations are either astronomical
(Ludendorff 1935; Thom pson 1936) or historical
(Seler 19011902; Berlin 1963, 1965; Kelley
1962a), while Spinden (1924) proposed an in
terpretation partly astronomical and party histor
ical. Seler (1901 1902, pp. 1719), Berlin
(1963), Kelley (1962a) and D tting (1965) have
discussed the idea that the three temples each may
com memorate a different deity whose mythologi
cal existence is noted in inscriptions recording
events as occurring from the fourth until the first
millennia b . c . Seler did not try to identify these
gods. Berlin noted a distinctive god-glyph (fig.
III-135) at each tem ple (GI at the Tem ple of the
Cross; G il at the Tem ple o f the Foliated Cross;
GUI at the Tem ple o f the Sun). Kelley believes
that the Tem ple o f the Sun was dedicated in M ex
ican natal-day style to Kukulcan as N ine Wind; the
Temple o f the Foliated Cross to Ah Bolon Tzacab
as O ne Flower, and the Tem ple o f the Sun to Ek
Ahau as Thirteen Death.
N either Berlin (in 1963) or Kelley discussed
the historical dates in Cycle 9 at the Palenque tem
ples. Before Berlin only Spinden was concerned
to separate mythical time from historical events in
the inscriptions (1924, pp. 181 ff.). Berlin later on
(1965) proposed a historical interpretation with
convincing precisin for the Tem ple o f the Cross.
Spinden wanted to prove that the opening IS
inscription of the three sanctuary tablets pertained
to placing the w inter solstice in relation to the
vague year and the apparent long-term reversal of
the seasons between the fourth millennium b . c .
and the first millennium o f our era. H e distinguished m ore usefully, however, between mythi
cal and historical dates, and he regarded the righthand columns as containing im portant historical
dates, like the small blocks in the figural spaces of
the three sanctuary tablets. H e also pointed in
each tablet to the boundary between mythical and
historical time (Spinden 1924, p. 186). At the

303

Temple o f the Foliated Cross (fig. III-l 33) the D N

at the end of the mythical side contines in a CR
expression, 2 Cib 14 Mol, over to the beginning of
the right block. A similar statem ent appears at the
same place on the Tem ple of the Sun (fig. III-134),
marking the entry into history at the top of the
right block as (9.12.18.5.16), implied by 2 Cib 14
Mol, which appears at N 404.
O n the Tem ple of the Cross (fig. III-l 32) Spin
den (1924, pp. 18091) offered two determinations for the right-hand dates; they are mythical in
one case and historical in the other. His purpose
was to fix the solstice. But many parallels on the
other two tablets seem to support an interpreta
tion as mythical time on the left, succeeded by
historical time on the right for the Tem ple of the
Cross as well (Berlin 1965, pp. 33034).
O ther differences o f content distinguish the
three tablets. The Tem ple of the Cross abounds in
upended-frog signs followed by ames, suggesting
a dynastic listing or table (Berlin 1965, p. 337).
Sometimes no appellative accompanies the T740
birth glyph (for example at d 2 ) ; but in one passage the T740 glyph is twice paired (e14, e17) with
the same frog-and-leaf appellative. All told, the
T740 glyph reappears seventeen times (once as
glyph D at Al 1, where, however, it could serve a
double function, both calendrical in the lunar se
ries, and historical as a birth sign for the womans
head at b 11, serving as the 2 c glyph). At the T em
ple o f the Sun (fig. 111-134), however, birth-signs
appear only five times, with appellatives differing
from those of the Tem ple o f the Cross (one occurrence at b10 parallels the Supplementary Series
expression at a 11 , b 1 1 on the Tem ple of the Cross
tablets). At the Tem ple of the Foliated Cross (fig.
I I I - l33), glyph T740 appears as the lunar Supple
mentary Series glyph D at a 10, and once more at
n 2 , associated with a title (Spotted Ahau) which
appears twice at the Tem ple o f the Sun ( d I , k 1).
This title may connect the two tablets o f the Tem
ples of the Foliated Cross and o f the Sun histor
ically, although at the Tem ple of the Foliated
Cross the ame appears within the historical,
right-hand panel, whereas at the Tem ple of the
Sun it is given in the mythological left-hand panel.
Thus the glyph may be a lineage title, as well as an
individual ame. A nother appellative connecting

304

A ncient Am erica

the Tem ple of the Sun and the Tem ple of the Foli
ated Cross (figs. III-133, III-134) is the macaw sign
(T744) at the Tem ple of the Sun (n16) and at the
Temple of the Foliated Cross ( m 7 ) , in both
passages accompanying the Roman-nosed head
(T1010).
The identity o f the heads labeled by Berlin
(1963; 1965, pp. 3 3 8 -3 9 ) as G ods I, II, and III
(T1010; T1030f, g; and T594a) is still uncertain.
Examining these signs (fig. III-135) m ore closely,
one may describe God I as having fish attributes,
specifically gills, applied to a mythological human
head. It is mythological because of the divine
shell-sign inserted into the skull, the seashell ear,
the scroll in the eye, and the rimmed lower eyelid.
These traits resemble Schellhass G od D from the
manuscripts. G od II evokes the Toltec and MayaToltec recum bent figure called Chac Mool; it also
suggests a divine infant with divine shell inserts on
the limbs, wearing a necklace and bracelets. At
Quirigua, its abstract head-form (T1030d) was
proposed by Kelley (1962 b) as the birth sign,
equivalent at Quirigua to T740 (upended frog).
God III usually is glyph T594 resembling a checkerboard with a small affixed head and lordly title.
God I appears twice in a 3-glyph phrase ( d 7 - d 8 ;
d15-d16) and once with a numerical coefficient of
8 ( d 1 1 ) and near the sixth-day glyph on the m ytho
logical left side o f the Tem ple of the Cross tablet
(fig. III-l 32), as well as at b 2 on its west balustrade
(Ruz 1 9 5 8 , p. 8 6 ) .
God II appears at e l on the north balustrade at
the Tem ple o f the Foliated Cross associated with
an IS inscription of 7 . 1 4 . 1 4 . 1 1 . 1 2 5 Eb 5 Kayab
(Ruz 1 9 5 8 , p. 1 4 2 ) , and again on the tablet at d 2
where the God II glyph clearly is the subject, pre
ceded by a possible variant o f the rodent glyph,
signifying here is portrayed (Proskouriakoff
1 9 6 8 ) . The God II glyph itself ends the IS clause
with the date o f 1 . 1 8 . 5 . 4 . 0 in the 3rd millennium
b . c . At l 3 the oxglyph for all three gods appears at
an implicit date in Cycle 9. At N 9 -o l 1 the three
figures are shown in sequence again, as of shortly
before 9 . 1 3 . 0 . 0 . 0 .
At the Tem ple of the Sun (fig. III-134), a head
like that of God I is linked with God III (c6; d6)
on the mythological left side following the title of
Spotted Ahau at d 1. G od III also appears on the

historical side at 0 6 , after Spotted Ahau at o5,

and G 1 appears in the twelfth clause, at r 1 0 , again
in association with Spotted Ahau at r9- Thus the
G I/G II/G III glyphs seem to have historical as well
as mythical meaning, perhaps as priestly titles or
dynastic lines or both, appearingsingly, paired and
as a trio.
T he Tem ple o f the Cross (fig. III-132) is per
haps preem pted as recording the oldest events by
GI without m ention of G il or GUI (Berlin 1965,
p. 339). T he Tem ple o f the Sun (fig. 111-134) mentions only G I and G U I, but the Tem ple o f the
Foliated Cross (fig. III-133) shares all three, with a
preference for G il. Can this distribution relate to
the differing mythical antiquity of three lineages
o f rulers? The male aspect of GI, the female affix
o f G UI, and the infantile aspect of G il might signify a family among supernaturals, as parents with
their child between them. At Palenque they may
have been regarded as the ancestors o f the ruling
dynasty, whose progeny were com memorated at
the three Tem ples o f the Cross, Sun, and Foliated
Cross, and at the Tem ple of Inscriptions.
These three glyphs occur together again at c2
on the stairway tablet (fig. I I I - l36) at H ouse C
(Maudslay 18891902, vol. IV, pl. 23). The in
scription opens with IS 9-8.9.13.0 at a1-b4 fol
lowed by the birth sign (T740) over the head of an
aged and toothless lady at a5. H er ame reappears
at a 6 and c6. The right half o f block a5 contains a
title (T 74.184.42:511 ?) and an emblem. A DN
( 1 2 .9 . 8 ) follows the emblem, marking the begin
ning o f clause 2. Clause 3 contains a title, an
emblem, and the G I/G II/G III cluster at c2, pre
ceding a ame which includes an infixed notation,
11.11.13, at d 2 . Shield-Jaguar and the Yaxchilan
emblem are m entioned at c3. The final clause
again m entions the ame with the inverted ahau
postfix at c6 , whose ancestry may be outlined in
the preceding clauses, as deriving from the ancient
originating generations o f GI, G il, and G III.
Among the 6 2 0 glyph-blocks at the Temple of
the Inscriptions, the reading order o f the tablets
proceeds, as suggested by Thom pson ( 1 9 3 2 , pp.
3 9 3 - 9 4 ) and Berlin ( 1 9 5 1 , p. 1 2 4 ) , from eastern
(fig. III- 1 3 7 ) to middle (fig. I I I - l 3 8 ) to western
panels (fig. III- 1 3 9 ) . The western wall concludes
at t 12 with the Palenque emblem. The east and

M ythological A ncestries in Classic Maya Inscriptions

west panels both contain mythological ames
(Thompson 1 9 6 0 , p. 3 1 4 ) em bedded among peri
ods o f the eighth order of magnitude on the west
panel. B ut the middle panel deais only with events
of the eleventh and twelfth katuns ( a . d . 6 5 2 - 9 2 )
without the large Distance N um bers. First mention o f the current ruler, named Sun-Shield, oc
curs on the east panel at r 1 1. Sun-Shield appears
three times on the middle panel. At g 3 his ame
follows the twelfth katun-e nding glyph. O n the
west panel his ame appears twelve or thirteen
times. As this tem ple may contain the tomb of
Sun-Shield (Kubler 1 9 6 9 , pp. 2 0 2 1 ) , the tablets
would tell o f his mythological ancestry on the
western and the middle ones in the shrine, and of
his deeds on the eastern panel. Two more appearances o f his ame, at c 4 and e 8 (middle), are
accompanied by six appearances o f the beaded
helmet glyph (T678) o f Teotihuacan style. This
occurs also on Lintel 2 ( g 2 ) at Piedras Negras (fig.
I I I - l 5 1 ) , where six kneeling warriors are shown
wearing the helmet. The botun-e nding date o f this
lintel ( 9 - 1 1 . 1 5 . 0 . 0 ) is contem porary with the events o f Sun-Shields life at Palenque, and its fig
ures may reflect the same events that are recorded
at Palenque in the middle Tem ple of Inscriptions
tablet.
The doorjam bs o f Tem ple X V III at Palenque
(fig. III- 1 4 0 ) likewise tell a very ancient story. It
opens at 9 . 1 2 . 6 . 5 . 8 with the birth o f FeatheredSkull ( T 1 0 6 8 at b 1 3 - b 1 4 ) , who was a contem po
rary at Palenque o f B erlins Ruler C ( 1 9 6 8 a , p.
1 4 0 ) . A bout fourteen years later, Feathered-Skull
reappears with Berlins 3 x a tB l8 (Berlin, 1 9 6 3 ) .
The middle glyphs (top o f south jamb) are destroyed, but Feathered-Skull appears again at d 1 6 .
H ere the final clause begins with a back Distance
N um ber about three millennia ( 3 , 0 8 0 tuns) ear
lier. As of this mythological date, several ames
appear, possibly as indicators of the divine lineages to which their appellatives, including Feath
ered-Skull, belonged.
This record recalls the seven lintels o f Structure
12 at Yaxchilan (fig. 111-141). Proskouriakoff regards them (1964, p. 184) as recording legendary
accounts o f the heads o f different lineages in
ten num bered expressions. N o long-distance calculations accompany these records o f the reign o f

305

Bird-Jaguar, which are dated as of 9.4.11.8.16

( a . d . 525). Barthel (1968) has interpreted the
same clauses as relating to eight high priests, and
six celestial or deceased warriors. Either interpretation marks the remotely mythological nature of
the record. q 4 (T1078) repeats g 8 at Tikal on
Stela 10 (fig. III-125). In both passages this head
accompanies the same unidentified emblem. The
parallel returns us to the beginning of this study,
with the renewed assurance that both glyphs refer
to the same mythological personage.
Stela 31 from the N orth Acropolis at Tikal (fig.
I I I - l56) may also belong in this group of inscrip
tions, as com memorations o f historie rulers by listing their mythological ancestors. The incomplete
inscription on the back of Stela 31, fragmentary
only in the bottom rows, is conventional, being
now segmented by sixteen dates spanning
( 8 . l 4 . 0 . 0 . 0 ) - 9 . 0 . 1 0 . 0 . 0 . O ther calendric statements and indicators mark off at least twenty-four
clauses. T hree o f these ( b 1 3 , e 2 5 , and g 8 ) desg
nate, to begin with, the half period o f baktun 1. N o
anniversary or count-back affixes accompany the
baktun glyph in any o f three clauses. If this statem ent refers to the present great cycle which began
in 3 1 2 0 b . c . then it is contem porary with the
mythical era m entioned at Palenque on the open
ing tablet-passages at the Tem ple o f the Sun
( 1 . 1 8 . 5 . 3 . 6 ) and at the Tem ple of the Foliated
Cross ( 1 . 1 8 . 5 . 4 . 0 ) .
At Tikal on Stela 31, the first o f these baktun 1
clauses introduces a series o f nine mythological
heads;2 the second one repeats the first two heads
o f this series; and the third one repeats only the
first head, com pounding it with the baktun glyph
and closing at g 9 with a name-glyph familiar also at
Quirigua: Two-Legged Sky (Kelley 1962). This
ame was earlier m entioned on Stela 31 at b20
immediately before the Tikal emblem at a 2 1 , pos
sibly as the same ruler who is named on Stela 26 (at
a 4 ) , as o f an undeterm ined Early Classic date.
Thom pson proposed (1962, pp. 81, 3 9 3 -9 4 ) that
the jawbone glyph (T846) has calendric meaning,
because of its chuen infix (T520) and its proximity
to cauac, tun, katun, year count glyphs, and CR
dates. Actually, this linkage appears unequivocally
only once, at c l l - D l l , where the jawbone pre
cedes a cauac sign, suggesting the day Eb, with its

306

A ncient Am erica

association of death and calamity (Thompson

1962, p. 81). The other linkages of the jawbone
(T846) allow us to suppose the opposite, that this
skeletal head is noncalendric and that it belongs to
the group of signs signifying human death. The
T846 glyph is rare, occurring only at two places,
Tikal and Tortuguero. M ore common and more
widely distributed is another jawbone glyph
(T590a, b). It appears twice on the great sarcophagus at Palenque (Ruz 1958, fig. 14), where
the context is unmistakably that of the death of
named persons. Berlin (1968, p. 14) noted the
absence of death records in most Classic Maya in
scriptions, although Proskouriakoff (1963, p. 163)
proposed death as the meaning o f glyph T204:575
(serpent-wing over scroll) at Yaxchilan on
Structure 24.
At Tikal on Stela 31, glyph T846 appears nine
times, each time in a different immediate context.
The nearby dates range between 8.14.0.0.0 ( a . d .
317) and 9.0.0.0.0 ( a . d . 435) in Early-Classic
time. At the seventeenth clause (e24-f27), a myth
ological date in baktun 1 (31202725 b . c .) intro
duces the explanation o f five glyphs, containing
T846 at f26.
This same mythological date earlier introduced
nine head-glyphs, naming persons who probably
were spirits or rem te ancestors or both. O f these
nine, only the first two are m entioned again here,
in a section ending at f 2 7 with upright bone
(T571), which reinforces the dea of death.
A third clause m entions baktun 1 at g 8 . H ere
only the first head o f the series o f nine is com
pounded with the baktun glyph, and the clause
ends at g 9 with the Two-Legged Sky glyph. The
next date is an Initial Series marking 9.0.0.0.0
( a . d . 435). It also concerns Two-Legged Sky at
g 17, in a long clause ending with the T846 glyph at
h 2 1 . H enee the same appellative appears in both
mythological and historical contexts, suggesting
that it is a title rather than a ame, and a title
relating historical persons to mythological ancest
ries.
The first linkage o f the jawbone glyph (T846)
is with T617 at a 1 2 - b 1 2 , explaining the opening
Initial Series date (9.0.10.0.0). Barthel (1965, p.
148) explains T 6 17 as a verb (to bury) and as an
emblem o f the underworld. Glyph T617 recurs at

c l 9 linked with a haab-ending, and at g15 linked

with 9.0.0.0.0. In these contexts, death and endings are conveyed. Returning to jawbone (T846),
as at c6-d 6, the linkage is with T849 (vulture?),
which Proskouriakoff identified as a title or clan
symbol (1960, p. 469). At d26-c27 the jawbone
precedes another title resembling the T1030i-n
(batab) group (Kelley 1962b, figs. 2, 3). At
e14-f14, it follows a birth sign (T740; upended
frog). At e26, two mythological heads precede the
jawbone, repeating the first and second o f nine
mythological heads at a14-b14. Therefore, the
jawbone precedes a jaguar head at g 5 - h 5 , perhaps
the same ame as at c5. Finally, at g 2 1-h 21, the
jawbone follows a glyph o f an armed hand atop a
cauac sign, which may be a title, repeated at h 28 as
9.0.3.9.18 ( a .d . 438).
We are far here from having explicated all the
raw significance o f these inscriptions. But the orig
inal purpose of the scholars who composed them
was perhaps to make the meaning clear to the man
in the Street, all while loading the statements with
esoteric meaning. The fact that we, knowing little
or nothing o f ancient language and ritual, can reconstruct the evident outlines o f the gross struc
ture is a testimonial to the clarity and the regularity
of the organization o f Classic Maya writing.
In conclusin, this review o f those Classic Maya
inscriptions which offer legendary and long-dis
tance dates shows that some principal figures of
Maya dynastic history boasted of mythological an
cestries. Such were Sun-Shield at Palenque, T ur
tleshell at Piedras N egras, Bird-Jaguar at Yax
chilan, or Two-Legged Sky at Quirigua. Their
ancestors are cited as flourishing in the most remote times and as occupying another cosmological
stage. Such interpretations o f historical content in
Classic Maya texts are based on textual methods
first proposed in 1960 by Proskouriakoff and by
Berlin in 1965 and 1968. In effect, these readings
fill out the barren chronographic structure of the
inscriptions with the themes o f a dynastic ico
nography o f ruling families, which Maudslay and
Maler first apprehended at the beginning of this
century. The present paper (written in 1970) also
reopens the question o f mythological ancestors,
first suggested by Berlin, writing on the Tem ple of
the Cross at Palenque (1965, p. 339).

M ythological Ancestries in Classic Maya Inscriptions

{Heinrich Berlin kindly read these pages in July
1971, preventing me from numerous mistakes and
omissions.}
NOTES
*Readers unfamiliar with Maya chronology, calendrics, and
its term inology might consult ageneral work like C urt M user,
Facts and Artifacts of Ancient Middle America: A Glossary of
Terms andWords Usedin the Archaeology and A rt History of PreColumbian Mxico and Central America (New York, 1978).
1 Joyce Bailey notes that d5, d7 repeat on Stela 31 at b 19 and

307

FO, as at Uaxactun on a vessel from Burial A31 (Smith

1955, fig. 8g).
2 Barthel (1963, p. 209) suggested that a1 5 -a1 8 recorded
seven primordial earth-rulers (U rsprungsherrscher ber
das E rdreich), but he om itted m ention o f the first two o f
the series o f nine whom we here suggest as mythological
ancestors.
[O f the same date is a microlinguistic approach to the same
subject by F. G. Lounsbury, Some Problem s in the Interpretation o f the Mythological Portion o f the Hieroglyphic
Text of the Tem ple o f the Cross at Palenque, Third Palenque
Round Temple, 1978, v ol. 5, part 2 (1980), pp. 9 9 -1 1 5 . Gk]I

III. 12
The Clauses of
Classic Maya Inscriptions

The Maya wriring format of paired vertical col

umns is usually ftted into a frame. O ften this framing breaks the linear sequence of the reading order
into several short paired columns. Their lengths
are determ ined by the height o f the frame rather
than by the divisions o f the message. To see these
divisions clearly, we need to take the paired col
umns out of their frames and arrange them as a
linear sequence in such a way as to display in Maya
order those segments o f the message which I am
here defining as clauses and captions. T he diagrams (figs. III-145, III-147, III-149, and III-151)
accompanying my examples are composed in lin
ear sequence. I shall om it extended discussion
both of the chronographic expressions and o f the
accompanying explanations, in o rder to focus attention clearly on the clauses and their infernal
divisions.

titled Principies o f Maya Glyphic W riting containsashort section (1950; 1960, p. 43)com paring
the clause to a single sentence. Jones (1964, p. 8)
prefers repeated group to clause. Thom pson
regards Initial Series (hereafter cited as IS), Distance N um bers, and Lunar Series as clauses. I
shall talk o f them as ways to start a clause.
For Thom pson, B eyers glyph pair and glyph
series are also clauses. In effect, any combination of glyphs which is repeated is defined as a
clause (1950; 1960, p. 43). Thom pson also admits long clauses as containing one or more
subordnate clauses, but his longest example (Pal
enque, Tem ple of the Inscriptions, middle, 14-17)
contains only seven glyphs. These are called
blocks. Thom pson ventured no interpretation
of this clause. H e also referred to the Lunar Se
ries as a long sentence dealing with nocturnal
m atters. I shall talk o f it as a way to begin a clause.
In general he left unresolved the difference, if
any, between clauses and sentences. H e was more
certain about subclauses, however, in respect to
glyphs G and F, or B and X within the Lunar Se
ries, which are to be read together. H e also recognized the ben-ich katun (Thompson 1962, glyph
no. 168, hereafter cited as T + glyph no.) at Yaxchiln (1950; 1960, p. 44) as em bedded in a frequent clause (seven examples, 1950; 1960, fig.
46). H e adm itted that the fusin o f several glyphs
into one block might constitute a clause of only
one glyph, because o f fusin. Thus several
glyphs appeared to him capable o f being fused
into one, but he avoided the idea o f a block as
being significant. H e preferred to retain the definition o f a glyph as a single elem ent, a main ele-

1. SENTENCES, CLAUSES, OR PHRASES?

Students o f Classic Maya writing still are not sure
about its nature as a system. Is it ideographic or
homophonic, logographic or pictographic, phonetic or syllabic? The combinations in which its
modes might occur is also uncertain. We continu,
nevertheless, to discuss its divisions as though it all
were punctuated writing, by referring to its sentences, clauses, and phrases.
Uncertainty still reigns about such fundamental
matters as the correct divisin among the parts o f a
long inscription. Usage as to clauses or blocks
is poorly defined in Maya studies. Thom psons
Maya Hieroglyphic Writing: A n Introduction (1950;
1960) nowhere carries in the index any entry for
the words block or clause. The chapter en308

T h e Clauses o f Classic Maya Inscriptions

ment to which have been added various infxes or
affixes (1950; 1960, p. 37). H e also noted that a
glyph block or its half or quarter divisin may contain one or two glyphs, arranged so that the edge of
one impinges on the edge of the second (1950;
1960, p. 37). H ere he supposed without enthusiasm that this welding together o f glyphs was practiced only when both formed part of the same
clause. If this belief were correct, then the block
would have to be accepted as the significant con
tainer of a grouping of glyphic expressions, but
Thom pson him self resists this view when he ob
serves (1950; 1960, p. 41) that overlappingglyphs
are not necessarily . . . related to one another. . . . In effect, many blocks divide (fig.
III-147) between the end o f one clause and the
beginning o f the next, as if to mark the independence of clauses from blocks.
Thom pson (1950; 1960, pp. 5051) was firm,
when writing o f sentences, in saying that the
m onum ents disregard the spoken word, as to distinctions between nouns and verbs, genders, ten
ses, and word order. H e stressed the vague order,
which allowed the reader to rearrange and supplement the glyphs and their affixes, w hether the
reader was a Yucatec, Chol, Tzeltal, or Chontal. . . . The argum ent here is in full support of
his view.
At this point some definitions will be useful.
Classic Maya inscriptions are taken here as texis
consisting o f statem ents explaining chronological
events and captions identifying persons and scenes.
Any day-count statem ent and its explanations
form a clause. Several clauses form a text, by
arranging explanations of dates in some chrono
logical order. Captions are similar to the legends
accompanying plates in m odern books, as expansions upon a writing which surrounds or encompasses the figural m atter. (Jones 1964, p. 12, distinguishes personal and scenic groups of
glyphs.)
The terms used here are chosen to correspond
to the construction of whole texts. This structure
is assumed to consist o f chronographic expres
sions followed by explanations. Chronographic
expressions are Distance N um bers (hereafter cit
ed as D N ), Calendar Round dates (hereafter CR),
Period Endings (hereafter PE), and Cycles. Expla

309

nations are still incompletely understood: their

present range embraces ames, titles, places, actions, colors, directions, and qualities. A date is
any specified position in time. It may be marked
either as a named day in a known period or as a
Period Ending chosen among orders o f magnitude
larger than the day. Distance N um bers by themselves can function as dates because they usually
count from a known date to reach another date.
Thus Berln has spoken of dates as being engendered by Distance N um bers (1965, p. 334). The
distinction is that dates differ from Distance N um
bers much as points differ from lines.

2. A GENERAL RULE FOR SCANNING

ENTIRE INSCRIPTIONS
In Classic Maya inscriptions, the sequence among
time-statements and explanatory remarks has never been discussed with a view to establishing the
intended divisions. Even the most recent studies
on the historical interpretation o f Maya inscrip
tions leave the m atter unresolved. The placing of
historical passages in relation to the dates they explain requires some study of entire inscriptions.
To resolve many existing ambiguities, a simple
and practical rule o f reading order is here proposed. It assumes that time statements normally
precede their explanations.
An illustration is the Tablet o f the 96 Glyphs at
Palenque (fig. III-143). In seeking for the
intended reading order, Berln (1968a) divided it
into nine rows, of which seven begn with Calen
dar Round expressions and five end with Distance
Numbers (fig. III-144). His seventh and ninth
rows lack both CR and D N expressions.1Y et they
too are concerned with giving information about
four rulers at Palenque. By the general rule proposed here, only seven clauses can be defined (fig.
III-145). T he first lacks a D N and opens at Al with
a CR. It ends at b 4 with an appellative. Clause 2
begins with a D N at a 5 , ending at D i, and so forth.
By this rearrangem ent, either a CR or a D N fol
lowed by a CR expression opens each clause, and
every clause ends in this inscription with an appel
lative.
Thus Berlins segmentation brackets the expla
n aron between a prefacing CR and a concluding

310

A ncient Am erica

D N . This is in contrast to my segm entaron, which

opens either with a CR or a D N followed by its CR
(fig. III-145). Each clause therefore concludes
with a ame o r an em blem, but not normally with a
chronogram.
It is assumed here that the num ber o f CR dates
in a text determ ines the num ber o f its clauses and,
conversely, that no clause normally contains more
than one CR date. For example, Relief Panel 1 at
D um barton Oaks (Fig. III-146), which is thought
by Coe and Benson to come from El Cayo (1966,
p. 5), contains seven clauses by our rule (fig.
III-147): (I) an Initial Series ( A l - D 2 a ) giving the
birth date o f an individual whose title or ame at
e l may be Turtleshell; (II) com memoration of the
third anniversary o f Turtleshells accession at
Piedras Negras ( 2 b - F 2 a ) ; (III) m ention o f a ceremony of accession by Turtleshell to rulership
( F 2 b - H 3 b ) ; (IV) notation o f Turtleshells death ?
( H 4 a - i 2 b ) ; (V) another birth recorded (j2a-j6b);
(VI) a death (7a-i8b); (VII) a terminal clause,
opening with a D N and closing with a CR fixing
the recorded span o f all these statements at about
eighty-four years (j8-jl 1), from a . d . 649 to 733. A
caption of one column o f six smaller glyphs inside
the figural space may give the ames and titles o f a
woman ( k 1 - k 6 ) .
Berln him self (1965, p. 335) has briefly considered this question as to w hether the text precedes
or follows the dates. H e was unwilling to frame a
hard-and-fast rule because he limited his study of
text glyphs only to the 819-day clause, the Era
clause, and the glyphs accompanying repeated
dates. These expressions, however, are all o f chro
nographic content, unlike the ames, titles,
actions, and places o f historie or mythological
events. If Berlins text glyphs be treated as the
chronographic expressions they are, our primary
assumption, that tim e-statem ents always precede
their explanations, needs no modification. To be
sure, the segmenting of an inscription may thereby
break it into many short clauses. Some may be no
longer or even less long than one glyph block.
O thers may open by a long time-expression explained very briefly or, conversely, open briefly
with a short date, followed by a long explanation.
But if the rule of segmentation proposed here be
applied as a hard-and-fast rule, the advantage we
gain is in having an easily defined unit o f study

which focuses on the explanations as well as the

dates. This rule proposes few anomalies, and there
are few inscriptions which cannot be segmented
by it.
W hether this was a conscious rule in Maya writ
ing is still to be determ ined. For the present, we
can use it as a scaffolding o f our own construction
with which to gain access to the intended divisions
in Maya writing, whatever those may eventually
appear to be.
The format o f the IS expression, which usually
begins an inscription (figs. III-146 and III-150),
appears to govern the form at of the other following expressions. As IS statem ent opens with an
Introducing Glyph (ISIG) containing a reference
to the regent of the m onth reached in the succeeding D N . That D N usually enum erates five periods, to reach the terminal day indicated by the
D N . The corresponding CR follows. Its month
ame usually is separated from the day ame by as
many as eight glyphs pertaining to other eyeles,
mainly lunar. The entire chronographic expres
sion ends with the day ame, followed by an expla
nation com pleting the clause.
Thereafter the same form at contines in the se
quence o f D N , occasionally prefaced by a Secondary Series Introducing Glyph (SSIG or DNIG).
The D N often has noncalendric glyphs following
it and preceding the CR expression. The CR in
turn may have another chronogram or explanatory
glyphs between the day and the month, and further explanation may follow the m onth position.
The repeating clause structure therefore typically consists o f an initial D N , followed by segments o f explanation intercalated before, be
tween, and after the term s o f the CR date.
A nother type opens with a CR, followed by an
explanation. It is common at Yaxchilan and in Late
and Post-Classic inscriptions. The calendric ex
pression is m ore stenographic. H ere the initial
D N may be om itted, and the reader is left to calc
late the days elapsed between CR dates.
3. IN T E R V A L S A M O N G T H E P A R T S OF
A DATE
An im portant variation upon the rule that a date
precedes its explanation appears wherever the day
and m onth ame are separated by explanations, or

The Clauses o f Classic Maya Inscriptions

when a D N and its CR are separated by noncalendric m atter. Although Morley (1 9 3 7 -3 8 ,
vol. II, p. 44) called these interpolations a highly
unusual procedure, we now can see that they
become m ore common during the latter half of
Cycle 9, and that their general purpose may have
been to allow the same date to introduce different
classes o f inform ation by opening slots or windows
among the elem ents o f a chronographic expression. By this device the Maya scribe acquired additional locations where he could m ention different
kinds o f inform ation bearing on the same date.
These interpolations are often brief enough to
suggest their use as pictorial memory-aids to a
narrator.
For example, Lintel 25 at Yaxchilan (figs.
III-148 and III-149) carries an explanation three
blocks long, separating the D N 01 -l1 ) from its CR
( n 1 - n 2 ):

(T714)
head (T I030b)
m 2 Imix (T501): cleft sky (T562)
m

1 fish-in-hand

l2

Proskouriakoff (1963, p. 156) relates the T714

glyph to the serpent ritual for dead persons portrayed on the soffit. The remaining seventeen
glyphs following the CR date contain several
ames o f women, concluding at v2 with the ame
of Shield-Jaguar immediately before the Yax
chilan emblem at w2. H enee two different subjeets are connected with the same calendrical interval a reference to ritual and a list o f women,
perhaps of the family o f the ruler by using the
CR as a divider.
A nother example having two subjeets to share
the same time-span is Wall Panel 2 at Piedras
Negras (fig. I I I - l50). The nature o f these messages is suggested by the narrative relief sculpture.
Inside the glyphic frame, six lance-bearers kneel
before a com m ander attended by another standing
warrior. All eight warriors wear helmets, which
resemble those o f the Teotihuacan warriors portrayed at Tikal on Stela 31.
The enframing inscriptions consist o f a main
text (fig. III-151) with four clauses2 and seven captions. The main text pertains to the whole scene
and to the commander. The captions identify the
seven secondary personages.
In the main text, the first clause opens with

311

9 .1 1 . 6 . 2 . 1. It is a normal clause, containing at g 2

the helm et glyph (T678), which refers collectively
to the six kneeling warriors portrayed below it
(Morley 19 3 7 -3 8 , vol. III, p. 97). The helmet
glyph is part o f a series o f signs which recurs in the
third clause as follows:

Clause 1
H1
G2
11
J1
12
J2
K1

Clause 3
action glyph 1
helmet
turtleshell
hand T671
C-head
Imix
T1010 head

W lb

(verb)

X6

(ame)

W6

(title)

Turtleshell recurs in the fourth clause at xlO .

Since this ame reappears in all parts o f the main
text, we may suppose that it refers to the principal
figure, who is possibly also portrayed on the relief
thought to be from El Cayo (fig. III-146), now at
Dum barton Oaks.
But the third clause (o l-x 7 ) inverts the usual
order by stating the CR prior to its D N , and it
separates the explanation into two segments, preceding and following the D N . The sequence is
CR item 1 D N item 2. The D N at x l-w 3
counts back 7.5.0.16 to 9.1.12.1.9. The CR (8
Chicchan 3 Yaxkin) is stated before the D N , at
o l - p l , instead o f afterwards.
This inversin places the CR before its D N . It
may indicare the subtraction o f the D N from the
IS. But their separation also allows the scribe to
present two themes connected with this timespan. The first, at o 2 -w l, opens with an action
glyph (T670). This hand, holding an ahau glyph
(T533), is interpreted by Barthel (1968, p. 133) as
signifying lordship. The second them e follows
the D N at x 3 , and it describes the capture of a
prisoner named D eaths H ead at x4-w 5, by T ur
tleshell at w6.
The last clause (w 8 -x l2 ) opens with the D N
(1.8.12.11) counted forward from an undeclared
CR (9.10.6.5.9 8 Muluc 2 Zip). This CR is declared as an IS on Stela 36; as a CR on Stela 33;
and as a Secondary Series on Stela 38, all forming
parts o f Proskouriakoff s Ruler Series 2 at Struc-

312

A ncient Am erica

ture R-5 (1960, p. 458). The D N ends with the

hotun-ending date expressed at x l l - x l 2 , as 4
Ahau 13 Mol, at the end o f the third hotun
(9.11.15.0.0). The D N and the PE are separated
by an explanation three blocks long, which consists of the toothache or accession glyph (T684) at
x9, an anniversary sign at w lO , and the ame of
Turtleshell (T64:626b), which also occurs in
clause 2 at x 6 , and in clause 1 at il. The timeframe, however, spans 9.1.12.1.9 to 9.11.15.0.0
or 10.2.16.11. This period o f about two centuries
takes us into past historical time. T he opening
clause describes actions o f the persons represented under the command of Turtleshell. The
fish-in-hand glyph (T714) at m 1 alludes to rites
commemorating dead persons (Proskouriakoff
1963, p. 156). Clause 3 draws the readers attention to events two centuries earlier, relating to
war and the capture of prisoners. Clause 4 returns
to the present, stating an anniversary o f the acces
sion of Turtleshell and giving the dedicatory date
as 9.11.15.0.0.
Henee the general meaning of the panel may be
that Turtleshell and his companions in Teotihuacn dress com m em orate their predecessors of an
earlier age, all while stating their own titles and
achievements. T heir lineage may perhaps be
stated in the w om ens heads at n 2 , u l , and N ' l .

4. A REVIEW OF THE PRESENT

POSITION OF SCANNING ENTIRE
INSCRIPTIONS
Thus it is possible to interpret roughly the gist of
entire inscriptions, guided by the rule that chro
nographic expressions precede their explanations.
The nature o f these explanations was frst proposed by Berlins study o f em blem glyphs in 1958
and greatly expanded by Proskouriakoff in 1960
and 196364 in her studies of dynastic history at
Piedras Negras and Yaxchilan. David Kelley ex
panded the m ethod to include Quirigu in 1962.
Berln proposed a series o f four rulers for Palen
que in 1 9 6 8 .1 suggested in 1969 that the occupant
of the sarcophagus found by Ruz at Palenque bore
a ame like Sun-Shield, represented by grapheme 624 in Thom psons Catalog (1962, p. 242),
which is frequently repeated in inscriptions at
Palenque.

If approxim ate readings of whole inscriptions

are now possible, thanks mainly to Proskouri
akoff, it is timely to review what has actually been
done by the historical approach. O ur table divides
the glyph-blocks for which calendric and noncalendric readings are available from those blocks
of which we still know nothing. These latter still
add up to nearly half o f all the glyph blocks we are
considering at the three sites.
The historical phrases at Yaxchilan or Palenque
are like those o f Piedras Negras and Quirigua in
being abbreviated aids to memory. These records
typically place a local ruler in time. They enumerate his ancestry, his titles, and his deeds, and his
marriages and offspring are noted. Deaths were
occasionally recorded, and it is likely that continuing study will add new inform ation about alliances
and wars to this compact notational system, which
appears to be m ore pictorial and ideographic than
phonetic. W hole phrases, words, and ideas, rather
than single sounds, are conveyed by these pictorial
glyphs o f humans, animals, and objeets which
compose the historical clauses. Their compactness
and the fewness o f their parts support the belief
that they are aids to memory rather than phonetic
equivalents o f speech.
It is significant that their context has been established even if their exact meaning is still undetermined. Furtherm ore, many affixes or subordinare
signs still require m ore study as historical ex
pressions. Thus it may not be unjustified to repeat
here that one-third o f the content o f Classic Maya
texts has been one-third deciphered. A possible
objection should now be raised. If nearly half the
glyph-blocks are still unintelligible, how can we
assume that they are going to tell us only more
about what we already know?
The historical format itself is our main hope that
this may be true. W hatever the inscriptions convey has been channeled and set into place by the
chronographic system o f recording. It seems unlikely that material alien to this system will appear
within its network. Nonhistorical information oc
curs in the pages o f the Post-Classic Maya manuscripts in the form o f auguries, rituals, and tables
designed for use outside historie time. The main
framework is not the frame o f the Long Count. In
the manuscripts, the framework of the ritual calen
dar of 260 days governs most pages. This ritual

Table III-3.
Block Percentages
Whole
glyph
blocks

Calendric
blocks

25
36
2
8
3
1
11
2
3
14
16
15
1
13

44
32
132
175
134
110
31
36
195
30
26
20
116
1,081
100%

20
22
46
64
69
68
22
23
49
21
21
8
18
451
42.5%

5
10
7
5
10
4
2
9
15
5
2
3
11
88
8.1%

79
106
55
38
7
4
131
4
3
9
87
542
50%

A
D
E
F
I
J
K
L
G
O
10

41
50
72
70
35
85
25
54
56
88
576
100%

14
29
44
41
13
31
19
28
38
32
289
49.8%

7
9
8
5
3
6
2
12
8
7
67
11.06%

20
12
20
24
19
48
4
14
10
49
220
38.09%

14
20
20
20
32
32
32
26
16
23
7
20

3
20
19
11
1
1
2
2
2
14
2
2

Piedras Negras
(Proskouriakoff 1960)
Stela
Stela
Lintel
Stela
Stela
Stela
Stela
Altar
Lintel
Stela
Stela
Stela
Throne

Quirigu
(Kelley 1962b)
Stela
Stela
Stela
Stela
Stela
Stela
Stela
Structure
Animal
Altar

Yaxchilan
(Proskouriakoff
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Stela
Stela
Lintel

1963, 1964)
6
Str.
Str.
29
Str.
30
Str.
31
Str.
49
Str.
37
Str.
35
32
Str.
Str.
39
Str.
6
Str.
3
14
Str.

1
10
10
10
12
12
12
13
16
20
20
20

Noncalendric
blocks
analyzed

5
3
7

6
3
16
7
5
4
7

Undiscussed
blocks
19

6
24
25
27
8
7
4
1
11
(continued)

314

A ncient Am erica
Table II1-3 (Continued)
Whole
glyph
blocks

Yaxcbiln
(Proskouriakoff 1963, 1964)
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Lintel
Stela
Stela
Lintel
Lintel
Middle door
Stela
Lintel
Lintel
Lintel

15
16
17
18
22
21
25
24
27
59
28
2
10
12
42
41
21
54
53
52
33

Summary
13 Piedras Negras
33 Yaxchiln
10 Quirigu
56

St
St
St
St
St
St
St
St
St
St
St
St
St
St
St

. 21
. 21
. 21
. 22
. 22
. 22
. 23
. 23
. 24
. 24
. 24
. 33
. 39
. 40
. 42

St
St
St
St
St

. 44
. 44
. 54
. 55
. 55

17
14
21
20
32
32
43
16
16
8
18
23
26
108
22
17
37
28
14
16
18
808
100%
1,081
808
576
2,465
100%

frame cannot by itself carry historical information

concerning secular events. Thus the Long Count
of the inscriptions is now m ore evident in its role
as a historical calendar, designed and used mainly
for the chronicle of events.
We are far from having explicated all the primary meaning of any inscriptions. But the original
purpose o f the Maya scholars who composed them
was perhaps to make the meaning clear to the
farmer from the fields regardless of dialect, all
while loading the statem ent with esoteric meaning
for the learned few. The fact that students today
are able to reconstruct from inscriptions the gross

Calendric
blocks
2
2
---19
8
2
6
2
3
2

Noncalendric
blocks
analyzed

Undiscussed
blocks

46
2
2
22
2
2
2
2
201
24.8%

7
4
1
8
2
7
7
6
7
5
7
5
3
26
2
3
2
14
1
6
5
195
23.9%

9
8
20
12
30
6
28
8
3
1
8
18
23
46
18
12
10
12
11
8
11
422
52%

451
201
289
941
38.06%

88
195
67
350
14.14%

542
422
220
1,184
47.9%

outlines o f the dynastic structure is a testimonial to

the clarity, the regularity, and the accessibility of
the organization of Classic Maya historical re
cords.

NOTES
1 T o Linton Satterthw aites question as to w hether Berlin
arranged his rows to fit the foldout page, Berlin replied
(personal com m unication, Feb. 26, 1972) that his row 7
contines row 6 to save space. Row 6 he regards as a
phrase, with a second section in row 7. Such phrases
and subsidiary texts are o f course not denied by the rule of

T he Clauses o f Classic Maya Inscriptions

segmentation here proposed. Row 9 he regards as a subsidiary text beginning with T 6 1.756.
2 Linton Satterthw aite, to whom I am indebted for several
corrections and im provem ents in wording, has reason to
propose six dates here, o f which two are implied but not
stated. H e reads a CR (8 Ben 1 Xul) at l1-k 2 , and he
corrects the Bowditch-M orley reading for o l - p l from 8
Chicchan 18 Zac to 8 Chicchan 3 Yaxkin (contrast Beyer
1939, who reads 8 Chicchan 3 Ceh and a sculptors error in
the D N [x l-w 3 ] as 7.5.1.16). These em endations yield
backward counts to 8 Ben 1 X ul (9.9.19.6.13)?; to 8 Chic
chan 4 Yaxkin (9.8.17.2.5); and to 13 Muluc 2 Xul (9.1.
12.1.9) implied for the D N at x l-w 2 . H e suggests that this
date may record the foundation (by a capture) o f the lineage
o f Turtleshell (Satterthw aite 1971).

315

As to the general rule proposed here, Dr. Satterthwaite

writes (personal com munication, Feb. 10, 1972) that it
holdsgood in his opinion for chains o f linked Secondary
Series dates. H e further observes that the problem of seg
m entation is not confmed alone to chains o f dates. H e
cites inscriptions w ithout dates, but requiring paragraphs.
[N o one has contested this rule for scanning, and other
students use it as if taken forgranted. N ew com ers to the field
will need the different m ethods by D. H . Kelley (Deciphering
tbe Maya Script [Austin, 1976]) and L. Schele (Maya Glyphs.
The Verbs [Austin, 1982}). g k

III. 13
The Paired Attendants of
the Temple Tablets at Palenque

Two years ago the iconography o f the tem ple tab

lets at Palenque was first discussed as referring to
historical persons and events (Kubler 1969, pp.
18-22). It then appeared that the same personages were portrayed on all three tablets (figs.
III-132 to III-134) and that the shorter figure,
wearing voluminous mufflers and wrappings in
some discomfort, had come from a colder climate
than the tall, bare-chested man who wears only
light hip-clothes. T he rattler (crotalus) signs on the
short mans chest and the four-way hatchings on
his mufflers recall the iconography o f Teotihuacan
and associated places. Extending this argument, I
suggested that a political unin between northern and Southern peoples, between highlanders
and lowlanders, is being celebrated by the coinmon acceptance o f the worship customary at Pal
enque (Kubler 1969, p- 19). As no one has yet
contested such claims, they may still deserve further discussion.
More questions may be asked o f these three
masterpieces of world sculpture:
1. Can the ames, titles, ages, and origins of the
persons shown on the reliefs be approximated
by analysis o f the adjoining inscriptions?
2. H ow can the symbolic content o f the three re
liefs be described, both as to historical and geographical origins?
3. W hat are the M esoamerican parallels for the
format of the three reliefs at Palenque?
Im portant Maya persons are so closely associ
ated with dates repeated in their biographies that it
is possible to judge their importance by the repetitions of such dates. A Calendar Round date (9
316

Akbal 6 Xul) heads the captions accompanying the

short figure. It appears three times at Palenque on
the Tem ples of the Cross and o f the Sun. It is a day
in a . d . 641.
The taller, bare-chested man in all three reliefs
is explained by captions beginning with a day in
the year a . d . 683 (8 Oc 3 Kayab).1 It marks an
interval of forty-two years after the dates accom
panying his short companion. T he short mans
date reappears on the two tablets at Palenque and
nowhere else in the Maya corpus, but the tall
mans date reappears fve m ore times in the in
scriptions at Palenque.2
These frequencies support the interpretation
that the tall man had m ore local importance, with
six distinct occurrences of the date associated with
him, than the short man, whose date appears only
twice and only on two o f the tablets. These dates
also lead us to believe that the short man was asso
ciated with events occurring forty-two years earlier than those connected with the tall man. But
both men appear to be about the same age of ma
nir manhood. If the sculptor chose to show them
at the same historical m om ent, rather than each in
his own time scheme, then the dates in the cap
tions may be less about these men than about the
event or person they are both celebrating. We
shall return later to the nature o f that event, which
may have concerned the life-history o f a ruler at
Palenque named Sun-Shield, who is buried in the
great sarcophagus beneath the nearby Temple of
the Inscriptions.3
The ames or titles of our two men, tall and
short, are likely to appear, as in other Classic in
scriptions, near the date at the beginning of each

T he Paired A ttendants at Palenque

caption.4 The short man is named immediately
after the date, both at the Tem ple o f the Cross and
the Tem ple of the Sun by the pyramid glyph
(Thompson 1962, glyph no. 685; hereafter cited
asT + glyph no.), with affix 74, which is p a rto f the
title o f lords at Palenque (Berlin 1968a, p. 140).
The tall man is likewise named near the opening of
his caption as Jaguar-Snake (T762), and this ame
appears six times on the three tablets.
We can now examine w hether the representations show significant differences. They must first
be dated. The known historie span of the subjeets
of the three tablets covers about fifty years be
tween 9.10.10.0.0 (a .d . 642) and 9.13.0.0.0
( a .d . 692). But the span of their style as determined by Proskouriakoff (1950, p. 192) ranges
from 9.14.10.0.0 ( a . d . 721) until 9.15.0.0.0
(a . d . 721). Imagery and glyphs pertain to events
between a . d . 642 and 692, but the sculpture itself dates from between 681 (721 40 = 681)
and 771 (731 40 = 771). In making her stylistic study, Proskouriakoff (1950, p. 137) reached
different judgm ents, one as to motifs and another
as to overall composition. As to motifs, she put
the Tem ples of the Cross (TC) and o f the Sun
(TS) before that o f the Foliated Cross (TFC). As
to composition she proposed the Sun as earliest, the Foliated Cross as m ore advanced,
and she implied that the Cross was the latest of
the three.5 (See table III-4.)
If we now examine the physiological signs of age
in the tall figure, still another order appears. The
body o f the figure in the tablet of the Tem ple of
the Cross (fig. I I I - l32) is the oldest and heaviest,
with a large paunch and a deep facial furrow from
nostril to mouth. At the Tem ple o f the Sun (fig.

317

I I I - l34) the same person appears years (ten?)

younger, and at the Tem ple o f the Foliated Cross
(fig. III-13 3), the slender body and uplifted glance
suggest the youngest of the three portraits. Using
this criterion, which seriates the sculpture by the
age o f the subject, we obtain the Foliated Cross as
earliest, the Sun in the middling position, and the
Cross as the most recent of the three works. The
short figure provides another valuable clue in resolving the question of seriation. Proskouriakoff
(1950, p. 137) noted the archaic frontal mode in
the perspective rendering o f the mat pectoral in
the tablet in the Tem ple of the Foliated Cross (fig.
III-133), and the Late Classic mode of the same
form as a fat mat (halved) in the Tem ple of the
Cross (fig. III-132). Thus Proskouriakoffs estimates based on motifs diverge from her estimates
based on composition, but her assessment o f the
perspective convention as archaic or as Late
Classic confirms the seriation as to the age o f the
portrayed subject. (See table III-4.)
The question about the geographical origins of
our two figures can be approached through the
ame glyphs. The tall mans ame is the more certain because it is preceded by the jog sign, T757,
which Proskouriakoff (1968, p. 249) interprets as
directing attention to the subject o f the accompanying picture, in the sense of here is por
trayed or recorded. The ame glyph of the tall
man, T762, or Jaguar-Snake, appears only at Pal
enque, w here Thom pson (1962, p. 362) has catalogued twenty-seven examples on eight monuments and buildings. Thus Jaguar-Snakes existence is spanned between c. 9.10.10.0.0 and
9.13.0.0.0. B ut his birth is probably recorded and
his death about fifty-seven years later. The fre-

Table I I 1-4

Dates
Motifs
Composition
Age of subject
Perspective
convention

First

Second

Last

TC (9.10.10.0.0)
TC or TS
TS
TFC

TS (9.13.0.0.0?)
TS or TC
TFC
TS

TFC (9.13.0.0.0)
TFC
TC
TC

TFC

[TS]

TC

318

A ncient Am erica

quent association o f his name-glyph (T762) with

the T74.184 prefix o f lordship suggests his high
rank and may refer to his status as a regent or ruler
succeeding Sun-Shield (T624) after the latters
death.
The short mans glyph (T685) can be taken both
as an object and as an activity, in the sense of he of
the pyramid, or pyramid-builder. The ame
seems non-Maya, like the costume, which prominently displays another rare and outlandish glyph,
T65 6 (four-way hatching atTeotihuacn; see Kubler 1967, p. 15, no. 61).
This sign appears five times on the three tablets
as a badge gathering the muffler ends. The center
of its distribution is at Early Classic Teotihuacn
(Kubler 1967, table). T here it appears painted in a
feathered diadem (Seler 19 0 2 -2 3 , vol. V, pp.
51215), as well as incised upon a candelero and in
a feathered surround like a birds-eye sign (Gamio
1922; Sjourn 1964, p. 156). A Toltec mold
from Tultitlan (H. von W inning 1962) in the Valley of Mxico shows the sign beside a ruler figure,
perhaps as a ame or title, from a period ascribed
to the twelfth or thirteenth century. To these ex
amples the Codex en Croix in Paris should be added (Dibble 1942). It is illustrated on three leaves
with a 52-year schema recording events from a d
1403 until after 1569.
An origin in the Valley o f Mxico for the grapheme seems assured by such examples from different periods. It should, m oreover, be noted that the
function, form, and meaning o f pyramids differed greatly between Maya lowland and Mexican
highland peoples. T he most splendid Maya example is the Tem ple o f the Inscriptions at Palenque,
with crypt and sarcophagus, com memorating a dynastic ruler, as in several other rich Maya burials.
In the Valley of Mxico, as well as elsewhere in
highland Mxico, the burial function is usually absent from pyramidal platforms. Pyramidal decorations and texts about pyramids point instead to an
association with w ater and its worship. Thus
Muoz Camargo (1947, pp. 224ff.), the colonial
historian o f Tlaxcala, relates that at the time o f
Cortss arrival in Cholula, the inhabitants firmly
expected their dolo Quetzalcatl to destroy the
invaders with great torrents o f water flooding

. .

from his tem ple and shrines. Torquem ada (1723,

vol. I, p. 440) later added that during C ortess stay,
the people o f Cholula believed that the floods
would burst forth from the flanks of the great pyr
amid itself if they broke through its crust.
Such water associations reappear in Aztec beliefs reported by Sahagun: because springs and
rivers come from mountains, the mountains were
huge vessels o f w ater constantly in danger of being
broken and destroying the earth by floods (Robelo
190508, pp. 22627). The connection with pyr
amids as artificial mountains is clear in Cholula,
where the great platform was known as tlachihualtepetl (Kubler 1968, p. 120), or the mountains
made by hand. Parts o f it were o f Teotihuacan
date (M arquina 1951, p. 118). A t Teotihuacan, the
stone-clad platform, traditionally called the Pyra
mid o f Quetzalcatl in the Ciudadela (fig. III-119),
bore carved and painted designs o f waves and seashells, indicating the watery symbolism of these
platform terraces.
The outlandish cold-weather costume worn by
the Pyramid can now be shown to resemble
types familiar at Teotihuacan. In the Tem ples of
the Cross (fig. III-132) and Sun (fig. III-134) we
see the w earers right profile, and at the Tem ple of
the Foliated Cross (fig. III-13 3) the left is shown.
The costume has many parts: (1) a neck muffler
hangs down the back in a twisted coil; (2) a short
cape o f three panels falls to the rear and sides with
disks marked by four-way hatching; (3) a corselet
o f overlapping segments shows knotted bands
over the chest and abdomen; (4) a loincloth wrapped about the waist has a doubled apron-fold
hanging in front; (5) on the chest is a plaited-mat
pectoral, from which a tasseled rope hangs to below the knees and which is flanked at the Temple
of the Cross by crotalus rattler elem ents;7 (6) a
pointed cap seems to be made of many stiffly
folded and intersecting strips o f cloth or paper
with indistinct glyphlike markings.
The capelet and the muffler resemble those
worn at this time at Teotihuacan. Examples are the
priests o f the Teopancaxco mural (fig. III-117),
who wear stiffly stylized mufflers like collars over
capelets of two or m ore panels. A nother example
from Las Colinas near Calpulalpan (fig. III-113)

T he Paired A ttendants at Palenque

shows four processional figures wearing threepaneled capelets and mufflers falling nearly to
their heels.

II
As to Maya traits in the three panels, the pedestals
beneath the officiants and the objects in their
hands belong to Classic Maya writing and ritual.
On the tablet o f the Tem ple o f the Sun (fig.
III-134), the officiants stand on the backsof supernatural figures indicated by glyphs on their flesh.
The officiants hands hold effigy figurines o f other
supernaturals. Flanking the central panoply with
its sun-shield image are two glyphs o f skeletal
long-nosed heads bearing affixes and coefficients
for seven on the left and nine or ten on the right.
Beneath the panoply, supporting a ceremonial
bar, are two m ore seated supernaturals. The one
on the left is clearly the same as the smoker on the
east side of the sanctuary doorway o f the Temple
of the Cross (fig. 111-132). This large assembly of
supernaturals, shown as offerings, as glyphs, and as
heraldic supporters, is unique in Maya figural art,
and it suggests the obeisance of supernatural
forces to the rulership o f Sun-Shield, whose glyph
and panoply dom nate the scene, spread out over a
ground frieze o f earth signs (caban) alternating
with the sun-heads (TIO 17). M ore sun-heads ter
minare this ground frieze as if to mark the limits of
Sun-Shields power by the scope o f the sun.
The other tablets have fewer supernatural fig
ures. At the Tem ple of the Cross (fig. I I I - l32) a
skeletal head (Selers triadic sign) supports a
stylized corn plant. A mythological bird perching
atop the plant faces Jaguar-Snake, who proffers a
reclining supernatural figurine upon his stiffly outstretched arms. Facing him, the short pyramid
figure stands upon the glyph we saw flanking the
panoply on the side o f Jaguar-Snake at the Tem ple
of the Sun (fig. III-134). H e carries a scepter repeating the forms o f the pedestal of the plant,
which are a skeletal long-nosed head, wearing a
badge of shell, crossed bands, and leaf forms.
These have been interpreted as a symbol (Selers
triad) pertaining to the ritual burning o f the fields
in Maya agriculture (Kubler 1969, P- 46). Beneath

319

his bare feet is a glyphic head accompanied by the

numeral nine.
The head itself rests on a glyphic band containing the signs o f night, moon, and Venus. At the
right beneath Jaguar-Snakes feet are the signs of
the sky (T561), sun (T1017), and day. Thus the
Pyramid is shaped as a nocturnal figure in contrast to Jaguar-Snakes diurnal aspect. But the
T1030 head under his feet suggests gathering
clouds, so that we may also think o f the Pyra
mid as standing in a lofty regin above the clouds,
perhaps indicated as a mythological level o f the
skies by the numeral nine or as a high country
designated in this way.
At the Tem ple o f the Foliated Cross (fig.
III-133), the planetary ground frieze is omitted,
but the corn plant rising from a skeletal head, with
a mythological bird perched upon its crest and fac
ing Jaguar-Snake, repeats the scheme of the T em
ple of the Cross. As at the Tem ple o f the Cross, the
short officiant stands on the south side o f the tab
let with his feet resting upon the leaf of a corn
plant rooted in a conch, where a supernatural figu
rine dwells like a herm it crab. The relationship
resembles that at the Tem ple o f the Sun, where
the short figure stands upon the back o f a super
natural with the head o f a sun-glyph (TIO 10). Jaguar-Snake at the Foliated Cross stands upon a
three-faced skull (frontal view com pounded with
two profiles). From the top of the skull stylized
corn plants grow. As at the Sun, the tall officiant
holds a figurine, and his short companion holds a
scepter bearing a long-nosed mask.
Thus the tablets at the Tem ples of the Cross and
Foliated Cross share the same scheme of worship
addressed to a stylized plant rising from a skull and
supporting a mythological bird. This scheme of
worship reappears at the Tem ple of the Sun, but it
is addressed to a panoply, as if to enhance that cult
by the adjacent comparisons with the worship of
natural phenom ena such as plants and birds.
In these three cult scenes, the short figure ap
pears as an honored officiant, but in foreign dress.
His pedestals and his instrum ents, however, are all
of traditional Maya forms.
The claim that the six figures on the three tab
lets represent only two persons still requires ex-

320

A ncient Am erica

amination. The various ruler portraits at Palenque

have not yet been completely sorted out (Berlin
1968a, p. 140), and there is no reason to doubt the
existence of others. Sun-Shield is probably portrayed on the Tablet o f the Slaves (Ruz 1952, fig.
12), and other likenesses will perhaps also be identified when the ruler-lists have been established.
If the tablet figures be adm itted as portraits, and
as portraits o f only two persons, it is striking to
note that the shorter figure shows no signs of
aging, unlike the tall figure, who appears youngest
on the Foliated Cross and progressively older on
the Sun and Cross.
If this subject had aged between portraits, it is
obvious that the date with which he is associated
(9-12.11.12.10 8 Oc 3 Kayab) fixes some event.
These indications on the three tablets appear to
docum ent a term o f office by Jaguar-Snake under
the regency o f B erlins triad o f gods, following Sun-Shields reign.

III
The compositional scheme o f the three reliefs has
on central axis a frontal cult-image symmetrically
flanked by officiants in profile. The scheme is rare
in Maya art of any period and in any province.
There is at Tikal an early prefiguration o f it in the
program o f Stela 31 (fig. III-156), where a rulerfigure on the face o f the stone is flanked on its left
and right sides by warriors o f differing heights in
Teotihuacan dress, each with his own identifying
caption of glyphs above his head. Satterthwaite
favors a dedicatory date at least two centuries earlier than the Palenque tablets.
This comparison leads us to look again at T eoti
huacn. T here the compositional scheme we are
discussing seems m ore at home than among the
M ayapeoples, for we find it used in mural painting
at Teopancaxco (fig. III-l 17) and on the upper
wall atT epantitla (figs. 111-74 and 111-75), as well
as in pottery designs w here offerings are brought
to a cult-image (H. von W inning 1949, p. 147)(fig.
III-203).
This compositional scheme reappears also at
M onte Alban in Tom bs 104 (fig. 111-56) and 105
(Caso 1965, pp. 86668). The muris there are of
about a . d . 500. The processional profile figures

advancing upon the central cult-image wear heavy,

full clothing like that o f the Teotihuacan muris.
Both sites were closely connected in Early Classic
time, as we know from other resemblances and
from the evidence o f interregional trade. Thus a
case is possible for the reliefs at Palenque as being
non-Maya in the compositional scheme as well as
in regard to the costume of the smaller officiant.
The source o f the scheme would by this argument
be central and Southern highland Mxico.
A nother much later example, of the axially symmetrical com position of cult-image and flanking
officiants, is the case o f the completion-symbol
image of the Sacred Year o f 260 days, as represented in Codex Madrid and Codex FejervaryMayer (Liverpool; Seler 1901-02). H ere the format of the cruciform completion-symbol is clearly
Maya, but the pairs o f deities in the panels of these
Post-Classic manuscripts which do not antedate
a . d . 1350 are distinctly o f highland Mexican origin. In this situation, the matrix is Maya, but its
figural iconography is o f highland Mexican, PostClassic origin. Behind the axial symmetry of the
panels, we can suppose Early Classic prototypes as
to compositional schemata, pointing to origins or
parallels in the Valley o f Mxico and in the Oaxacan highlands. The apparent parallels between
the Fejervary-Mayer page and the Palenque
temple tablets are confined to compositional devices and to axial plant forms with birds in their
branches. The nine human representations, however, are of Mexican deities, unlike the two historical personages thrice repeated at the dynastic
court of Palenque. Thus the differences between
the Maya reliefs and the Mixtee manuscript recall
the differences between Classic and Post-Classic
societies and between highland and lowland cul
tures.
It is time to assemble and interpret these findings. The two men shown in the three temples are
officiants in rites celebrated during and after the
rule o f the personage named Sun-Shield. He was
buried in the sarcophagus beneath the Temple of
Inscriptions in a . d . 683. The date 13 Cimi 4 Paxis
repeated three times on our tablets in the captions
describing the taller o f the two men, whose ame
is Jaguar-Snake (T762). Since a . d . 683 is the date
of Sun-Shields sarcophagus (Fuente 1965, p. 124)

321

T he Paired Attendants at Palenque

and as Sun-Shields remains in the sarcophagus are
ofaman forty to fifty (Ruz 1955a) we may suppose
that he was born about a . d . 640. Let us therefore
suppose that the two officiants stand in rites o f the
beginning and ending o f a period defined by the
life span of Sun-Shield.
Striking differences of size and costume distinguish these officiants. The smaller man wears
costume reminiscent o f mountain climate and Teotihuacn iconography. H e is a foreigner in the
rain forest o f Palenque not only by costume but
also by his am e which is the rare pyramid
glyph. This ame is linked with Sun-Shields in the
opening clause o f the Tablet o f the 96 Glyphs
(Berlin 1968a), as of a . d . 652 when Sun-Shield
was still a boy. T he pyramid may connote the
builder, or even the tu to r, perhaps in the
sense of the Alttolteken about whom W alter Lehmann (1933, p. 149) w rote in order to account for
the Teotihuacn presence in Maya lands.
Having begun with questions, we end with
more questions. Was the Pyramid a m em ber of
the same lineage as the ruler or a ruler in his own
right, preceding Sun-Shield? The same questions
arise with Jaguar-Snake. Was he or not SunShields successor? U ntil the ruler-lists of Palen
que are further sorted out, these questions cannot
be answered definitively. But it remains significant that these two men, thrice repeated, are differentiated by costume as a native and an outlander and by their associated dates as persons
connected with the beginning and the ending o f a

reign. Was it an ecumenical rule harmonizing

highland and lowland spheres, in the repeated
linking o f these two figures?

NOTES
1 T. Proskouriakoff (letter, April 26,1972) regards 8 Oc 3
Kayab as related to the seating glyph, and the meaning as
6.1.6 [after] the seating to office forward count to the
sacrifice (T568). H enee she interprets the CR as a seat
ing o f som eone, w ithout rejecting the identification as
Jaguar-Snake.
2 Tem ple o f the Cross: l1 -l2 , u l 6 - u l 7; Tem ple o f the Sun:
l1-m1; T em ple o f the Foliated Cross: E1-E2, n 5 -o 5 ; T em
ple o f the Inscriptions, west: t8 -s1 0 ; and Ruz 1: k 6- l6.
3 Ruz arges that the occupant was named for his natal day, 8
Ahau in a . d . 655, and that he died on D ecem ber 19, 694.
4 Proskouriakoff doubts that the pyramid glyph is a ame,
because she knows o f no text w here a ame immediately
follows a d ate (letter, April 26, 1972).This passage, however, is a caption and may desgnate portrayed persons in its
vicinity differently from a stela text.
5 Proskouriakoff w rites (letter, April 26, 1972) that she
thinks the Tem ple o f the Foliated Cross is probably the
latest temple.
6 O n the identification as Sun-Shield, see K ubler 1969, pp.
20 - 22 .

7 Proskouriakoff thinks that these are m erely the knots on

the bands o f the corselet (letter, April 26, 1972), but my

sketches from the original confirm Maudslays drawings
m ore than photographs do.
[Mayanists are tacitly agreed to let any plausible reading stand
until the irrefutable one appears. This interpretation has not
been refuted but only replaced by others. I am no m ore certain o f it than o f its replacem ents. g k

111.14
The Doubled-Portrait
Lintels of Tikal

similar seated figure on Lintel 3 at Tem ple IV (fig.

III-l 55) portrays another ruler who built this tem
ple about ffty years later. Barthel (1968) carried
the epigraphic analysis in another direction by
studying the jaguar figures and glyphs on these
lintels from Tem ples I and IV. Barthel surmised of
Temple I that the jaguar-head glyphs in the inscription o f Lintel 3 refer to two jaguar-gods in
the relief, one behind the ruler and another on the
staff in front o f him. Glyphs c 1 - d 2 ame the god,
and glyph c5 gives the rulers ame (Thompson
1962, glyph no. 561:23.1030; hereafter cited as
T +glyph no.), which the present w riter, following
the custom o f approxim ate English equivalents,
paraphrases as Sun-Sky-Rain. D tting (1972, p.
223) reads this expression as above in the sky
god B. H e notes that in Lintel 3 o f Tem ple I at
Tikal it precedes the em blem and follows T181.
630:130 as a clause (contrast K ubler 1974a)
which may refer to a high priest or king ruling
Tikal.

At the temples o f Tikal our usual ideas about

rooms and passages undergo a change. We have
been accustomed since the Renaissance to an architecture o f thin-wall doorways opening upon
large rooms. But at Tikal the tem ple doorways are
large trabeated chambers which overwhelm the
tiny vaulted rooms no larger than cracks between
the walls. The passage dominates the room; the
path dwarfs the goal; the means overcome the end.
In passing beneath these carved tim ber ceilings,
we are held by the rich scenes overhead. Between
them dark vaulted slots act as quick changes of
scenes, like frames between pictures which we are
forced to view askance rather than upright. The
design stresses the doorway at the expense o f the
vault. The display o f the lintel, out o f reach and
protected from the weather, was a dom inant purpose.
The seven surviving wooden lintel reliefs of
Tikal, which were carved and installed during the
eighth century, now are scattered among Tikal,
Basel, London, and N ew York. Their original
placing and dating w ere reconstructed by Coe,
Shook, and Satterthwaite (1961). Radiocarbon
measurements in 1959 by Ralph o f samples from
two of the lintels have confirmed the epigraphic
dating of Lintel 3 from Tem ple IV near them iddle
o f the eighth century, and o f Lintel 3 from Tem ple
I near a . d . 700. An unpublished study by Jones
(1970) identifies the seated figure in the latter (fig.
III-153) as a ruler who ascended to power in the
late seventh century and who may be buried in the
tomb under Tem ple I known as Burial 116. A

II
Tem ples I and IV at Tikal have in common the
survival in each o f two carved lintels. These paired
lintels can be shown to depict doubled portraits
of rulers at Tikal, one pair in each temple, with
both lintels portraying the same person. In each
temple the paired lintels were intended to be seen
together. The spectator entering the tem ple passed first under a plain lintel before Crossing the
first narrow chamber. Thereafter he passed underneath the middle lintel, carved with figures facing
outward from the interior. After Crossing the sec322

T h e D oubled-Portrait Lintels o f T ikal

323

Table III-5
D epth

Width

Area

Lintel
Chamber
Lintel
Chamber
Lintel
Chamber

1
1
2
2
3
3

2.25
4.85
2.15
4.95
1.93
3.50

X
X
X
X
X
X

1.34
1.15
1.42
1.00
1.98
.94

=:
=
=
=

Lintel
Chamber
Lintel
Chamber
Lintel
Chamber

1
1
2
2
3
3

2.06
6.50
2.12
4.64
1.94
4.38

X
X
X
X
X
X

3.10
1.24
2.31
.74
2.37
.70

=
=
=

T em ple I

Lintel
Lintel
Lintel

1
2
3

area
3 035
3.05
382
9.905

Tem ple IV

Lintel
Lintel
Lintel

1
2
3

T em ple I

Tem ple IV

6.39
4.9
4.6
15.89

Chamber 1
Chamber 2
Chamber 3
Chamber 1
Chamber 2
Chamber 3

3.035
5.58
3.05
4.95
3.82
3.29
6.39
8.06
4.9
3.43
4.6
3.07
area
5.58
4.95
3.29
13.82
8.06
3.43
3.07
14.56

Metric dimensions from M aler 1971, plans 19 and 25.

ond chamber, he stood beneath the figures on the

third lintel, which again faced outward with their
heads adjoining the wall on the spectators left and
their feet at the wall on his right. Behind the third
lintel was a dim inutive term inal chamber.
In Tem ple I the total chamber area exceeded
the lintel area, but in Tem ple IV, the lintel area
exceeded the aggregate area o f the chambers (see
measurements in table III-5). In addition to vaulting underneath the great roof combs, the display
of these lintel reliefs with their figures larger than
life was evidently among the main concerns o f the
architect. His design treated the chambers as dark
and narrow voids separating the pictorial scenes
overhead, perhaps also serving to store cult objects and ritual paraphernalia.
D oubled portraits here refer to those Classic
Maya stelae carved on back and front with depictions o f the same person in different dress. All
such stelae were classified by S. G. Morley (1937

38, vol. I, p. 152) in class2 as having two opposite

surfaces sculptured with human figures. H e listed
twenty-seven examples, of which seven were at
Piedras Negras, seven at Quirigua, four at Copan,
three at Yaxchilan, and one each at Uaxactun,
Cancuen, N aachtun, M otul, and Yaxha, but none
at Tikal (Morley, 19 3 7 -3 8 , vol. IV, table 108).
Some of these, like Stela 10 at Uaxactun, dated
before a . d . 435 at the end o f Cycle 8, are too
fragmentary to determ ine w hether the portrait fig
ures are of the same person. Stela 19 at Yaxchilan,
dated 9.12.8.14.0 ( a . d . 600) doubly portrays
Shield-Jaguar. The rulers of Series 3 at Piedras
Negras probably appeared in doubled portraits on
Stelae 4 , 5 , 7 , and 8, dated from 9.13.10.0.0 ( a .d .
702) to 9.14.15.0.0 ( a . d . 726) (See Proskouriakoff 1960, p. 456). T he doubled portraits at
Copan, like all the rem ainder in M orleys class 2,
still need detailed study. The dynastic stelae of
Quirigua, spanning the years from a . d . 761 to

324

A ncient Am erica

810, were first identified by D. Kelley (1962a) and

were discussed as doubled portraits by K ubler
(1969, pp. 14-18).
The inscriptions on the two lintel reliefs at T em
ple I allow the seated figure to be identified as the
same person in different costumes. The ame
glyphs at b 4 on Lintel 2 (fig. I I I - l52) and preceding the Tikal em blem at c5 and E l i on Lintel 3
(fig. III-15 3), correspond to Sky-Rain, or
T 5 6 l:2 3 .1030. O n Lintel 3, C hristopher Jones
(1970, p. 3) identified the clause containing the
nominal glyph at El 1 as occurring in an inaugural
clause, opening with the seating glyph T644. This
personage is named and portrayed again at Tikal
on Stela 16 at b 3 - 4 and on Stela 5 at a z2 . These
references at Tikal span the years from before
9.14.0.0.0 ( a . d . 711) until 9.15.13.0.0 ( a . d . 744),
according to the dedicatory dates tabulated by Satterthwaite (1958, pp. 12021).
Sky-Rains ame and portrait reappear at
Naranjo, forty kilometers east of Tikal. H. Berlin
(1968b, pp. 17-18), w ithout naming him, suggests that his birth was recorded on Stelae 22, 24,
and 29 at N aranjo as o f 9.12.15.13.7 ( a . d . 687)
and that his inaugural cerem onies at N aranjo occurred between a . d . 698 and 702. Joyce Marcus
(1971) adds that he was the son of a woman from
Tikal whose ame glyph includes the num ber six.
H er son is portrayed seven times at N aranjo in
different regala from a . d . 694 until 721, being
shown at inauguration (Stela 20) and in other roles
as priest (Stela 21) and warrior (Stelae 2 and 30;
fig. III-180).
Tem ple IV at Tikal, like Tem ple I, contains two
portrait-bearing lintels which depict the same p er
sonage. His ame may be translated from the
glyphs as Sun-Sky-Rain. These appellatives
(T16 544}:561.1030) appear on Lintel 2 (fig.
III-154) at b 5 and c2; and on Lintel 3 at d 2 and e 9
(fig. III-155), and they appear to ame the seated
persons on both lintels. O th er references to the
ame o f this ruler occur on Stelae 5, 19, and 22,
where he appears in profile bearing different insig
nia of rulership. The dates of these monum ents
span the years from a . d . 741 (lintels o f Temple
IV) to 790 (Stela 19). C h risto p h erjo n es (1970, p.
13) named him Ruler B at Tikal and was first to
suggest that his portrait might appear on both lin
tels as the builder o f Tem ple IV in a . d . 741.

At Tikal there is no surviving example of a stela

belonging to M orleys class 2, which includes all
stelae with human figures sculptured on back and
front. Y et Stela 6 at Yaxh, thirty kilometers
southeast from Tikal, which resembles early sculpture before 9-5.0.0.0 at Tikal, had human figures
both front and back according to Morley (1937
38, vol. III, p. 472). H e placed it in his class 2, even
though the back design was weathered beyond
recognition. It is therefore o f some interest to find
these doubled portraits at Tikal in an architectural
setting, w here the sculptors idea o f portraying the
same person twice on the same m onum ent was
translated into front and rear lintel positions in the
temples.

III
We may now turn to the question of the meaning
of the doubled portraits at Temples I and IV at
Tikal. For the present, our only way to approach
their meaning is through the different costumes
and attributes w orn by the rulers. In both lintels in
Tem ple I the format is the same: seated in profile
beneath an overarching anim al-protector figure,
the ruler faces an upright shaft.
On Lintel 2, nearest the entrance of Tem ple I
(fig. III-152), he is portrayed as a seated warrior,
bearing lances and a shield in his left hand, beneath
towering serpent-head forms. But on Lintel 3,
deeper inside the tem ple (fig. III-153), SkyRain reappears beneath a giant jaguar standing
over him, and he is seated on an rnate throne,
bearing a manikin scepter in his right hand and
attended by a richly dressed dwarf.
Many details o f costum e and setting seem to
mark categorical differences between the two por
traits, going beyond the primary contrast of fig
ures bearing a scepter on Lintel 3 and bearing
weapons on Lintel 2. Considered as agroup, these
differences appear in the animal protectors, costuming, seats, terraces, and standing shafts. All
these differences point to an ethnic distinction be
tween the insignia o f lowland Maya and highland
Mexican peoples, between the Peten and the Valley o f Mxico, and specifically between Tikal and
Teotihuacan.
The most im portant o f these distinctions is in
the giant animal protectors. O n Lintel 3 ofTem ple

T he D oubled-Portrait Lintels o f T ikal

I (fig. III-153) the lotus-jaguar head appears twice,
both as the giant animal protector and on the head
of the staff between the ruler and the dwarf. The
lotus-jaguar is peculiar to lowland Maya symbolism, being also a Maya glyph (T 2 13:75 Ib) signifying according to Barthel (1968, p. 151) a
bringer o f wealth and abundance, often represented at Copan and in Classic Chama pottery
painting.
On lin te l 2 ofT em ple I (fig. 111-152), however,
the protector is a giant mosaic-encrusted serpent,
repeated in smaller form above the upright shaft.
The seated ruler below them likewise wears a m o
saic-encrusted headdress repeating the serpent
forms in the upper register.
Mosaic-encrusted headdresses in the Southern
Maya lowlands appear both in glyphic writing and
in the m onum ental sculpture as Mexican highland
attributes o f Teotihuacan origin. The warrior
helmet, o f the type shown in Teotihuacan muris
and figurines, appears on Stela 31 at Tikal (fig.
I I I - l56), dated after a . d . 435 near the beginning
of Cycle 9 (Satterthwaite, Tikal MS, checklist II).
These warriors recur at Piedras Negras on Lintel 2
(fig. III-150) as o f a . d . 667 (9.11.15.0.0), where a
glyph at g 2 refers to these warriors; it represents
their distinctive helmet, and it reappears at Palen
que in the tablets of the Tem ple o f Inscriptions
(figs. 111-137 to III-139), dated by Ruz (1973) as
commemorating the death o f a ruler buried there
after a . d . 694 (9.13.2.14.6). A nother versin of
the mosaic headdress assumes serpent form at
Piedras N egras on Stela 7 (9.14.10.0.0; a . d . 721)
and Stela 9 (9.15.5.0.0; a . d . 736). It designates
rulers in w arrior dress, as at Tikal, early in the
eighth century.
Between the rulers feet on Lintel 2 ofT em ple I
(fig. III-l 52) is an im plem ent like the featheredjaguar censer shown in a Teotihuacan mural at
Zacuala (fig. 111-99)- Such feathered-jaguar forms
occur at M onte Alban and in the Maya-Toltec art
of Chichn Itz, but censers o f this type are not
present in standard Classic Maya iconography.
The short lances carried by the ruler on Lintel 2
ofT em p le I display another m otif which is common in mural painting at Teotihuacan. O n the
shafts are stylized human-eye forms, which at T eo
tihuacan often appear on stylized riverbands, as in
the Tepantitla mural (fig. 111-73), perhaps to con-

325

vey the idea of the sparkle of flowing water

(Kubler 1967, p. 15, no. 57).
The terraced platform at the northern base of
the same lintel contains (in addition to another
mosaic-encrusted serpent head) three steps decorated with motifs which also resem ble wall paintings and stucco decorations related to the art of
Teotihuacan. O f these the most striking is at the
right on the top step; it portrays from right to left
the flower, body and thorns, and root of the barrel-cactus biznaga (Kubler 1967, p. 15, no. 29).
Representations o f this arid highland plant are
known in Classic time only at Teotihuacan, as in a
mural at Zacuala (Sjourn 1959, fig. 9). Below
the barrel cactus on the bottom step a stylized
bird-wing appears: cise parallels to it are figured
in stucco on the facade o f the Early Classic temple
at Acanceh ( a . d . 3 0 0 -5 5 0 ), in northern Yucatn,
among other forms o f Teotihuacan derivation. Finally, the upright staff at the right bears three
bindings containing one round spot in each: this is
the folded-triangle sign o f Teotihuacan (Kubler
1967, p. 15, no. 54; see fig. 111-84) in the halfrepresentation meant to be read as the profile
view. We may therefore suppose that the frontal
view showed the com plete folded-triangle sign
with two dots em paneled and suggestive of the
goggles o f the highland rain deity.
Lintel 2 at Piedras Negras (fig. III-150), carved
about a . d . 667 (9-11.15.0.0), takes us back about
a century and a half to describe the early history
there o f warriors in Teotihuacan dress. The inscription discusses also their descendants (Kubler
1974). A similar situation may be present at Tikal:
the attendant warriors o f early Cycle 9 on Stela 31
(fig. III-156) belonged to a lineage at Tikal which
later enjoyed the ruling rank described in the costuming o f Lintel 2 at Tem ple I (fig. III-l 52). Thus
the doubled portraits o f Sky-Rain, costumed in
Maya dress and in that of Teotihuacan, may mark
the convergence o f distinct dynastic lines in the
rulership o f Tikal.
Tem ple IV com m em orates another ruler at
Tikal, named Sun-Sky-Rain, who reigned about
half a century later, as we know from the inscrip
tions on two m ore lintels bearing his doubled por
traits. As at Tem ple I, each lintel displays different
ethnic associations, but at Tem ple IV the formats
are different. Lintel 2 alone (fig. III-154) repeats

326

A ncient Am erica
Table III-6

M onument

Stylistic date

Dedicatory date

TI
TI
TIV

L2
L3
L2

9 .1 7 .1 0 .0 .0 2 katuns ( a . d . 780 40)

9 .1 6 .0 .0 .0 2 katuns ( a . d . 751 40)
9 .1 5 .1 0 .0 .0 2 katuns ( a . d . 741 40)

TIV

L3

9 .1 6 .0 .0 .0 2 katuns ( a . d . 751 40)

the giant jaguar-god protector rising above the

ruler seated in profile on a cylindrical throne.
Lintel 3 (fig. III-155), deepest inside the shrine,
portrays Sun-Sky-Rain again, but here he ap
pears in a frontally seated position, enframed by a
double-headed ceremonial serpent-bar beneath a
winged-bird crest, and enthroned on a threestepped litter or carrying frame. The mosaic-encrusted serpent-jaw helm et resembles that o f
Lintel 2 at Tem ple I (fig. III-l 52), but the format
of the front lintel at Tem ple IV (fig. III-154) resembles the format o f both lintels at Tem ple I
(figs. 111-152 and I I I - l53)- In effect, theT eotihuacn trait o f the mosaic serpent-helm et in Temple
IV has been prom oted from the outer to the inner
position and from the profile presentation (fig.
III-l 52) to a frontal one (fig. III-155), suggesting
an increase in im portance for Mexican highland
lineage in this generation at Tikal. In Lintel 3 of
Temple IV (fig. III-155), Sun-Sky-Rain also is
endowed with warlike associations such as the
owl(?)-like crest (Kubler 1967, pp. 910; see Fig.
III-l 12), the death signs on his shield (cimi) and
mantle (deaths-eye borders), and the skeletal fig
ures on the steps. In general the whole configura
ro n is m ore Maya than Mexican but the highland
associations just noted seem to mark continuity
with similar traits on Lintel 2 at the front o f T em
ple I.
A nother peculiarity o f Tem ples I and IV may
now be discussed. Tem ple I com m em orated the
earlier ruler, named Sky-Rain, and Tem ple IV
commemorates a later one, named Sun-SkyRain, but the stylistic traits indicate that the lin
tels of Tem ple I were carved later than those o f

(9 .13.10.0.0) ( a . d . 702) (L. Satterthwaite)

9 .1 3 .3 .0 .0
after 9 .1 5 .1 0 .0 .0 ( a . d . 741); probably
9 .1 6 .0 .0 .0 ( a . d . 751)
after 9 -1 5.10.0.0 ( a . d . 741); probably
9 -16.0.0.0 ( a . d . 751)

Tem ple IV (Coe and Shook 1961, pp. 22 and 64).

That is, the lintels o f Tem ple IV may have been
carved about the time of its dedication, but those
of Tem ple I were carved some years later than
those o f Tem ple IV and decades later than their
dedicatory date. In other words, the sculptors
commissioned to com m em orate the earlier ruler
were at work later than those working for the second ruler. H enee the presence of Teotihuacan
traits, which is so pronounced on Lintel 2 of Tem
ple I, records events before a .d . 702, but the
beams were not carved until about a . d . 780 40,
and after those in Tem ple IV. To conclude, if
Tem ple IV were carved earliest, its lintels re
corded very recent events. But if Tem ple I was
carved 5 0 -1 0 0 years later, it is about events which
were then about three generations in the past,
when Teotihuacan was being abandoned.
If Lintel 2 in Tem ple I (fig. III-152) shows a
ruler in Tikal in Teotihuacan attire and regala, was
it because he had in some way been entitled to
continu the lineage from Teotihuacan at Tikal
around a .d . 700? If so, did the historie importance
o f the decline and fall o f Teotihuacn em erge only
gradually among the dwellers in the Peten? Or
does the inform ation on Lintel 2 in Tem ple I re
cord an actual transfer o f lineage and power from
Teotihuacan to Tikal? The answers to these questions will probably be learned only from a systematic study o f all Teotihuacan traits in Maya art
(H ellm uth 1973).
[ An admirable and independenc expansin o f these observations appears in the study by C hristopher Jones (Inauguration Dates of T hree Late Classic Rulers o f Tikal, Guatemala,
American Antiquity, 42:1 [1977], 2860). g k J

III. 15
Mythological Dates at Palenque
and the Ring Numbers
in the Dresden Codex

For many years the thesis o f calendrical uniformity

among all the Preconquest Maya has inhibited
thought about regionalism and historie change in
matters o f time-count. Inconsistency was noted by
E. W. Andrews IV (1965a; 1965b; 1968) between
northern Maya lowland radiocarbon dates, which
favor a 12.9-0.0.0 correlation, and Southern
lowland ones favoring a 11.16.0.0.0 correlation
(tables III-7 and III-8). These correlations of
course differ by about 259 years in their interpretations o f the Initial Series dates o f the monuments in terms o f European chronology, by keying the end o f the last katun, 13 Ahau, before the
Spanish Conquest, either to 12.9.0.0.0 in 1536 or
to 11.16.0.0.0 in 1539 (R. L. Roys 1933, p. 205).
A dilemma is that neither correlation fits both
Southern and northern archaeological data. During his many years of excavation in Yucatn, An
drews added two periods to the sequence. In the
process he found 11.16.0.0.0 too short to contain
the history o f the Puuc cities (his Pur Florescence), plus Mexican domination (his Modified
Florescence), plus his now-disputed Black-onCream Period, so that on these grounds he preferred the 550 years o f the 12.9.0.0.0 correlation
to the 300 years o f the 11.16.0.0.0 equation without being able to resolve the dilemma that he presented in a table in 1965 (table III-7), together
with his assessment o f the radiocarbon dates favor
ing 12.9.0.0.0 (table III-8). A ndrews dilemma is
not easy to sidestep or ignore, as Willey pointed
out in 1971.
In 1975, E. W. Andrews V (table III-9) deleted

all transition periods, thereby discarding Andrews

IV s alternate correlation at 12.9.0.0.0. His reasons are that the architectural, stratigraphic, and
ceramic data from Dzibilchaltn do not justify creation o f a separate and intervening culture peri
od. The dilemma still remains, however, because
of the unresolved inconsistency between northern
and Southern radiocarbon dates noted by E. W.
Andrews IV (1965a; 1965b; and 1968). W ithout
questioning the ceramic evidence used by E. W.
Andrews V, we still need to examine the implications o f the divergent radiocarbon datings. Andrewss dilemma is a true dilemma in being a
choice between equally unsatisfactory alternatives, either too little time or too much, either too
early a correlation or one too late. But a dilemma is
only a form o f argument, in this case about the
correlation between Maya and Christian time. If
we focus now on Maya time by itself, the dilemma
may be transformed.
There is, for example, evidence in the Dresden
Codex for the need of fluid placements. The
D resden Codex, which Thom pson (1972, p. 15)
dates c. a . d . 12001250, contains twenty-seven
Initial Series, spanning eleven centuries from the
dates 8.6.16.7.14 to 10.19.6.1.8. They lack both
Introductory and Period Ending Glyphs, as well as
Supplementary and Secondary Series. They are
more like Distance N um bers between stated Cal
endar Rounds. Satterthwaite (1964, p. 50) calis
them also Long C ount Series to distinguish them
from the Initial Series in inscriptions. O f these, six
occupy Cycle 10, beginning at 10.11.3.18.14 in
327

328

A ncient Am erica

the mid-eleventh century (c. a . d . 1051?). All

twenty-seven are carefully specified as belonging
on the 4 Ahau 8 Cumhu Base D ate by its expression, either at the cise or the opening o f the
Initial Series number, as though to distinguish
such a count from another or others in current use.
In contrast to the inscriptions, where such a
statem ent was rare (seven examples) and probably
unnecessary, these Initial Series numbers in the
Dresden Codex suggest reference to a foreign or
disused system. It is like specifying European
dates as Od Style, prior to the acceptance of
Gregorian reforms (1582), or explaining that such
a notation differs from the one in local use.
There may exist in the D resden Codex, moreover, ten series in another count, differing from
the Initial Series era in having a different Base
Date, which is 9 Kan 12 Kayab (table III-10) instead of 4 Ahau 8 Cumhu. Beyer (1933, p. 2) saw
ft to cali them, appropriately, Serpent Series.
The idea o f special counts in the Maya calendar
is generally accepted. R. W. Willson (1924), the
astronom er who coined the ame for the Ring
N um bers, first proposed that they indcate periods in special counts before the beginning o f the
Initial Series at 4 Ahau 8 Cumhu.
To summarize information on the Ring-Number Base Dates, ten o f sixteen are related by
glyphic expressions naming the same supernatur
als. These ten are all Serpent Series N um bers
counted from Base D ate 1, which is 9 Kan 13
Kayab, some 34,000 years ago. Three other Base
Dates accompany tables of the mltiples of 364
days. Being useful in converting Maya tun-counting into vague solar years, these tables and their
associated Base Dates probably connect with
mythological persons and events.
D resden 6 1 , 6 2 ,
Palenque TC
D resden 2 4 / a - b

69

Only two Base Dates are related to astronomical

tables: one is about the revolutions o f Venus (Base
Date 3), and another is about lunations (Base Date
16). Three Base Dates (6, 8, and 9) are marked by
the accompanying glyphic expressions as con
cerned with divination. In effect, all but two RingN um ber Base Dates resemble historical records,
by not being Period Endings and by being spaced
in irregular, short intervals.
The inscriptions include several dates of this pe
riod before 3000 b . c ., and one of them at Palen
que resembles a R ing-N um ber Base Date.
The first two columns of the tablet in the
Tem ple of the Sun record at b 1 7 the birth in
12.19-13.0 l Ahau 18 Zotz o f a woman at Palen
que here named Feather-Down (T69:1000a). Her
ame relates to weaving, like the ame o f the deity
of weaving who is frequently named in the Base
Date passages o f the D resden Codex, as having to
do with the Serpent N um bers and related calculations.
Let us suppose that the oldest Base Dates ( d a d h , e a , e b ) at 9 Kan 1 2 Kayab (see table III- 10)
refer to an age of creation, which the texts on
Dresden 6 1 / a - b and 6 9 / c - d presumably explain.
Although these messages are still far from clear to
us, they seem to convey the same ideas as the Cre
ation o f the W orld in the Chronicle o f Chumayel
(Gates 1 9 3 2 ; Roys 1 9 3 3 ) , as we soon shall see.
The mythological section at Palenque on the
Tablet of the Cross occupies the same period as
most of the D resden Ring-N um ber Base Dates.
This period is about 30,000 years later than the 9
Kan 12 Kayab Base Date, and it occurs before
3000 b . c . The Palenque date also precedes the
third Base D ate after 9 Kan 12 Kayab by 260 days:

(1 .1 1 .1 5 )3 .1 6 .1 4 .1 1 .4
1 2 .1 9 -1 3 .1 6 .0

and it follows the second Base D ate by about 134

Palenque TC
D resden 6 3 /c, 3 la /c

1 2 .1 9 -1 3 .3 -0

12.19.13.3.0
-1 2 .1 2 .1 7 .3 .1
+ .6.15.17.19

Kan

Kayab
1 Ahau
Ahau 1 8 Kayah
12

18

B.D.l
Zotz
B.D.3
years:

1 Ahau 18 Zotz
13 Imix 9 U o
B.D.2

T able III-7.
Summary of the Stratigraphic Column in the N orthern Maya Area (after E. Wyllys Andrews IV 1965, p. 62)
Maya
Calendar

Correlation at Approx.
12.9.0.0.0

Gregorian
Calendar

Correlation at Approx.
11.16.0.0.0

Maya
Calendar

A .D .
12.15.0.0.0
12. 10.0 .0.0

Colonial Period

1600

Colonial Period

12. 0 .0 .0.0

Decadent Period
Second Phase (Post- Monumental

1500

! Decadent Period
Second Phase(Post-Monumental

11.15.0.0.0

12. 5.0.0.0

1400
Decadent Period
First Phase (Mayapn)

12 . 0 . 0 .0.0

11.15.0.0.0

11. 10.0.0.0

1300

11 . 10 .0 .0.0

j Decadent Period
First Phase (Mayapn)

11. 5.0.0.0
11. 0 .0 .0.0

1200

1100

Transition (Black-On-Cream)

1000

11. 5.0.0.0 ^ Florescent Period

Second Phase (Modified)

900

^Transition (Black-On-Cream)
Florescent Period
Second Phase (Modified)
M Florescent Period
0
First Phase (Pur)
'Transition

I
Florescent Period
First Phase (Pur)
Transition

800
700
600

Early Period
Second Phase

(Tepeu 3?)

Transition

o. y o.o.o
10. 0.0.0.0

(Tepeu 1-2)

9.10.0.0.0

(Tepeu 3?)
400

Early Period
First Phase

10. 10. 0 . 0.0

9.15.0.0.0

500

Early Period
Second Phase

10.15.0.0.0

(Tepeu 1-2)

9. 5.0.0.0
Early Period
First Phase

(Tzakol)

300

8.15.0.0.0

200
Transition
(Tzakol)

8.10.0.0.0
(Matzanel)

100

8. 5.0.0.0

8. 0.0.0.0

(Matzanel ?)
100

Dzibilchaltn Formative
Phase 111

9. 0.0.0.0

Dzibilchaltn Formative
Phase III

7.15.0.0.0
7.10.0.0.0

300
7. 5.0.0.0
400

Dzibilchaltn Formative'
Phase I IB

500

Dzibilchaltn Formative
Phase IIB

600
Dzibilchaltn Formative"
Phase IIA

700

Dzibilchaltn Formative
Phase IIA

7. 0.0.0.0
6.15.0.0.0
6 . 10.0 .0.0
6. 5.0.0.0

800
Dzibilchaltn Formative
Phase I

6. 0.0.0.0
900

Dzibilchaltn Formative
5.15.0.0.0

1000

5.10.0.0.0
5. 5.0.0.0

329

A ncient Am erica

330

Table III-8.
Radiocarbon Determinations from the Northern Maya Area (After E. Wyllys Andrews IV 1965, p. 63)

No.

Site

Lab.
Number

Culture
Period

C14 Date

Correlation
Indicated

1*

Balankanche

LJ-272

F-II

860 90

12.9.0.0.0

2*

Balankanche

LJ-273

F-II

860 100

12.9.0.0.0

Chichn Itz

Y-626

F-II

790

12.9.0.0.0

4
5

Chichn Itz
Chichn Itz

Y-626b
LJ-87

F-II
F-II (?)

810 + 100
810 -+ 200

6*

Chichn Itz

TBN-313-1

F-I

7*

Chichn Itz

TBN-313-2

8*

Chichn Itz

9*

70

12.9.0.0.0
12.9.0.0.0
12.9.0.0.0

F-I

600 - 70

12.9.0.0.0

TBN-313-3

F-I

610 70

12.9.0.0.0

Dzibilchaltiin

LJ-505

PE-I

-975

340

--

10*

Dzibilchaltun

1-171

PE-II-III

-3 2 0

80

--

11*

Dzibilchaltun

LJ-279

PE-II-III

-2 5 0 90

--

12*

Dzibilchaltun

LJ-531

EP-I

430 200

(11.16.0.0.0)

13*
14*

Dzibilchaltun
Dzibilchaltun

M-857
W-707

EP-II
EP-II

458 200
508 - 200

(12.9.0.0.0)
(12.9.0.0.0)

15

Mayapn

GRO-1166

D-I

1310 55

16
17

Mayapn
Tulum

GRO-452
Y-393

D-I
D-I

1250 3: 95
1070 - 60

18

Uxmal

Y -621

F-I

560 50

19

Uxmal

GRO-613

F-I

885

O
00

70

100

12.9.0.0.0
(12.9.0.0.0)

Notes

Cave shrine, Twig

charcoal
Cave shrine. Twig
charcoal.
Castillo. Zapote
lintel.
(Re-run o f Y-626).
Las Monjas. East
Patio (?).
La Iglesia. Original
beam.
La Iglesia. Another
original beam.
Casa Colorada.
Original beam.
Str. 605. Hearth
charcoal.
Str. 226. Carbonized jack beans.
Str. 450. Hearth
charcoal.
Str. 612. Hearth
charcoal.
Str. 1. Lintel.
Same sample. Rerun.
Str. R-87. Burned
beam.
Str. 1. Charcoal.
Str. 21. Zapote
beam.
Casa del Adivino.
Lintel.
Monjas Quadrangle. Lintel.
Average

190
190
260
240
240
150
330
320

0
135
85

390
60
199

*Few determ inations produced during current program o f research.

In the succession o f the Base Dates, only 1,654

days separate Base D ate 4 from Base D ate 16. It is
notable that Base D ate 8 lies exactly 360 days be
fore Base D ate 16. This spread o f thirteen dates
over about six years suggests a closely spaced se
ries of events, as o f the births o f founders o f lin-

eages o f equal rank or the institution of worship

for supernatural beings. The intervals m irror the
irregularities o f historical life. N one is longer than
101 days or shorter than 16 days, and none has any
known calendrical character, other than the 360day (tun) interval between Base D ate 8 and Base

Table III-9.
Chronological Chart o f Culture Periods in the N orthern Maya Area and Ceramic
C om plexes at Dzibilchaltn (after J. W. Ball and E. W. Andrews V 1975, p. 235)

Maya Long
Count
12. 0.0.0.0

Northern Maya Area

Culture Periods

Dzibilchaltn
Ceramic
Complexes

Colonial

1600

11.15.0.0.0
11.10.0.0.0

Gregorian
Calendar

1500
D ecadent

Chechem

1400

11. 5.0.0.0

1300

11. 0.0.0.0

1200

10.15.0.0.0

M odified florescent

Zipche

10.10.0.0.0

1100
1000

Pur florescent
Copo 2

10. 5.0.0.0
10. 0 .0.0.0

900
800

(Tepeu 2)
9 .1 5 .0 .0 .0
9 .1 0 .0 .0 .0

Copo 1
Early II ..................... ........
(Tepeu 1)

600
500

9. 5.0.0.0
9. 0 .0.0.0

700

Piim

Early 1
(Tzakol)

8.1 5 .0 .0 .0

400
300

Xculul 2

8.1 0 .0 .0 .0
8. 5.0.0.0

200
100
A .D .

8. 0 .0 .0 .0

Late Formative

Xculul 1

0
B .C .

7.15.0.0.0

100

7 . 10.0.0.0

Komchen

200

7. 5.0.0.0

300

7. 0 .0.0.0

400

6 .1 5 .0 .0 .0
6 .1 0 .0 .0 .0

Middle Formative

Nabanche

500
600

A ncient Am erica

332

T a b le 111-10.
Ring-Num ber Base Dates

Base
Dates
1
1
1
1
1
1
1
1
1
1
2
2
2
3
3
4
5
6
7
8
9
10
11
12
12
13
14
15
15
16

Passages
(Satterthwaite
1964)
DA
DB
DC
DD
DE
DF
DG
DH
EA
EB
DM
DN
FC
AA
AB
EE
EC
CA
DI
GA
CB
DK
EF
DJ
FA
ED
HA
DL
FB
BA

Page
Locations
6 1 /C -D , B l.
6 1 /C -D , Rd.
61/E-F, B l.
61/E-F, Rd.
6 2 /A -B , B l.
6 2 /A -B , B l.
6 2 /C -D , B l.
6 2 /C -D , Rd.
69/E-F, B l.
69/E-F, Rd.
6 3 /C , B l.
6 3 /C , Rd.
31a/C
2 4 /A -B
24/A -C
70a/B
70a/A
58/E-F, B l.
62/E
43b /C
58/E-F, Rd.
6 3 /A
70b/B
62/F
3 la /A
70b/A
45a/A -B
6 3 /B
31a/B
51a/A , B l.

Date 16, which might be named the tun of the

N ine Base D ates, or the N ine Beginnings.
This presence of N ine and Thirteen evokes the
colonial text about the end of the dawn o f the
world (ahalcab) in the Chronicle o f Chumayel
(Roys 1933, p. 99). D u rin g th e Creation, theT hirteen Gods o f the heavens were overcome by the
N ine Gods o f the underworld (Gates 1932, p. 81),
and the Four Bacabs were set in place.
H owever this may be, in the D resden Codex a
hypothetical Creation D ate is followed after about
30,000 years by a cluster o f beginnings spread

Notations
9
9
9
9
9
9
9
9
9
9

Kan 12 Kayab (1.11.15.) 3.16 .1 4 .1 1 .4

Kan 12 Kayab (1.11.15.) 3.16 .1 4 .1 1 .4
Kan 12 Kayab (1.11.15.) 3.1 6 .1 4 .1 1 .4
Kan 12 Kayab (1.11.15.) 3.1 6 .1 4 .1 1 .4
Kan 12 Kayab (1.11.15.) 3 .1 6 .1 4 .1 1 .4
Kan 12 Kayab (1.11.15.) 3 .1 6 .1 4 .1 1 .4
Kan 12 Kayab (1.11.15.) 3 .1 6 .1 4 .1 1 .4
Kan 12 Kayab (1.11.15.) 3 .1 6 .1 4 .1 1 .4
Kan 12 Kayab (1.11.15.) 3 .1 6 .1 4 .1 1 .4
Kan 12 Kayab (1.11.15.) 3 .1 6 .1 4 .1 1 .4
13 Imix 9 U o 12.12.17.3.1
13 Imix 9 U o 12.12.17.3.1
13 Imix 9 U o 12.12.17.3.1
1 Ahau 18 Kayab 1 2 .19.13.16.0
1 Ahau 18 Kayab 1 2 .19.13.16.0
9 Ix 2 Chen 12.19.15.7.14
9 Ix 7 Xul 12.19.18.5.14
13 Muluc 2 Zac 1 2 .19.18.10.9
3 Kan 17 Mac 12.19.18.13.4
3 Lamat 1 Uayeb 12.19.19.0.8
13 Muluc 17 Zec 12.19.19.5.9
3 Chicchan 13 X ul 12.19.19.6.5
4 Eb 0 Mol 12.19.19.7.12
13 Cauac 7 Ceh 12.19.19.11.19
13 Cauac 7 Ceh 1 2 .1 9 .19.11.19
9 Ix 2 Kankin 12.19.19.13.14
13 Oc 18 Pax 12.19.19.16.10
13 Akbal 11 Kayab 12.19.19.17.3
13 Akbal 11 Kayab 12.19.19.17.3
12 Lamat 16 Cumhu 13.0.0.0.8

over six years immediately near the Era Clause

at 13.0.0.0.0 4 Ahau 8 Cum hu (Berlin 1965, pp.
3 3 4 -3 5 ), when history began. These beginnings
are marked by Ring N um bers as relating to the Era
and as com peting with the Era itself for special
consideration and worship.
Such texts, both in the D resden Codex and in
the inscriptions, suggest that they are about the
mythological origins prior to 3000 b .c . o f the prin
cipal ruling Maya families of the Classic era
(Thompson 1973; K ubler 1973), which by hypothesis came to be used as zero points or depar-

M ythological Dates at Palenque

rure dates for local dynastic histories. This tradition continued in the katun count o f the colonial
era, as the Maya Chronicles or Books o f Chilam
Balam.
To conclude: long before the Initial Series
count fell into disuse after 10.4.0.0.0, it is proposed here that its uniform era began to com pete
in Maya historiography with other Base Dates. Its
forms were replaced by local option not only as to
the well-known shift in day ames for year bearers, but also as to varying Base Dates or starting
points for dynastic histories.
These Base Dates, which had been associated
with different lineages and dynasties since the fifth
century or sooner, carne into increasing use during
Cycle 9 and were ultimately continued in the ka
tun count.

333

In 1950 Paul K irchhoff tried to resolve many

discrepancies in the katun-count chronicles by
suggesting the existence o f separate chronologies
from different starts, serving as family or dynastic
annals. H e was led to propose as many as thirteen
different histories, corresponding to the districts
of Yucatecan communities. But at this point A n
drews'j Dilemma becomes Kirchhoffs 13-beaded
Hydra, which we cannot examine because his
manuscript is still unavailable.
[Appearing next to mine in the same issue, a study
Lounsbury treats similar material very differently
tionale for the Initial Date o f the Tem ple o f the
Palenque, Segunda Mesa Redonda de Palenque.
[1976], pp. 2 1 1 -2 4 ). g k ]

by F. G.
(A RaCross at
Part III

III. 16
Aspects of Classic Maya Rulership
on Two Inscribed Vessels

Thus, 9-17.11.5.19, unstated but implied, may be

taken as the date com m em orated in the scene be
low the rim. We may also take it as near the time of
the execution o f the carving, if we can prove that
the scene is historical rather than mythical.
The inscription reads from left to right, as is
usual with writing on Maya vase rims. But the
scene below it turns leftward, or clockwise, with
the result that any correspondence between the
glyphs o f the rim and the persons below is limited
to the beginnings o f both series, that is, to corre
spondence only of the opening Glyphs 13 with
person A below them.
The persons are discussed in the inscription, for
Glyphs 1921 o f the caption flanking C recur as
Glyphs 810 on the rim. The inscription on the
rim, however, begins with a Distance N um ber and
Period-Ending date in Glyphs 14 above the head
o f the lady at A (fig. I I I - l57), who is named in a
caption o f two glyphs (17 and 18) in the space
before her face. She sits cross-legged and in frontal
position, with only her head and arms shown in
profile. H er body covers part of the back o f a serpent-throne.
This throne is shown m ore completely as the
regal seat o f the bearded man at B (fig. III-158),
behind her and to her left. It closely resembles the
skeletal heads on Lintel 25 at Yaxchilan and on the
front o f Stela D at Copan (fig. III-170). Having no
glyphic identification, and being shown with body
in profile, figure B appears to be o f lesser impor
tance than A, but the fact that he is enthroned and
perform ing the same gesture o f displaying a gly
phic com pound suggests that he is o f equal rank.
In front of the lady at A and to her right, in the
Maya position o f honor, sits the man at C (fig.

A SCENE A N D A T E X T O N R U L E R S H IP :
A N O N Y X M A RB LE BOW L A T
D U M B A R T O N OAKS
Three seated figures carrying glyphic insignia are
incised on the wall beneath an inscription on the
rim of an onyx marble bowl (figs. III-157 to
III-l5 9 )1 which was carved in the southwestern
Maya regin. The bowl is now in the R obert
Woods Bliss Collection at D um barton Oaks.
S. K. Lothrop (Lothrop etal. 1957, pp. 2 5 9 -6 0 )
thought that the bearded figure at B (fig. I I I - l58)
outranked the others, that the woman at A (fig.
III-157)2 was a man in a skirt, that the man with
glyphs by his head at C (fig. I I I - l59) was least
important in rank, and that the scene as a whole
portrayed the ceremonial offering o f religious
symbols representing the earth. H e also interpreted the inscription as a functional one referring
to the day Imix and the N o rth Star god and
fixing the date as after 9.15.0.0.0 (a.d . 731) and
before 10.8.0.0.0 (a.d . 987), possibly at an inter
medate Period-Ending. All these suggestions
about the vessel stand today in need o f further
discussion.
The date (fig. II1-160) consists o f a Distance
N um ber (Glyphs 12: 13.12.1)andaP eriod-E nding statement. If Glyph 3 specifies the m onth Ceh
(Thompson 1950, fig. 23, 4; 1962, p. 186)3 and a
day, 4 Ahau, equivalent to 9-18.5.0.0 (followed at
Glyph 4 by a forward-count indicator, signifying
that the Distance N um ber is counted from an unstated date to the Period-Ending statement), then
(9.18. 5. 0. 0)
13.12. 1
(9 .1 7 .1 1 .5 .1 9 )

(13) Ceh 4 Ahau

(Sept. 1 5 ,7 9 5 )
(Feb. 23, 782)

334

Aspects o f Classic Maya Rulership

111-159), about whom the inscription on the rim
appears to revolve. H e is frontally seated like the
woman, and, like the others, he displays a glyphic
head.

The Glyphic Objects

Each of the three persons displays on an outstretched hand an object that is less like a thing
than a glyph. If the lady at A begins the clockwise
series o f three, we may identify the glyph in her
hands as the first in a sequence; the glyph held in
both hands by the bearded man at B is the second,
and the object in the left hand o f the man at C is
the last in the sequence. The glyph in the hands of
the person at B is a com pound consisting o f the
numeral 7 prefixed to a skeletal, long-nosed head
(T1035), which bears a kan-cross glyph (T281) as
its superfix; the postfx is a glyphic serpent head
with recurved snout (T794). The bearded man
may be engaged in pulling the serpent head from
its attachm ent to the 7-head, an effort conveyed by
his raised left shoulder and tense wrist. Behind the
bearded man is the named youth at C, who dis
plays the serpent head (T794) in an upright position with an em bellishm ent or fringe o f beads.
If this interpretation o f the action is correct, the
bearded man at B (fig. II1-158) is holding an abbreviated or shortened Cerem onial Bar consisting
only o f front (TV II.1035) and rear (recurvedsnout serpent) heads. The latter insignia reappear
in the hand o f the man behind him at C (fig.
III-159), who is, therefore, the recipient o f an
important charge and a person o f im portance, as is
marked by the glyphs surrounding him in his p er
sonal caption.
Given the above, the scene may represent a ruling lady, A, with her consort behind her at B; the
consort shares with the prom inently identified
person at C one part o f his insignia, in a rite that is
comparable to the ascensin to power identified
at Piedras N egras and elsewhere by Proskouri
akoff (1960). Specifically this transfer concerns a
recurved serpent-head insigne which appears to
have been detached from the 7-head in Bs hands.

The Inscription
W hether the foregoing interpretation of the scene
is confirmed by the inscription (fig. III-160) re
mains to be tested. Following the chronogram in

335

Glyphs 14a, there is the head o f an od man at 4b

(T10l4a), which may imply ancestry and begin a
noncalendric statem ent (Proskouriakoff 1974, p.
172).
Several themes o f sacrifice follow in Glyphs 5
and 6. Barthel has read Glyph 5a (1966, pp. 105
06) as sacrifice by decapitation. Lizardi Ramos
(1947), followed by Proskouriakoff (1974, p.
172), gives sacrifice as the meaning o f 5b. Affix
T 6 l and skull ( T I040) in Glyph 6 again imply
sacrifice and death, while the infix of quincunx is
taken by Thom pson (1972, p. 152) to signify
offering.
Closing the clause, Glyph 7 may allude to the
woman at A in the scene below as the subject of
these references to sacrifice and death. It may be
noted that both she and the bearded man at B are
seated on skeletal serpent jaws which desgnate a
celestial locus, as on the serpent jaws of the C ere
monial Bar carried by the ruler on Stela D at
Copan (fig. III-170).
Glyph 8, which begins another statement, ap
pears in the ame caption o f the man at C, as well
as on a shell discussed by John Graham (1971, p.
157), who regards the com pound on the shell, on
the D um barton Oaks bowl, and on the sarcophagus at Palenque as a title like dignitary or lord.
Glyphs 9 and 10 round out the ame caption,
which is repeated in Glyphs 1921 at C. This long
appellative phrase is followed at Glyph 11b by a
com pound that Barthel (1968, p. 126) reads as a
warriors title: it may refer again to the man at C,
perhaps m entioning the place w here he lived
(T168:518:130).
A final statem ent begins with a com pound like a
title at Glyph 12, including signs for day and
fiery, possibly designating the bearded man at
B. Glyph 13, with its skull (T1040), suggests
death. Glyph 14, evoking rattlesnake (T207) and
bat (T756), may be a personal ame. Glyph 15 has
the characteristics o f an unidentified emblem or
locative glyph with T 5 6 la as the main sign, and
Glyph 16 is the familiar expression am en or so
be it (Thom pson 1962, p. 96), marking the end
of a sentence.
In summary, the inscription tells in the first part
(Glyphs 17) o f sacrifice and death in relation to
the woman at A; this is placed below the date,
which may be that o f her death. In part two,

336

A ncient Am erica

Glyphs 8 -1 1 give title, ame, and place for the

man at C. Part three (Glyphs 1216) ames the
bearded man at B, whose death may also be noted
as having occurred at a different place.
M. Coe (1973, p. 18) believes that unstated
portions o f the sequence which he identifies as
Primary Standard appear on the two inscribed
onyx marble vessels at D um barton Oaks (Lothrop
etal. 1957, pls. 86, 87; von Winning 1963). These
would be the glyphs (fig. III-160) at4 b (G od N )
and 14b (bat). In figure III-178, they would be the
Palenque em blem and Glyph 1 (quincunx).
T h e G ly p h ic In sig n ia
All three persons bear in their hands related insig
nia which are actually glyphs, the distribution of
which, throughout the corpus of Classic Maya
sculpture, yields certain clues as to their meaning
and the meaning o f the scene as a whole.
The woman at A and the bearded man behind
her at B are displaying skeletal serpent heads of a
type classified by Thom pson (1962, p. 458) as
T1035. His examples wear T563 as a headdress or
superfix, but the serpent head (T1035) at A is prefixed with the numeral 9 and has T629 as a head
dress, and the serpent head at B is prefixed by the
numeral 7 and has T281 {kan-cross) as a head
dress.
As noted earlier, this serpent head also has a
postfix (T794), another serpent-head type with recurved snout, which reappears with an embellishment in the hand o f the man at C, possibly as part
of the insignia o f the bearded man at B. The embellished form o f T794 is known at Palenque, as
on pier C of H ouse D, the Tablet o f the Cross, and
the Sarcophagus (figs. III-161, III-132, I I I - l62).
The representations in these instances have been
provisionally labeled as tasseled corn showing
stylized rows ofgrains (K ubler 1969, pp. 40, 45).
Among the 7- and 9-numeral insignia now
known (fig. III-163), it is apparent that the T I 035
head appears in nearly all examples, the only absences being at Copan (Sculpture A", Structure
11; fig. III-1631) and at Piedras Negras (Stela 40;
fig. III-176). The 7-head commonly bears the kancross affix (T281), often augm ented by an affix for
mol(T580 or 581) and adorned with stylized cornplant leaves. T he 9-head commonly bears an affix

(T629) resembling a seashell or carapace, which,

in turn, bears other affixes such as T42, conveying
the idea of brilliance or sparkle, and T223, which
resembles footprints.
An early interpretation by H. J. Spinden (1924,
pp. 16368) o f the intended meaning o f the insig
nia proposed that the heads stood for baktuns, or
cycles o f 400 Maya years, commemorating, respectively, the seventh and the ninth cycles of
Maya history. Against Spindens view is the evidence o f Stela 7 (Glyph b 2 ) at Aguateca, where the
katun is identified by the numeral 7 as its Symbol.
Though it was never accepted, his view reflected
the dom inant interest in calendar mechanics then
governing Maya studies. Later on, Thompson
(1950, pp. 89, 214) proposed mythological explanations based on Colonial sources. T he 7-head was
described at Copn as an earth crocodile (ibid.,
fig. 12, 2) and as the jaguar god o f the underw orld (ibid., pp. 89, 134). In the same way, the 9head (ibid., p. 89) became a Chicchan, the celes
tial snake, in an interpretation contrasdng sky
and earth.
If, however, we analyze the examples and consider them in context, neither Spindens calendrical or Thom psons specific mythological explana
tion finds further confirmation.
T h e G ly p h ic A ffixes
The 7- and 9-heads are distinguished from one
another not only by their numeris but by their
affixes, which are kan-cross (T281) and mol (T580
or 581) for 7, and T629 (with T42 or T223) for 9.
The meaning of these affixes is far from clear;
kan-cross is related to maize, and mol is the ame
of a m onth, but the com pound is known to occur
only once (T850) in the Maya corpus: as a cartouche on Tikal Stela 31 at e 17 (fig. III-142),
where it accompanies the numeral 7. Kan-ctoss is
the main sign, and mol is the affix; the cartouche is
surm ounted by T359. The context immediately
before e 17 is D ate H (8.18.0.0.0, or a . d . 396)
given as 12 Ahau, end of haab, at e 16 f 16. Next
after e 17 is a cauacsign at f 18, written asT122:529
with a fame affix. This com pound is possibly
equivalent to the serpent head itself, T1035 (see
fig. III-172), but it is notable that no other text
makes use o f it, which suggests that the idea of an

Aspects o f Classic Maya Rulership

equivalence between 7 and cauac was m ore appropriate to insignia than to glyphic writing.
The same situation prevails with T629, which is
more common among the insignia than it is in
glyphic composition. At Copan on Altar T ' (fig.
111-164), this affix resembles the lunar sign
(T683). Elsewhere it is m ore like a crustacean shell
with a foliated outline. The Copan glyph appears
to take the form T628, as on Stela A at aIO , where
it serves as Glyph A o f the Lunar Series and ap
pears again at f6 with coefficient 9. O n two examples, moreover, T629 is hatched half-dark and
half-light (at Palenque on the Tablet o f the Cross
and at Copan on Altar T '). These expressions
and aspects support an interpretation o f T629 as a
lunar glyph o f undeterm ined function, possibly
marking a nocturnal aspect for the 9-head and implying a diurnal one for the 7-head, an interpreta
tion further supported by its association with corn
plants.

Glyphs or Objects?
The numeral heads on the D um barton Oaks bowl
are used not so much like objects as like glyphs
detached from an inscription: the numeris appear
to float in the atm osphere with little visible sup
port, and the other affixes obey glyphic rules,
without concern for pictorial realism. These are
more glyphic signs than pictures o f glyphic ob
jects, burdening the hands o f their supporters less
than their large size would suggest. Y et they are
held like objects and manipulated, as when the
bearded man appears to pul the parts o f his glyph
asunder, and when the man behind him holds his
portion pointed upward.
This ambivalence of being both glyphs and ob
jects reappears among the m onum ents and minor
arts in different ways. Some representations o f the
regala in the repertory are m ore like objects, and
others are distinctly glyphlike, such as those on the
incised obsidians from Cache 161 at Tikal, with
glyphs (T IX .78:33:629:1035 and TV II.78:
580.182:1035) correctly facing left as glyphs
should (fig. I I I - l65). Those shown as objects obey
gravity in resting on the earth or on a dais and in
serving as pedestals or seats, while those portrayed
as glyphs hover or float like words in a pictured

337

space. Still others are worn as parts o f costume or

carried as small effigies.
At Yaxha in the northern Peten, southeast of
Tikal, two stelae o f Early Classic date, which were
erected as a pair, show this ambiguity with clarity.
Stela 4 (fig. I I I - l66), dated about a . d . 465 (9.1.
10.0.0 2 katuns; Proskouriakoff 1950, p. 199),
displays the 9-head as a one-glyph inscription on
the framed background at the level o f the forward
knee o f the standing ruler figure. T he glyphic
compound has no connection with the costume of
the figure, or is it attached to the frame or ground
line.
Stela 2 (fig. III-167) is dated about a . d . 534
(9.5.0.0.0 2 katuns; Proskouriakoff 1950, p.
199). H ere the skeletal serpent-head glyph (T
1035) rests on the ground bearing a mol-kan affix
surm ounted by the numeral 7 and a corn plant.
The head faces the left toe o f the standing ruler; it
appears to be part o f the setting and an elem ent of
the pictorial space rather than an inscription.
Both stelae form parts o f a row o f five stones
along the east side o f Plaza C (fig. III-168), in front
of Structure X III. They flank the central Stela 3, as
though their rulers were paying homage to the
person portrayed there in the classic stance with
feet spread at a 180 angle (Morley 193738, vol.
III, p. 469). Stela 3 at Yaxha is otherwise illegible,
but the relationship between these three stelae of
varying Early Classic dates prefigures the mu
seum alignment on the G reat Plaza at Tikal, as
well as the pictorial composition of the Tablet of
the Sun at Palenque (fig. III-134), where repre
sentations o f the 7- and 9-head glyphs flank a cen
tral and dom inant image.
This comparison, furtherm ore, relates the
standing persons on the Tablet o f the Sun to the
D um barton Oaks bowl, in that the main person at
C has the 7-head glyph to his right and the 9-head
to his left, as at Yaxha. At Palenque, too, the sunshield panoply o f Lord Pacal is flanked on its
proper right by the glyphs o f the 7-head and, on its
left, those o f the 9-head, in cise association with
such regala as shield, lances, Cerem onial Bar, and
proffered images o f deities.
The comparison, however, does not clarify the
ownership o f the 7- and 9-heads: are they offered
to the ruler o r are they the insignia o f the rulers

338

A ncient Am erica

associates? O n Stela 2 at Tikal and Stela 88 at Calakmul (9.11.0.0.0 ?, or c. a.d . 652; Proskouri
akoff 1950, p. 185), this question is resolved in
part by ruler portraits wearing or holding both
heads (9 on the left and 7 on the right o f the spectator) like twin infants in both arms. At Tikal on
Stela 2 (fig. III-169; 9.3.10.0.0 2 katuns, or c.
a.d . 504; Proskouriakoff 1950, p. 195), the 7- and
9-heads appear to rest like epaulettes on the shoulders o f the standing ruler, who is burdened with an
immense Cerem onial Bar bending around to the
sides o f the stela. The 7-head on the rulers left
shoulder is at a lower level than the 9-head on his
right side; here, as on Stela 2 at Tikal, both the
height and the dextral position o f honor suggest,
but do not prove, a difference in importance.
At Calakmul (R uppert and Denison 1943, pl.
53a) the 9-head is held higher, at the level o f the
rulers mouth, while the 7-head is carried below
the rulers left shoulder.4 This presentation, without explaining their origin, suggests that the two
heads can become the rulers insignia, and also that
the heads were possibly o f different rank, 9 being
held higher than 7 and in the place o f honor.
The incised obsidians (fig. III-165) from Cache
161 at Tikal are provisionally assigned by W. Coe
(1970) to c. 9 -14.0.0.0 (a.d. 711). For the present,
they seem to be the last known occurrence o f the
7- and 9-head them e in the Peten. It is also noteworthy that this style o f presentation both p re
cedes and follows the hiatus (Willey 1974), as if,
at Calakmul, a ritual stance and costume reasserted the authority of older tradition at Tikal.
Stela D at Copan (9 .15.10.0.0 2 katuns, or c.
a.d . 544; Proskouriakoff 1950, p. 188) displays
the heads on a pedestal formed by a skyband
near the base on the sides of the stela (fig. III-171).
The heads are distinctly glyphic, but they also rest
like ob jects on a pictorial ground line, further confirming the idea o f the heads as a pair belonging
among the insignia of rulership. H ere the 7 is on
the left side o f the ruler and the 9 on the right;
both heads face the rear, or north.
The heads are differentiated not only by the
numeris and superfixes, but also by different vegetation. As expected, corn leaves and grains ap
pear behind the kan-cross of the 7-head, but above
the 9-head are leaves and a seed pod that may
relate to another food-plant, such as ramn (Bro-

simum alicastrum) or cacao. If so, both 7- and 9heads signify nourishm ent, among other meanings. These references to vegetation, and particularly to maize, may reinforce the nature of the
masked person portrayed on the south face (fig.
III-170), who possibly im personates a vegetation
deity analogous to the Mexican X ipe Totee.
The association at Copn of the 7- and 9-heads
with such a vegetation-renewal image, masked
in human skin, recalls a similar association at Pal
enque about twenty years earlier. Skeletal serpent
heads o f the T1035 type appear there on the Tablets of the Sun, Slaves, and Palace (fig. III-172),
resting on the hands of officiants as effigy-head
offerings tendered to, or received from, rulers. In
each, the effigy is related to a X ipe shield
(T537). The skeletal effigy heads have cauac droplets and the skulls are like foliated cauac signs of
the T528 type. These, in turn, recall the shape of
T629, if seen in profile. T629 is the distinguishing
affix o f the 9-heads. W hether these cauac-head
effigies can be proven to represent 9-heads is
doubtful, but Schele (1974, p. 49) associates them
with the giving o f pow er for kingship in Palen
que, which supports the conclusin reached here
for the meaning o f the 7- and 9-heads. The early
inscription on Stela 31 at Tikal (fig. 111-142) also
may support this association.
Stela 1 at Cancun (fig. III-174), upstream on
the Ro Pasin and south of Seibal, is dated
9.17.10.0.0 2 katuns, or c. a.d . 780 (Pros
kouriakoff 1950, p. 186). T he female ruler seated
on a dais has at her right knee a 7-head facing her
and resting on the dais as an object, like a vase at
her side. The affix is the customary kan-mol
group, surm ounted by T I 30 and the numeral 7.
The association here with a woman is the opposite o f that on the D um barton Oaks bowl,
where the woman bears the 9-head, thus invalidadng the idea that the heads might be male- and
female-associated. The early precedent at Yaxh
(figs. III-166 to III-168) leads to the supposition
that another (lost) stela may have shown a 9-head
in a parallel or symmetrical position flanking a
central m onum ent at Cancun.
This association at Cancun o f the ruler with a
nearby object suggests a context for the vessel of
late Manik III date (Coggins 1975, pp. 23842
and Appendix 18) found in Burial 132 at Tikal

Aspects o f Classic Maya Rulership

(fig. III-173) as an heirloom made about 9.6.0.0.0,
or a.d. 554. A central exterior quatrefoil cartouche contains the 7-head with kan-cross affix,
surrounded by five glyphs: these ame women at a
and b; a skull is at c; the glyphic com pound for
sacrifice and disease appears at d (Barthel 1963, p.
189); and at Eis a head glyph referring to lordship,
disease, and reptilian nature.

A
B
C
D
E

T l9 .1 0 0 0 f [528]
5 9 .1 0 0 0 ?
t 40? 1040
t

758. 5 0 6 :1 1 0 .1 9

5 9 .1 9 : 1 0 1 3 { 5 33 }

appellative?
title?
death
sacrifice, disease
title?

These five glyphs indcate that the bottom o f the

bowl was to be seen facing up, because the glyphs
on the vessel wall would be upside-down if the
bowl were placed base down. At the Morley M u
seum in Tikal, the bowl is accordingly displayed
bottom up. Possibly another bowl, carved with the
9-head and a different inscription, was its twin
companion, m eant to stand on its bottom and to
form with the surviving 7-head bowl a closed con
tainer like the one shown on the wall o f the Initial
Series vase o f Tepeu date from Uaxactun (fig.
111-179), as an offerin g o f two bowls tied together
for presentation to the seated ruler dressed in the
costume of the jaguar-serpent-bird o f Teotihua
can (Kubler 1967, table, no. 44).
The final example is Stela 1 (fig. III-175) at
Aguas Calientes (9.18.0.0.0, or a.d. 790), the
latest now known in our series. The front shows a
standing armed ruler, but the back is either damaged or incomplete. The most prom inent form is a
7-head with a kan headdress and a mol glyph in its
mouth, tilted above and behind an animal m outh
containing a human head. At the top is a kin sign.
The presence o f the 7-head suggests, as in other
contexts, an aspect o f the rulership commemorated on the stela.

Pedestals (Misnamed Altars)

One function o f the 7- and 9-heads was to specify
the meaning o f various places platforms, stands,
and thrones where authority was publicly exercised. A graphic example on the Tablet of the
Cross (fig. III-132) at Palenque (c. 9.14.0.0.0 2

339

katuns, or a.d . 721; Proskouriakoff 1950, p. 192)

shows the celebrant on the left side standing upright on the 9-head glyph, which is represented as
a pedestal. The head itself, which rests on a glyphic
band inscribed with signs of night, moon, and
Venus, was recently interpreted (Kubler 1974b,
pp. 10912) as designating a place above the
clouds or a high country. For the present discussion, this notion should be enlarged by the idea of
a pedestal (Clancy 1976), like the circular or quadrangular altars which are common at Classic
Maya sites. A ltar generally means a sacrificial
table in Od W orld religious usage, and the term
has little or no relevance to Mesoamerican ritual,
although it has been used uncritically from the
time of the archaeological writings of the early
travelers.
An example o f such stone pedestals is the Late
Classic Altar T ' at Copan (fig. I I I - l64), described by M orley (1920, pp. 3 7 5 -7 6 ) as a small,
drum-shaped stone 30 cm. high and 49 cm. in diam eter, having the periphery completely covered with four glyphs, one in each quadrant, and
all alike. Morley thought in 1920 that the affix
might desgnate Cycle 9, but he later rejected the
idea. It is here proposed that the 9-head on the
Tablet of the Cross at Palenque (fig. I I I - l32) is in
the class o f glyphs representing actual objects, and
that the officiant at Palenque is standing on a ped
estal inscribed with the 9-head glyphic compound,
of the type preserved at Copan.
Stela 40 (fig. III-176) at Piedras N egras (c.
9.15.0.0.0 2 katuns, or a.d . 731; Proskouri
akoff 1950, p. 193) may allude to the same use of a
pedestal to proclaim authority. The glyph in
scribed on the base o f the dais at the bottom o f the
scene records only the numeral 9, followed by the
affix T629 capped with T223. The skeletal head is
lacking, possibly being regarded by the scribe as
redundant. The image conveyed by these affixes is
that of the footprints o f a ruler standing in the
privileged 180 stance on the still-unexplained 9head affix T629-

Seats
At Copan and Quirigua, persons of high rank, as
indicated by their rich costumes, are often shown
seated on large glyphs, as on the carved doorway
steps o f Tem ple 11 or the sides of Altar O at

A ncient Am erica

340

Copan (Robicsek 1972, pls. 165, 171 and 72). The

glyphs are seats, depicted in the ambivalent manner discussed above, and, like the pedestals (altars), they may be construed as places of
authority.
Altar T at Copan (fig. III-l 77) is o f this type,
and it could have served either as a seat or as a
pedestal, being 6 1 -7 6 centim eters high. T he top
surface measures 1.27 X 1.90 meters, and it is
ampie enough to have served, if needed, as a dais
for several people. It com memorates a date in a . d .
782 (9-17.12.5.17). Carved in low relief on the top
and sides, an alligator skin with human hands
spreads out over the length and width of the block.
The snout hangs down on the wide south face
(Maudslay 1 889-1902, vol. I, pls. 9596), and the
tail on the north. The rear legs hang down over the
short east and west sides. The faces and top o f the
stone are also carved with reliefs of one parrot and
nineteen seated persons, some o f whom impersonate animals. In the place o f honor, at the right o f
the alligator snout on the south face, is a relief
rendering o f the 7-head shown as a seat for the
chief person among four principis. All four are
seated on large glyphs, and a vertical inscription
divides them into two pairs. The meaning o f this
elabrate program is unknown. Thom pson (1970,
p. 220) connected it with the Mexican belief that
the earth was the top of a saurian monster, and
Morley (1920, p. 335) emphasized its relationship
to the Maya numerical system, but neither expla
nation is irrefutable in Maya iconography.
If historical material is present, as we may de
duce from the Calendar Round dates (which are
non-Period Ending) and from the individual portraiture o f the human beings who are shown, a
comparison would be in order with the D um bar
ton Oaks bowl, where the dedicatory date is only
358 days earlier and w here the bearded man bear
ing the symbol of a 7-head precedes a younger
one. A nother possible comparison would include
the Saenz throne (Easby and Scott 1970, no. 174),
which portrays a woman and a bearded man as a
ruling couple, and is later by seventeen years.

Conclusions
That the place o f origin o f the D um barton Oaks
bowl is Campeche may be doubted. The configu-

ration of historical text and portrayed persons resembles m onum ental relief sculpture m ore than
pottery painting. The scene and its text seem to
derive from or be related to standard types of his
torical record in the Late-Classic period. Both text
and persons resem ble, in drawing and glyphic
characteristics, the sculptural style o f the lower
Usumacinta regin. T he bearded man at B (fig.
III-158) resembles the figure, accompanied by the
emblem glyph for Palenque, on another onyx marble bowl (fig. III-178) said to come from Jaina
Island. The bowl is dated by von W inning (1963,
p. 118) as 9.16.0.0.0-9.19-0.0.0, or a . d . 751810. The man and the woman on the Dumbarton
Oaks bowl, seated on skeletal serpent-head
thrones, may be forebears or ancestral figures like
those in the upper medallions on Stela 10 at Yaxchiln, or on the palace walls o f H ouse A at Palen
que (Maudslay 18891902, vol. IV, pl. 6), where
the same skeletal serpent snouts appear. These
enthroned figures on the bowl may be forebears of
the third person, who carries an insigne familiar
from the reliefs of Palenque as the tasseled maizeserpent em blem (fig. I I I - l72). These resemblances are not enough to fix beyond doubt the
precise provenance o f the figural style of the
D um barton Oaks bowl, but they suffice to identify
its general regin.
Table III-12 shows that the glyphic T1035 skeletal-serpent heads, with coefficients of 7 and 9 and
corresponding affixes, were in use as images
throughout the Southern Maya lowlands during at
least four centuries before a .d . 800. Their use
spread first northward from the north Peten, then
westward to Palenque, and, later on, southeastward to Copn. T heir meaning can be approached
by a review of the contexts in which they appear.
In general, the 7- and 9-heads, w hether singly or
paired, may be interpreted as enhancing the depiction of rulership; like Selers triadic glyph (T272),
they are related to the rulers agricultural duties
(Kubler 1969, PP- 3346; G reene Robertson
1974), but with the difference that the T 1035
heads do not appear in headdresses. They were
used as glyphs, as regalia, as seats, and as portable
insignia, but not as elem ents of costume, perhaps
because they retained their identity as glyphic ex
pressions related to rulership as such, rather than

341

Aspects o f Classic Maya Rulership

Table III-11.
Transcription o f the G lyphs on the Onyx V essel
1. Unilinear rim inscription
1
1 kin 12 uinal ! D N 13.12.1 from
stated D D (5
2
13 tun
Cauac 2 Ceh
9 .1 7 .1 1 .5 .1 9 )
m onth Ceh
3a ) 1:44:563{153]
4 Ahau
3b 1(9 .1 8 .5 .0 .0 )
count forward
4a 561a
od man
4b 1014a
decapitation?
5a 61 ?: 7 56
sacrifice
5b 568c:4?
sacrifice affix
6a 61
skull quincunx
6b 7 4 :1 0 4 0 [5 8 5 ]
7a 1000b :87:188 1
ame o f woman
7b 25:1004 b ?: 130 j
1
f 8a 12
ame or title
cf. 19
{ 8b 35:1016:23 j
disease
9a 18
cf. 20
501:23
l 9b
| 10a 126:683
cf. 21
1 10b 35:1016
l i a 204 ?: 129?; 133?
locative?
11b 168:518?: 130
184?
title?
12a
ame
12b 67:563:25
13a 63:758
skull
13b 32:1040?
14a 23
14b 207:756
15a 32
locative?
15b 168:561a?
16a 501 1
ending phrase
16b 501? J
II. Caption for woman at A
17
1002a 1
ame
18a 1004?
18b 765? )
III. Caption for man at C
1 19a 12
cf. 8
1 19b 33:1016:23
disease
| 20a 18
cf. 9
1 20b 501:23
| 21a 126?:683
cf. 10
121b 33:1016:23
74:528:1016
count o f the year
22

(baab)

to its functions and duties. Some evidence exists

for an interpretation that places the 7-head in a
diurnal aspect and the 9-head in a nocturnal or
lunar character, as if to mark rulership o f night and
day by the persons using both these insignia.
The known distribution and chronology o f the
7- and 9-heads, as outlined above, make it unlikely
that the D um barton Oaks bowl would have been
fashioned outside the Southern lowlands, or after
its dedicatory date (9.18.5.0.0, or a . d . 795). In
addition, no evidence is now known to contradict
a . d . 782 as the terminuspost quem for its manufac
ture. To present knowledge, the Dumbarton Oaks
bowl is among the latest known examples of its
iconographic type, a type that spans nearly four
hundred years. W herever it appears, it seemingly
refers as an emblem to rulership itself. Furthermore, the D um barton Oaks bowl informs us about
the recurved serpent-maize head (fig. II1-159),
which enlarges knowledge both o f rulership emblems and o f the possible meaning o f scenes at
Palenque (figs. III-132, III-161, III-162), where
that form also appears. Finally, the Dumbarton
Oaks bowl supports the argum ent (Kubler 1969,
pp. 4 7 -4 8 ) that structural relationships among
meanings are clearer in late examples than in less
fully articulated early examples of the same ico
nographic theme.

THE INITIAL SERIES VASE FROM

U A X A C T U N RECONSIDERED
Introduction
The accepted views about the Initial Series vase
(fig. III-179) from Uaxactun (Morley 193738,
vol. I, pp. 23033, and later writers) hold that it
was made and painted during Tepeu I (c. 9.12.
0.0.0, or a . d . 672) and that the inscription really
refers, in M orleys em endation, to 8.5.0.0.0 ( a .d .
146). The corrections proposed here are that the
date of the vase be moved to Tepeu II (c.
9 .1 6 .8 .1 1.0, or a . d . 759) and that the inscription
be accepted as stating 7.5.0.0.0 (248 b .c .) and as
having to do both with ancient history prior to the
em ergence of Classic Maya civilization, as well as
with history as it was happening when the vase was
painted. Clemency Coggins (1975, p. 508) thinks
its multiple-resist technique may come from the

342

A ncient Am erica
T a b le 111-12.
7- and 9-H eads in Chronological Sequence (see Fig. I I I -l63)

a
b
c
d
e
/

g
b
i
j
k
l
m
n
0
P
<7

a . d . 396
46540
50440
54440
before 554
65240
711 ?
72140
72140
73140
7 4 l 4 0
78040
78040
78040
782
793

Tikal, Stela 31, f.17, D ate h

Yaxh, Stela 4
Tikal, Stela 2
Yaxh, Stela 2
Tikal, Burial 132, bowl
Calakmul, Stela 88
Tikal, Cache 161, incised obsidians
Palenque, Tablet o f the Sun
Palenque, Tablet o f the Cross
Piedras Negras, Stela 40
Copn, Stela D
Copn, Structure 11, Sculpture A"
Cancun, Stela 1
Aguas Calientes, Stela 1
Dumbarton Oaks bowl
Copn, Altar T
Copn, Altar T'

northwest Petn. If ancient history was portrayed,

it was, as in Shakespeares histories, in the dress
and conventions o f the times o f the artist.
W riters o f Maya archaeology say little about
what the iconographer calis the programmatic intention o f burial assemblages. It is generally assumed, without proof, that pottery in burials was
made for funerals rather than for daily use. Abundant contrary evidence, however, appears in fragments of painted figural designs on sherds among
the rubbish near dwellings and in the rubble fll o f
substructures and vaults (R. E. Smith 1955, vol. I,
p. 13; Adams 1971, p. 31). The position taken
here is that fine pottery was made for the use o f the
living, and that meaningful selections from among
the household collection were made by the survivors for tomb furnishings when the owner died.

The Glyphic Inscriptions: The Rim

The rim inscription (fig. III-179), which has never
been discussed in detail, follows closely the gly
phic sequence called Primary Standard by
Michael Coe (1973, pp. 18-22), who thinks that
such glyphs describe an episode from the Popol
Vuh about the descent o f the H ero Twins to the
Underworld . . . (ibid., p. 22), ending with their
apotheosis and the possibility o f the same event

(8.18.0.0.0)
9 .1 .1 0 .0 .0 2 katuns
9 -3 .1 0 .0 .0 2 katuns
9 -5 .1 0 .0 .0 2 katuns
before 9-6.0.0.0
9 .1 1 .0 .0 .0 ?
9 .1 4 .0 .0 .0 ?
9 .1 4 .1 0 .0 .0 2 katuns
9 .1 4 .1 0 .0 .0 2 katuns
9 -1 5 .0 .0 .0 2 katuns
9 .1 5 .1 0 .0 .0 2 katuns
(9.17.10.0.0?)
9 .1 7 .1 0 .0 .0 2 katuns
9 .1 7 .1 0 .0 .0 2 katuns
(9 .1 7 .11.5.19)
(9 .1 7 .12.5.17)
>

for the soul o f the person for whom the vase was
painted.
Y et the variety o f pictorial subjects accompanying this sequence o f glyphs, and its appearance on
many vessels w ithout scenes, also permits the
reading o f the sequence as a loose group o f ritual
phrases that refer to vows or dedications used by
professional pottery craftsmen, in the sense of
dedicating the vessel itself to a person or institution or o f offering the work to those supernatu
rals such as spirits o f fire and earth who are
like patrons to potters and whose mighty aid is
implored.
On the rim o f the Uaxactun vase, Coes se
quence begins at Glyph 8, continuing in Coes
order through Glyphs 10, 18, 2, 3, 4, 5, and 6. Six
more glyphs correspond to those o f Coes se
quence, but their o rder is different from his. Thus,
o f eighteen glyphs, fourteen correspond to those
o f C oes Primary Standard inscriptions. In the
scene beneath the rim, the two standing blackface
figures would be interpreted by Coe as Hero
Twins, but they also closely resem ble the ruler
figures of N aranjo on Stelae 30 (fig. III-l 80) and
33 (fig. III-181), as well as the courtly attendants
on Tikal Altar 5 (fig. III-182). O n Lintel 2 in
Tem ple III at Tikal (fig. III-183), ruler and atten-

Aspects o f Classic Maya Rulership

dants all carry trident flints, as do two figures on
the Initial Series vase. In all these parallels, the
accompanying inscriptions are unmistakably historical rather than mythological, reinforcing the
impression that the blackface figures are historical
persons related to N aranjo and Tikal, who are either rulers themselves or participants with rulers
in a ritual using trident knives.

The Initial Series

As painted, the 7.5.0.0.0 (8 Ahau 13 Kankin) date
was unacceptable to M orley (193738, vol. I, pp.
23031), who thought it contained scribal errors.
H e therefore changed it, on the principie of the
least num ber o f corrections, to 8.5.0.0.0 (12 Ahau
13 Kankin). Morley, however, was primarily con
cerned with calendrical structure and its consistent
epigraphic interpretation. H e died before the his
torical content o f the inscriptions began to be deciphered by H einrich Berlin in 1958 and Tatiana
Proskouriakoff in 1960, 1963, and 1964. In the
light o f those discoveries, it is now necessary to
reconsider the Initial Series date as marking the
end of the first quarter o f Cycle 7, as well as referring to an actual event in Maya history.
If 7.5.0.0.0 is correct, to what event could this
date have referred? It is possibly the beginning of
the last katun o f an older era, before the inaugura
ro n o f a new Initial Series Base D ate (or starting
point) at 7.6.0.0.0 (11 Ahau 8 Cumhu), which
John R. Teeple (cited in Thom pson 1932, p. 370),
shortly before his death, presum ed to be the
beginning o f the Long Count. T eeples argument
was that twice 73 katuns in the Calendar Round
count had elapsed since the older starting point at
13.0.0.0.0 (4 Ahau 8 Cumhu), when some 365
leap days that had accumulated over 1,460 vague
years of 365 days were finally used to correct the
divergence between the seasons and the Calendar
Round. T he new start at 7.6.0.0.0 also obtained an
historical base with the same m onth position as the
older one (8 Cumhu), and, according to Maya calculations, the same position in the solar year as at
the 13.0.0.0.0 base on 4 Ahau 8 Cumhu, 3113
B.C.
Thom pson (1950, p. 152) accordingly reconstructed this early history of the Calendar Round,
the Initial Series, and the Long Count as follows:

A
B
C
D
E

13.0.0.0.0
7.0.0.0.0
7.0.0.0.0
5.0.0.0
_________
7.5.0.0.0

343

base; last creation of the world?

theoretical passage o f time
inauguration o f CR
time elapsed between the inauguration o f CR and IS
inauguration o f LC

Thom pson regarded B and D as guesses, but he

believed E to be the very earliest at which the LC
could have been inaugurated. Thom pson did not
mention the Initial Series vase in this connection,
perhaps because he had always accepted M orleys
em endations o f its inscription; but the coincidence
is striking, and it may help in explicating the meaning of the scene on the vessel.

The Scribes Purpose

Morley observed o f the vase we are discussing
(193738, vol. I, p. 228) that the Initial Series of
two columns o f eight glyphs each is the middle of
the com position and the center o f esthetic interest since all other elem ents o f the design the
five human figures and the one animal figure
face toward it. The inscription would, therefore,
have been under closest scrutiny in ancient times,
and it is unlikely that an error o f 400 Maya years
(tuns) would have been tolerated. It is also un
likely that Maya priests, nobles, and artists would
not have known the meaning in detail of any Initial
Series statem ent. We must, therefore, begin again
with the inscription as painted, avoiding easy recourse to correcting the scribe for illusory errors
without examining other possible interpretations.
For example, it is evident from G oodm ans
1897 tables that 8 Ahau 13 Kankin lies only thirtyfive tuns before 9.5.0.0.0, at 9.3.5.0.0, as a tunending that cannot recur in either direction for
949 Maya years (that is, tuns). It is plainly more
laborious to calclate this tun interval than to ob
serve by m ere inspection that the cycle is wrong.
In other words, the likelihood o f scribal error diminishes with the larger units o f time; for the reader it is easier to detect errors o f centuries than
errors o f months and days. Conversely, it is less
likely that the scribe will make large errors than
small ones. Morley has pointed out (193738, vol.
IV, pp. 3 0 8 -0 9 ) that errors in Initial Series dates
were few (less than 0.27 percent), and he observes

344

A ncient Am erica

that their correction would have been very easy

but for a hypothetical taboo against changing wrirten records for any reason. Indeed, among eight
errors in 412 Initial Series then known, Morley
found only the Uaxactun vase as having any error
in the baktun order. This unique status makes his
judgment o f Cycle 7 as a mistake very doubtful,
and it encourages us to return to reading the in
scription as it is unmistakably written.
M odern scholars constantly correct Maya in
scriptions w ithout crediting Maya scribes with
being good at arithmetic. Y et it seems unlikely
that people so obsessed with calendrical arith
metic could have made gross errors o f calculado n
more often than our bank tellers do today. In fact,
so many errors o f calculation have been claimed
against the most carefully carved inscriptions that
it is tempting to seek another explanation than the
one merely showing off our own ability at summing up.

Telescoped Inscriptions?
N o one yet has thought much about considering
telescoped statem ents where two or more
things might be said at once as we could expect
when the effort o f writing anything is so laborious
as in Maya notation. O ne possibility is to examine
a wrong expression to see if it contains different
correct chronograms. For example, skilled Maya
arithmeticians, in a society that depended more on
memory than on writing, would not have balked at
a statem ent so contracted that both an ancient
event and a recent one were compressed into the
same Initial Series declaration. Such arithm eti
cians would have known by instant reference to
memory that 7.5.0.0.0 (or first quarter-cycle, recurring every 400 tuns) was inconsistent with 8
Ahau 13 Kankin because this particular Calendar
Round marked the beginning o f the Long Count
that was as familiar to all as the older Year o f the
Era, or 13.0.0.0.0 (4 Ahau 8 Cumhu), when the
world was last recreated.
Clues that now escape us might have meant a
separation between day count and Calendar
Round in order to condense two into one, such as
the first quarter-cycle o f baktun 7 (25 4 b . c .) with a
tun ending on 8 Ahau 13 Kankin, which could
occur after 254 b . c . only at 9-3.5.0.0, or a . d . 500.
That the scribe might have confused two refer-

ences is unlikely on the score o f the rarity o f large

errors as noted above, and it is equally reasonable
to suppose that erro r conceals a complex
intention.
It is also necessary to m ention that 9.3.5.0.0 (8
Ahau 13 Kankin) is cise to the dedicatory date of
Stela 3 at Uaxactun, and that this katun is commemorated thirteen katuns later on Stela 2 in the
same plaza (Morley 1 9 3 7 -3 8 , vol. I, pp. 212 -1 3 ;
vol. V, pls. 5, 62). By this interval o f 260 Maya
years between two rulers, at the end o f thirteen
katuns, history was expected to repeat itself
. . . this belief was so strong at times as to actually
influence the course of history (R. Roys 1933, p.
184). Such a period separates 9.3.5.0.0 from the
probable date o f the painting o f the Initial Series
vase (c. 9.16.8.11.0).

The Lunar A ge (Glyph D)

Further support for such redating o f the painting
on the vase comes from an examination o f the
lunar date o f 8 Ahau 13 Kankin as written. This
Calendar Round date can recur at intervals of
2.12.13.0 (18,980 days, or 52 vague years o f 365
days) on only eight dates in Cycle 9, o f which the
real lunar ages can be com puted (L. Roys 1945) as
follows:
a
b
f
d
<?
/
g
h

9.0.12.5.0
9.3.5.0.0
9.5.17.13.0
9.8.10.8.0
9.11.3.3.0
9.13.15.16.0
9.16.8.11.0
9.19.1.6.0

moon
moon
moon
moon
moon
moon
moon
moon

age
age
age
age
age
age
age
age

24.3
16.1
8.5
20.4
21.1
13.0
4.8
28.1

days
days
days
days
days
days
days
days

Among them all, only the italicized penultimate

one approximates the lunar age o f three days given
by Glyph D at a 6 on the Initial Series inscription.
Roys allows a range o f about 2.2 days for Maya
observational errors; henee date g comes nearest
the age noted on the vase, confirming the date
suggested by style and iconography below.

Late Imix Figural Style

A nother approach is to regard the vase as a composition reflecting m onum ental sculpture, such as

Aspects o f Classic Maya Rulership

stelae with various figures and inscriptions on the
front, sides, and back, like Stela 31 at Tikal (fig.
III-142). If this were so, the scribe may have condensed and shortened a long inscription by selecting the day count o f the Initial Series ( a 1 - b 4 )
from the beginning o f one long statem ent, and the
Calendar Round with Supplementary Series ( a 5
b 7 ) from another inscription or from a subsequent
portion o f the same model, thereby telescoping
and abbreviating the texts he was following.
On the principie that no work o f art is likely to
postdate its latest chronologically secure trait, the
Initial Series vase can confidently be reassigned to
the last quarter of Cycle 9. A t least three late traits
support this dating.
1. The appearance o f ruler portraits at N aranjo
with banded staff and tridentate knife (fig.
I I I - l80) dates no earlier than Stela 30 (9.14.0.0.0
2 katuns; Proskouriakoff 1950, p. 191), but
this ruler wears zserpent-he&d helm et and an Usumacinta-style apron. Stela 33, which portrays a
ruler at N aranjo forty or m ore years later
(9.16.0.0.0 2 katuns\ Proskouriakoff 1950, p.
192), has a banded staff, tridentate knife, and a
jaguar-head helmet. This triple cluster o f attributes is identical with those o f the second visitor on the Initial Series vase. Marcus (1971, p.
15) identifies the Stela 33 figure as a ruler o f Se
ries VI, G roup B; Mathews (1976, p. 48) ames
him as Ruler 4, and he places Stela 33 at
9.17.0.0.0.
2. The large feathered flabellum, or fan, shown
in profile representation, is known on reliefs at
Yaxchiln on the upper step o f the middle doorway (fig. III-184) o f Structure 44 (dated by Pros
kouriakoff {1963, table 1] to 9.15.0.0.0) and on
Lintel 16 (9-16.0.13.17; Proskouriakoff 1964, p.
198) as well as at Bonampak in the mural paint
ing o f Room 1, Wall 4 (fig. III-185), w here such
fans are carried by courtiers as o f 9.18.10.0.0
(Ruppert, Thom pson, and Proskouriakoff 1955,
fig- 27).
3. The angular scrolls on the back masks carried
by the regal visitors on the vase appear frequendy
at Piedras Negras after 9.15.0.0.0, as on Stelae 4
and 25 (see Proskouriakoff 1950, fig. 15). The
sandals o f these figures are o f a type seen only
about 9-16.0.0.0 at Xcalumkin. O ther angular
scrolls are at Quirigua on Stelae F and I (both

345

9-17.0.0.0 2 katuns; Proskouriakoff 1950, p.

194).

The Scene Described Anew

Although it was described with care by Morley
(1 9 3 7 -3 8 , vol. I, pp. 2 2 8 -3 0 ), the scene accompanying the Initial Series has never been fully interpreted in a m anner consistent with all its many
clues (and it is unlikely that this present attem pt
will suffice because each new approach to Maya
iconography opens further possibilities for
explanation).
The compositional and narrative balance is
clear: an enthroned figure sits on a jaguar-skin
stool, facing two rigid, standing visitors in the
place o f honor on his right. Two courtiers or attendants stand at his left. Between the visitors, an
open-m outhed jaguar, whose human legs and
arms betray him as a man in jaguar costume holding a package, perhaps bears the visitors offering.
Morley (ibid., p. 229) believed this to be composed o f two bowls which contained a human
skull, such as was found between two identically
shaped bowls in Structures E-II and E-III at Uax
actun. His jaguar-head mask resembles the rulers
headdress in the flamelike outlines rising from the
muzzle, and a tail similar to his rises from the
rulers headdress like a panache.
N oting the Tzakol (Early Classic) shape and
color o f the paired and tied package o f bowls held
by the jaguar-man, Coggins (1975, p. 240) offers
the following possible explanation o f the anachronism: the seated jaguar, who represents the Underworld, offers an unearthed cache bowl (made
about 9-5.0.0.0) . . . to the enthroned ruler, for
rededication in another interm ent. She notes also
that she believes this tom b at Uaxactun to be contemporary with Burial 132 at Tikal (ibid., p. 238),
which contained late Ik pottery together with ear
ly Ik cache materials. She concludes that this reuse o f cache materials once dedicated to the royal
founding ancestor o f the group would have been
appropriate for descendants who were resuming
ceremonial activity at this tom b. Such thoughts
may also be entertained about the composition of
the much later burials A2 and A4 at Uaxactun.
The garm ent worn by both men who face the
ruler is described by Coggins (ibid., p. 239) as an
apronlike cloak that dips low in front and is high in

346

A ncient Am erica

back, which she compares to a very different overgarment on the female ruler o f Stela 9 from La
Florida (fig. III-186) in the northw est Petn. M or
ley, in 1944 (in I. Graham 1970, p. 454), wrote
that the garm ent looked like a chasuble worn over
a skirt.
Actually the garm ent o f the visitors on the Ini
tial Series vase is neither an apron or a cloak, or
is it like a chasuble, which is a sleeveless mande
covering body and shoulders. (It is a chasuble
which appears on Lintel 3 of Tem ple IV at Tikal
[fig. III-155].) T he garm ent in question at Uaxactn closely resembles a Mexican tilm atl o f the type
worn by rulers and nobles at the time of the Spanish Conquest. It is here illustrated (fig. 1-63) from
the Tovar m anuscript of c. 1585, where it designates the m onth named for rulers and nobles as
Tecuilhuitontli. O th er examples from the Codex
M endoza (fig. III-187) are illustrated and described by Seler (1904, pp. 5 15-19). This garment
was a bordered rectangular cloth worn as a shoul
der mande, ded by the corners o f the short side at
the right shoulder and displaying the undergarm ent or loincloth at the opening on that side.
Although the two men on the Initial Series vase
wear such shoulder m andes, the painter did not
understand that the right arm of the wearer would
have to be covered by such a mande or be thrust
out around its edge. Instead, he drew the arm outside the tilmatl, as though it had come through an
armhole or slit. The undergarm ent is showing on
the rear figure, nevertheless; he wears a loincloth
apron that falls in front between his legs below the
mande fringe. The painter not only misunderstood this m ande, he also took liberties with its
relationship to the arms o f the wearer.
Both mantles share the prevailing jaguar symbolism of the vase in the squared pattern with
yellow frames, each containing three dots, a pat
tern also seen on the pelt of the seated jaguar-man
between them.
Separated from the visitors by the Initial Series
inscription, the enthroned ruler may be regarded
as part of a symmetrical arrangem ent flanking the
inscription. The axial center is the Initial Series:
both the seated ruler and the standing visitors can
be said to Ilstrate or discuss it. This situation has
no parallel among known relief sculptures, where

the Initial Series invariably frames, backs, or

flanks the ruler portrait. H ere the inscription is
the true compositional and narrative axis, and its
placing encourages us to consider it as the prin
cipal subject.
A lthough his pose recalls Oaxacan funerary
urns, the ruler is not a mummy or a statue, for his
posture is both pliant and active as he leans forward attendvely, raising his face to the visitors,
either to speak or to listen. It is possible that his
right hand is lacking: the stump of an amputation
may protrude from his forearm.
The short figure beside him bears a tridentate
knife pointed at the ruler. An object in his arms
may be another package like the one in the
hand-paw o f the jaguar-man. His importance is
marked by the fan-bearer to his rear, who shelters
him (from the sun?) m ore than the ruler is sheltered.
The rulers ame appears in a caption above his
face. The main sign is much faded, but the affix
T74 (Thompson 1962, p. 47), which Berlin (1968a,
pp. 13940) idendfies as part of the title of ruler
or lord at Palenque, remains plainly visible.
The ruler is identified also by his water-lily-jaguar headdress (T213:751b). Barthel (1965, pp.
151, 158) connects these signs with the beginning
o f the vague year at the m onth Pop (of which the
lily.-jaguar is a glyphic expression) and also with
resurrection, with a rain-and-corn complex, and
with wealth and abundance. The richest known
expression o f the water-lily-jaguar them e appeared at Tikal in Burial 196, where a jade sculpture six and one-half inches long and weighing
three and one-half pounds (Coggins 1975, p. 552)
accompanied the remains o f the younger son of
R uler A at Tikal (Jones 1977). This younger
brother o f Ruler B at Tikal died in 9.16.4.9.8
(Coggins 1975, table 6), about four years before
the Calendar Round recorded at Uaxactun on the
painted inscription, read here as corresponding to
9.16.8.11.0. If our view is correct, the Initial Se
ries vase is related to events at Tikal both by the
water-lily-jaguar symbolism and by the dates of
Burial 196 at Tikal, only four years before the date
on the Initial Series vase at Uaxactun.
The water-lily-jaguar headdress on the vase may
also mark this ru lers clan or lineage relationship to

A spects o f Classic Maya Rulership

the second standing visitor, who wears an upright
jaguar-head helmet. It, too, has a feathered surround, and the jaguars m outh holds a plumed ornament. Both jaguar headdresses may distantly reflect the Mexican highland symbolism o f the
jaguar-serpent-bird icn, during the ascendancy of
Teotihuacn, although the visitors upright-head
helmet conforms to Maya fashion more than does
that o f the ruler, which resembles parallel forms in
Teotihuacn figurines (fig. I I I - l88).
These insistent and redundant portrayals of jag
uar regala have been noted by Coggins (1975, p.
270) as becoming at nearby Tikal in Imix times
(after a .d . 711) a prerogative of the nobility and
the emblem of their pow er.
At Tikal, Altar 5 (fig. III-182) is dedicated to
the dead wife o f Ruler A at about 9.14.0.0.0 ( a .d .
711; Coggins 1975, pp. 4 5 4 -5 5 , 459, 550). H er
bones, piled on the floor, are flanked by kneeling
men carrying banded staffs and stone knives, of
which the one on our left is tridentate. Both attendants wear the eyeloops seen painted on the faces
of the visitors on the Initial Series vase. Coggins
has noted (personal communication) a resemblance betw een the unusual pleated capelet worn
by the altar figures and the enthroned Lord on the
cylinder.
Lintel 2, Tem ple III, at Tikal (fig. III-183)
shows the same attendants carved in wood more
than a. century later (9.19.0.0.0 2Vi katuns;
Coe, Shook, and Satterthwaite 1961, pp. 76-77).
Both carry tridentate knives and banded staffs
and flank an obese ruler in jaguar-skin clothing,
who also carries the banded staff and tridentate
knife. The attendant on the right side is in danc
ing posture. The absence o f death symbols is
striking. H ere the tridentate knives appear more
as ritual instrum ents than as the sacrificial knives
that Helfrich (1973, pp. 19, 13941, 173) who
insists on human sacrifice and ritual death to the
exclusin o f all other possibilities o f interpreta
tion claims they are.
The ruler is seated between a visitor and a
courtier (or priest?), both o f whom bear threepronged instrum ents of a type known in light-colored flint from excavations at El Bal (Thompson
1948, p. 40), and which here may be painted black
to denote that they are made of obsidian.

347

Such tridentate knives and banded staffs (see

figs. I I I - l80 to I I I - l83) were at first carried by
rulers on stelae at N aranjo (Stela 30, 9.14.0.0.0
2 katuns; Stela 33, after 9.17.10.0.0). At Tikal,
their use was not restricted to rulers: they are
borne by priestly attendants at a funeral rite shown
on Altar 5 and by the courtiers on Lintel 2 of
Temple III, w here the fat ruler also bears the
banded staff and tridentate knife.
This lintel lacks death symbols and it shows the
same pattern as the Uaxactn vase, with a ruler
figure between others bearing tridentate knives.
The scene is like a play acted out in a later production with the costumes and properties of the Initial
Series vase jaguar suit and head, banded staff,
tridentate knife, and flanking attendants with tri
dentate knives.

The Chronology of the Burial Deposits

The new dating late in Tepeu II, here proposed for
the Initial Series vase, now requires some review
of the chronology o f the burials in Pyramid E at
Structure A-I at Uaxactn. We have been talking
about Burial A2 as the most recent o f those on this
platform, among others which are earlier by several katuns.
The published excavation details suggest a possible revisin o f the burial sequence. The remains
on the center line o f A-I (Pyramid E) are likely to
be the oldest (fig. III-190). Crypt II, on the east
side, either accompanied or followed the principal
burial on the axis (fig. III-189). Crypt I, which
contained the Initial Series vase near the head of a
male skeleton 1.72 m eters tall, occupies the honored right side o f the central burial. Finally, Burial
A4 lies at the feet of both the axial deposit and the
burial in Crypt I. It lies with its head to the west at
the Southern entrance to Crypt I, which was walled
up, covering the large polychrome tripod pate labeled as Vessel 2 (fig. III-191).
Barthel (1965) believes the iconography o f Vessel 2 portrays a cosmic diagram o f upper and
nether worlds. It would be useful, in testing Barthels interpretation, to know how the vessel was
placed under the masonry in the doorway. If his
U nderworld scene in the lower half lay in Crypt I
and the upper half outside it, Barthels argument
would gain in credibility. This might be a portrayal

348

A ncient Am erica

of a dying human being plunging head down in

late-Gothic style among infernal jaguars into the
Underworld. In addition, the orientation o f the
pate with its upper half to the east would support
our proposal that both burials (A2 and A4) were
parts of the same event.
If so, the four burials may correspond to two
events: an earlier inhumation o f Tepeu I date on
axis and in Crypt II, and a burial o f late Tepeu II
date, about one century later, when Crypts I and
III were built.

The Date of the Tripod Pate

We need now to test the stylistic date of the tripod
pate, excavated from under the wall separating
the two crypts. The proof is incomplete because of
the differences between the conservatism o f mon
umental sculpture and the relative experimental
freedom o f vase painting, but, in the absence of
any general history o f Classic Maya painting and
drawing, proof must now rest again on relief sculptures dated by style. P roskouriakoffs (1950) tables and drawings yield only Tepeu II dates for the
latest forms:
1. the dancing posture, with lifted heels, of the
tall figure on the right, shown in frontal body and
profile head (9.14.0.0.010.0.0.0.0; Proskouri
akoff 1950, p. 26, IJO;
2. the profile walking figures on the serpent
path (9.14.0.0.0 ?9.17.10.0.0 2 katuns; ibid.,
P- 21, IB]_());
3. the banded staffs (9.14.0.0.0 2 katuns;
ibid., p. 99, X III: L3).
Relying again on the principie that its latestdated traits fix the period o f the manufacture o f an
object, we may propose Tepeu II for the pate as
well as for the vase.
A possible iconographic relationship between
the Initial Series vase and the tripod pate may
appear in the resemblance between the giant fig
ure at the right o f the pate and the small person at
the rulers left on the vase, whose importance is
marked by the large fan held over his head by a tall
attendant. This resemblance may be fortuitous,
but it is apparent in the drawings made for R. E.
Smith (1955, pl. 72b, f), which show both figures
ciad in vertical streamers or bands of dark mate
rial. It may be that the vase shows an im personator

and that the pate portrays a supernatural, whom

the courtier (or priest?) represents on the vase.
Actually the cise proximity of the two vessels
the vase at the head and the pate at the feet o f the
tall skeleton in Burial A2 suggests that they both
have to do with events o f his life.
Barthel (1965, p. 134) interprets the pate as
showing this world, above the horizontal and me
dial serpent body, and an underworld below it;
both worlds are dominated at the right by a stand
ing giant who holds a trophy head in his left hand
and a banded staff in his right.
Equally plausible, however, and equally difficult
to prove, is the resemblance of the composition to
that o f the serpent facade on the west building of
the N unnery at Uxmal, in terms of a symbolic
illustration of the concept o f Maya rulership. The
pate may also reflect another level, that of theatrical performances, exemplified by the figure of
an acrobat standing on his elbows and flanked by
jaguar-men. In the upper register, the two kneeling women who hold monkeys also suggest a folk
play. Behind the women two men in black body
paint tread the serpent path. The central one
moves in front o f a thatched shed or sheet of falling water. W hether the meaning is secular or religious cannot now be answered, although the occurrence, among the rim glyphs o f the upper half,
of a threefold repetition o f an inverted vase glyph
meaning wife o f suggests secular history or genealogy rather than mythology.

The Present State of the Problem

The suggestion that two different chronograms
are stated on the vase provides an epigraphic base
for Thom psons reconstruction o f the early history
of the calendar that had been denied by M orleys
emendation. N evertheless, the message conveyed
by these telescoped dates is not clear. Are we
being told that the enthroned ruler in his MexicanMaya jaguar helm et was com memorating the start
of the Long C ount at 7.5.0.0.0? O r is the message
only that this Calendar Round event, at 8 Ahau 13
Kankin, was counted from 7.5.0.0.0 by an elided
Distance N um ber rather than from 13.0.0.0.0?
In any case, 8 Ahau 13 Kankin can correspond
only to 9-16.8.11.0 in Cycle 9 because of the lunar
age stated in the inscription. This date is proposed

Aspects o f Classic Maya Rulership

here on stylistic and iconographic grounds as the
date of the Initial Series vase itself. The representation of a pair o f funerary Tzakol-period bowls in
the hands of the seated jaguar-man then best corresponds to 9.3.5.0.0 (8 Ahau 13 Kankin). The
purpose o f the scene, accordingly, may be to re
cord a ritual and to com m em orate the reuse of an
ancestral tomb by descendants of the founder thirteen katuns later, in one of the periodic reenactments, or new beginnings, o f history every thirteen katuns, that recur in Late Classic dynastic
records (as on Stela 3 at Uaxactun) and in other
possible calendric adjustm ents in Cycle 9 (Kubler
1976b).
The action on the Uaxactun vase involving
persons with tridentate knives and banded staffs
reappears two other times: earlier, in a deathly
context on Altar 5 at Tikal (fig. III-182), and,
later, on Lintel 2 in Tem ple III (fig. III-183),
where an obese jaguar-suited ruler stands between
attendants who, like him, bear tridentate knives
and banded staffs. At Tikal, about a hundred years
separate these scenes, making it likely that a ritual
act is represented. In this act, a ruler on Stela 33 at
N aranjo (fig. III-181) also took part, costumed in
a jaguar helm et and carrying a banded staff and
tridentate knife, at about the time o f the Initial
Series vase.
Three motifs on the Initial Series vase jaguar
helmet, shoulder mantles, and tridentate knives
suggest non-Maya origins.
The jaguar helm et on the ruler has a widely
opened muzzle which frames the whole head of
the ruler as at Teotihuacan (fig. III-192). This
m otif was rare among the Maya especially the
dislocated-jaw effect but it was in common use
in the Valley o f Mxico and in Oaxaca during the
Early Classic period. The Maya headdress using
jaguar forms differs by imposing the entire animal
head upright on the wearer, as on the standing
figure at the left of the seated jaguar (fig. III-179).
O ther examples are at Bonampak in the muris of
the battle scene (Room 2).
The shoulder mantles are also m ore familiar in
Mexican highland use than in Maya costume, al
though depictions o f the Aztec tilmatl (figs. 1-63,
III-187) are only o f Conquest date and later. The
portrayal here, in the mid-eighth century, marks

349

its earliest known appearance, but the painter has

misunderstood the relation of the wearers arms to
the mande, thereby marking it as an outlandish or
unusual garment.
Tridentate knives belong in the archaeological
classificadon o f eccentric flints that Willey
(1972, p. 207) explains as possibly having been
diffused to the Maya lowlands via a Teotihuacan
trade that began even prior to the Early Classic
period. The hand-hole and tridentate shape were
believed by Follett (1932, p. 388), followed by
Helfrich (1973, pp. 13941), to rep resen ta jaguar
pawand to have been used in sacrifical killings in
imitation o f the jaguars m ethod. But no Maya
representation shows any trident in use as such a
weapon, and other possibilities should be examined.
At N aranjo, the tridentate knife carried by the
ruler on Stela 30 (fig. III-180) bears on each o f its
rounded, droplike points the glyphic cauac signs
(T528) with pluvial associations such as rain
clouds. In this context the m otif common at Teotihuacn called the trilobal drop (Kubler 1967, p.
15, no. 51) comes to mind. It was in use also at
M onte Alban, Tajn, and Xochicalco. If the triden
tate knife is a form cognate to, or descended from,
its Mexican ancestor, its meaning to the Maya as a
ritual im plem ent for controlling rain might be
considered.
As in other instances, such as the doubled-portrait lintels of Tikal (Kubler 1976a), survivals
among the lowland Maya of costumes and instruments related to Teotihuacan iconography appear
long after the agreed end o f the history o f that site
at about a . d . 650. These survivals signify the acceptance o f Mexican symbols among the ruling
class on a basis of equality, as though Mexican and
Maya dynastic lines had mingled so thoroughly as
to assure the continuity of highland lineages and
highland ritual customs long after the assumed
collapse o f Teotihuacan.

Colophon
M esoamerican studies often impress readers as a
Sahara o f guesses, w here travelers crazed with a
thirst for certainty suffer various mirages. Among
these mirages there is an od Egyptianizing illusion, where all life is reconstructed as being des-

350

A ncient Am erica

tined for the tomb. A nother is Gothicizing, as

when all life is made to seem ruled by the hope of
salvation, the fear of judgment, and the pains of
Hell. Then again, all life is made to appear crushed
beneath the terrible gods o f antiquity in an illusion that human sacrifices were perform ed everywhere and always, in Aztec numbers and with
Aztec cruelty. And there is also the Bergsonian
illusion (which is this w riters) that a vital impulse
informs everything human, to transform even
death and the tomb into mirrors o f life.
Mirages can be not only compelling delusions,
but they may also prevent visin. O ne example of
induced blindness is the long delusion that there
were no women rulers portrayed in ancient Maya
art; another is the long resistance to the idea of
historical meaning in the inscriptions due to the
theory of Maya number-worship; another is the
recent belief that millennial change and multiplicity were absent from M esoamerican history.
Yet these delusions and illusions are the raw
stuff of humanistic study: the humanist today is
primarily concerned with the history o f error, with
the shadows and failures that surround even the
most perm anent expressions of the thought o f the
past. The charting o f these submerged reefs and
wrecks is an unavoidable necessity because the
history o f error may be the history o f much that we
now accept as true.
For the purposes of Mesoamericanists, the situation is quite different. Surrounded by error past
and present, they m ust ignore it rather than accept
it as part o f their field o f study. T he need may be to
sort out the many mirages that fill all visin from
those limited but useful working models like the
historical hypothesis in Maya epigraphy that
may some day be reduced, when accepted, to unassailable lists o f dynasties and reigns.
The present reconsideration o f a long-neglected

and unique historical vase-painting will be justified when it finds a line or two in that emerging
manual o f M esoamerican history.
NOTES
1 Onyx marble is a dense, crystalline form o f lime carbonate
(Thrust et al. 1968, p. 767) usually deposited from coldwater solutions in caves, w here it forms in compact, cryptocrystalline, banded deposits. It is variously called cave onyx
(Gary, M cAfee, and W olf 1972, p. 112), Mexican onyx,
tecali, and aragonite, but calcite (C aC O ,) is the general
term for this translucent stone which, in Mesoamerica, was
regarded as a precious material.
Maya examples are rarer than those from Veracruz or the
Mexican highlands. A group o f six vessels was excavated at
Uxmal (Ruz 1955b, pp. 6064). O ne o f them is 25.5 centim eters high and bears a carved panel portraying a ruler
receivinggifts from a visitor. Both figures are identified by
separate captions of six and four glyphs each. Like the Dum
barton Oaks bowl, the vase from Uxmal is o f Late Classic
style, both as to the figural poses and the glyphic inscrip
tions, but the D um barton Oaks bowl and its companion at
D um barton Oaks (D um barton Oaks Collections 1963, p.
12, no. 50) differ from the example at Uxmal by their vertically ribbed and scalloped carving, which resembles that
of an onyx marble vessel from the Isla de Sacrificios in
Veracruz (Kelem en 1943, vol. II, pl. 257c).
2 Benson and Coe (D um barton Oaks Collections 1963, p.
12) corrected the identification o f the figure at A as a
robed w oman on the evidence o f women in Maya art
assembled by Proskouriakoff (1961).
3 In a letter (N ovem ber 9, 1975) to the author, Pros
kouriakoff questioned T hom psons reading o f Ceh as the
bright-fire glyph, T44:563b, a t c 'l on Quirigua Zoomorph G. She prefers aplacem ent at (9 .15.0.0.0) 4 Ahau (13
Yax), as on Copan Stela A at f 1 1. Riese (1971, pp. 233-35)
has offered proofs for the use ofT 563 as meaning, in eleven
certain instances, count forward to nearest PE [Period
Ending], either hotun or m ltiple th e r e o f (authors
translation).
4 R uppert and D enison (1943, p. 120) describe the ruler as
clasping horizoncally across his breast the double-headed
ceremonial bar, but the heads facing one another have no
parallel am ong Cerem onial Bars.

III. 17
Renascence and Disjunction in the
Art of Mesoamerican Antiquity

I
A ttention has seldom been drawn to the notion
that the literary habits o f Clio resem ble the weaving o f Penelope. H istorians often write o f the web
of happening, the tapestry o f history, as though it
were a creation o f patient persistence only. The
other face of happening, however, which is all disruption and broken threads, gets little attention. It
is much m ore difficult to describe change than to
report continuity. This may be why historians prefer to describe change as a continuity disturbed,
rather than as change and disruption per se. For this
reason perhaps, the writing of history has often
been lacking in sharply contrasted opposites.
Let us take, for example, the Renaissance. We
speak of The Renaissance with ease and familiarity as though it were an operational reality. But
one o f the tests o f reality is the presence o f an
opposite: night and day, wet and dry, hot and coid.
The idea o f the Renaissance suggests a field of
forces and, though it is seldom noted by historians,
the presence o f an opposite pole. H ow are we to
apprehend these counterforces in history?
We can begin by laying out a scale of the magnitude of periods within the historical field. The Re
naissance, as everybody understands it, is the
largest species in the m useum o f history. In magnitude it is like brontosaurus by far the greatest
example o f its kind but structurally it is similar
to much smaller and m ore recent species.
The next smaller historical instance has been
studied mainly by medievalists. Less ampie than
Renaissance, the ame it usually bears is renas
cence. Erwin Panofsky presented those many m e

dieval episodes in which a renascence o f some part

o f classical antiquity was attem pted, in a volume of
1960 entitled Renaissance and Renascences. 1 The
idea o f renascence as a fragmentary restoration of
some portion o f classical antiquity to actual use
mcludes the special case o f the Carolingian renovatio. Panofsky and his colleague Richard Krautheimer finally enum erated so many renascences,
from the Early Christian centuries through to the
quattrocento, that these formed an almost continuous tissue o f classical substance. From their work
there appeared a new definition of tradition. For
them, tradition could be characterized as a closemeshed sequence of efforts to restore portions of
classical antiquity to positions of authority in m e
dieval culture. In this light, renascence could be
seen as a phenom enon o f the persistence of tradi
tion. W e can now observe as well the persistence
of tradition throughout the entire scale o f magni
tudes o f historical periods.
O ur museum o f history, now a museum of continuities, contains even smaller species o f the same
genus as Renaissance and renascence. The history
of art abounds in examples of closely related persistences o f tradition called revivis o f taste.
When many different revivis coexist, as in the
nineteenth century, their intermingling is called
an eclectic style. Revival differs from renas
cence, however, as fashion differs from historie
style. G reek or Gothic o r Egyptian revivis are
episodes o f taste. As such, they transform the
scene. Doric columns, Theban pylons, and Flamboyant vaulting reappear side by side. They function m ore as quotations from the past than as
whole texts. T he Renaissance was an attem pt to
351

352

A ncient Am erica

live from the whole book o f antiquity; renascences

make use o f fragments; and revivis are selective,
obeying the rule o f taste in ransacking the whole
treasury o f history and archaeology.
O f even briefer duration than revivis are the
revolutions and cycles o f fashion in clothing. Two
different rhythms appear in fashion: the slower
one skips generations and the faster one turns
back upon its track several times in each generation. The slower rhythm reasserts the continuity
with the grandparents generation. The faster cycle is fashion itself, as charted by Richardson and
K roeber,2 and is manifested in the regular rise and
fall of skirt length, degree o f fiare, waist height,
and depth o f dcolletage. O n the slower cycle of
fashion, the present often returns to the modes of
about six decades, or two generations earlier,
thereby skipping the parents style to renew an
interest in the style o f the grandparents.
These terms and concepts all pertain to con
tinuity and the reassertion o f tradition, but there
are few correspondences on the other side o f the
ledger, w here discontinuity and rupture might be
recorded. In an experimental accounting o f such
polar terms we might have, in descending order of
magnitude:
Types of Continuity
Renaissance
renascence
renovatio
revival
fashion

Means of Discontinuity
disjunction
disjunction
disjunction
discard
discard

Discard needs little comment. B ut disjunc

tion is a term brought into the history o f art by
Panofsky in 1944,3 following the lead o f Adolf
Goldschmidt and Paul Frankl.4 Goldschm idt had
noted in 1937 the phenom enon o f the separation
of classical form from classical meaning in m edi
eval art. Panofskys principie o f disjunction
comes from his long examination o f the modes of
survival o f classical antiquity during the Middle
Ages. His analysis was iconographic, and it was
only toward the end o f his lifelong study that Pan
ofsky adopted the idea o f disjunction to explain
the medieval reuse of classical forms and meanings. In the process o f his investigation he came to

believe that a principie o f disjunction which governed these survivals existed in consistent and recognizable form.
The axioms fundamental to this m ethod, and
first stated in 1939 by Focillon,5 are, first, that
visible form often repeated may acquire different
meanings with the passage o f time and, second,
that an enduring meaning may be conveyed by
different visual forms. Panofsky extended this
perception to the systematic study o f medieval
Christian iconography in 1944 with the formulation as follows, w herever a [medieval] sculptor or
painter borrows a figure or a group from classical
poetry, mythology or history, he almost invariably
presents it in a non-classical, viz., contemporary
form .6 In 1960 Panofsky called this the princi
pie o f disjunction,7 and he amplified it to cover
not only the reclothing o f classical meanings in
medieval forms, but also the converse bestowal of
medieval meanings on classical forms. In these
terms Panofsky presented the entire fabric o f clas
sical art as disrupted during the M iddle Ages, with
classical forms torn from their meanings and reorganized as medieval art on the pattern o f classi
cal meanings expressed in medieval form and clas
sical forms yielding medieval meanings.
Panofskys works treat only of symbolic ex
pressions in classical and medieval European literature and art. W hen wider ranges o f useful objects
and ordinary Communications are considered, the
question arises w hether the disjunctive process of
combining od forms and meanings with new
meanings and forms does not vary along a gradient
between choice and necessity. Indeed, useful ob
jects and everyday expressions usually display a
greater conjunction o f form and meaning through
time than do the m ore fragile expressions of
religious symbolic systems. For instance, stability
of form in utilitarian pottery is evident to all.
Cooking ware changes less rapidly than carved and
painted pottery made for ceremonial use. The
most useful symbols also endure with little
change. For instance, the letters of the alphabet,
which are nonritualistic symbols, continu essentially unchanged for long periods because they are
in constant, universal use. Although thegrand disjunctions described by Panofsky all concern re
ligious beliefs and symbols rather than the ico-

R enascence and Disjunction in M esoamerican Art

nography o f everyday life, the principie can be
carried over to ancient and medieval technologies,
which also underw ent change in form and m ean
ing, if less rapidly than did the corresponding systems o f religious iconography. W ere we to read
the succession o f classical and medieval civilizations from cooking ware alone, the many differences between them would vanish into the continuity of ceramic technology.
The disruption between religious forms and
their meanings yields an objective measure o f the
extent to which antiquity was replaced by the
Christian and Islamic M iddle Ages. This disjunc
tion, which is a m ode of renovation, may be said, in
an even wider frame o f reference, to happen
whenever a civilization refashions its inheritance
by discovering new meanings in the forms of the
preceeding civilization and by clothing in new
forms those od meanings which remain acceptable. The successors thereby unconsciously obey a
rule of least effort: w ithout realizing it they salvage
large parts of their inherited tradition and so avoid
having to discard and then reinvent everything.
The cumulative character of the succession of
cultures in a given regin can occur only with the
selective discarding which is implicit in disjunc
tion. Panofskys invesdgations o f medieval and
Renaissance materials dem onstrated that the carry-over of a cultural tradition can be quantified by
the extent o f the disjunction between its forms
and meanings. The quantification may be coarse,
but it is undeniably a measure o f od and new
matter.
W hen observing disjunction, not only do we
pace out the boundary between eras in occidental
history, but we also face the difficult notion of
discontinuity in a temporal fabric whose weave we
know to be unbroken. Continuous form does not
predicare continuous meaning, or does continuity of form or o f meaning necessarily imply
continuity o f culture. O n the contrary, prolonged
continuities o f either form or meaning, on the
order of a thousand years, may mask a cultural
discontinuity deeper than that between classical
antiquity and the M iddle Ages. This is a particularly im portant caveat in considering civilizations
for which literary sources are unavailable, as, for
example, in the study of the older stages of the

353

native civilizations o f ancient America. T he princi

pie of disjunction, once accepted, brings into
question every ethnological analogy by insisting
upon discontinuity rather than continuity wherever long durations are under discussion, but it also
provides a serviceable explanation for many complex mechanisms o f cultural change.
The relation between renascence and disjunc
tion is o f particular interest in considering the
Mesoamerican archaeological record. Let us consider in detail, then, both the architectural profiles
of Teotihuacn and their continuation as renascent
forms in later periods, and also, through the dis
junction of form and meaning, the changing signifkance of the enduring image of the jaguar in
Mesoamerican iconography.

II
The iconography o f architecture, both here and in
Europe, has been under intense study for several
decades. The m ethods o f that study can also yield
useful results when they are applied to the archi
tecture o f ancient Mesoamerica. It is now apparent
that no building is w ithout some conventional
meaning which is conveyed by its spatial order as
well as by its ornamental themes. It is also appar
ent that such meanings can be recovered from the
spatial designs o f peoples who left no written re
cords when their societies vanished long ago.
At Teotihuacn (fig. 111-46) the ancient archi
tecture built from 300 b . c . to about a . d . 700 includes many forms which centuries later reappear
at distant places in Guatem ala and Yucatn, signifying at least some continuity o f meaning both in
time and in space. The most distinctive and dura
ble physiognomic trait o f the architecture at Teoti
huacn, known as the terrace-profile, was used
to articlate huge pyramidal platforms which
served as bases for shrines (fig. III-193). It is often
called by its Spanish ame, even in English writing,
as the talud-a.n-tablero profile. Talud means
talus in English, or, here, the receding slope at
the base o f the pyramid. The tablero, which means
apron in English, is a panel rising vertically
above the slanting talus. This panel, or tablero, is
the facing on a horizontal ledge of slate which is
cantilevered o u t over the top o f the sloping talus.

354

A ncient Am erica

Seen frontally, then, the first level o f construction

in the pyramid is the talus, which slopes away from
the viewer, and the second level is the tablero
which overhangs the talus and rises vertically
above it (fig. III-194). There are large and small
terraces, depending on the size o f the platform and
its context in space, but both the small and the
large terraces follow the same format.
The projecting tablero protects the talus from
weathering, and the design as a whole assures a
certain degree of stability by its cantilevered form.
Stonework projects from the face o f the tablero
around its perim eter to form a frame for the recessed panel (fig. III-l 19). This frame rested on
stone ledges destined sooner or later to collapse.
Early stone frames such as the Quetzalcoatl plat
form in the Citadel are wider, deeper, and thicker
than late frames and are made o f massive masonry
blocks. In contrast, the tableros of the later centuries had thin, shallow frames.
The visual effect o f this form o f construction is
remarkable. The massive, framed tablero ovcrhanging a relatively small talus casts a deep shadow
when the sun is high in the sky. This shadow causes
the massive frame o f the tablero to seem to levitate
off a supporting cushion o f darkness. The effect is
especially striking at midday when seen from the
small courtyard dwellings o f the last periods.
The main difference between public, or religious, and prvate, or secular, constructions of
ancient M esoamerican architecture was probably
one of size. Large platforms were for public use,
small ones for dwellings or household shrines.
Within the household, a difference in proportion
echoed the difference between the divinity and
the people o f the compound. The shrine rose
upon its high slanting base the entire pyramidal
platform while the surrounding megaron-Iike
chambers occupied lower pedestals over shadowed taluses.
Tablero and talus are om nipresent at T eotihua
can. As the privileged form, chosen to distinguish
the facades o f temples and their platforms, it dominates all parts of the vast city. N o other exterior
profile competes.
The tablero and its base, like the pedim ent in
M editerranean antiquity, may have connoted sacred architecture. The domain o f cult and ritual

would thus be marked off from secular building by

the notched and cantilevered profile o f the plat
form. The tablero may or may not bear sculptural
indications of the specific cult, but it is likely that
its main purpose was to set sacred edifices apart
from dwellings and other secular buildings. If this
guess is correct, then the tablero and talus are significant forms in themselves, w ithout additional
information which m ight have been supplied by
written record but which, if it had been offered,
might be merely corroborative. In this case, we
may suppose that the architectural profile is in and
of itself a m ajor indicator o f meaning, specifying
both the function o f the building and the ethnic
identity o f its builders.
The m ention o f ethnic identity brings us to ask,
how are the terrace profiles at Teotihuacan related
to those o f the rest o f ancient Mesoamerica? The
other varieties o f profiles assembled by Marquina
in 19518 differ significantly from those of Teoti
huacan (fg. III-195a). Five m ore types are easily
recognizable during the two-thousand-year era
from Late Formative until the sixteenth century
(fig. III-195). They are:
b. the dentated profile o f M onte Alban
c. the slanted and undercut profile of the
Southern Maya lowlands
d. the so-called binder profile o f the northern Maya provinces
e and f. the outsloping profile o f Tajn and Xochicalco
g. the double-pitched profile o f Aztec ar
chitecture.
Each profile marks out an architectural sphere
of influence, and each has a distinct duration. Cer
tain profiles especially the tablero and talus of
Teotihuacn or the dentation of M onte Alban
reappear far away from their origins, both as colo
nial forms, as at Kaminaljuyu, and as revival or
renascent forms, as atT u la and Chichen Itz. Each
probably has a distinct meaning in its characterization o f different architectural traditions, different
cult practices, and different ethnic identities.
The terrace moldings at M onte Alban are often
treated as though they were merely another vari-

R enascence and D isjunction in M esoamerican Art

ety o f the tablero and talus profile, but their design
and construction differ radically from those of T e
otihuacan. H ere, the talus is not at the base of the
platform, but rests upon a rectangular plinth (fgs.
III-195b and III-196). Above the talus hang sever
al short bracketlike moldings which repeat in two
or more parallel receding planes. Above these
moldings there may be a short outsloping cornice
or another plinth. The effect in the bracketlike
moldings is o f planes and outlines alternating in
highlight and deep shadow, giving to the base and
the roofline the character o f an interm ittent or
rhythmical system. These separate planes o f relief
are like a fringed fillet, or headband. The molding
above the talus is not a tablero: it has no frame, and
it endoses no panel. In large compositions its profiles are slanting. N o supporting ledges cantilever
the projecting portions, which are generally corbelled out only enough to east the desired shadows. T he dentated profile o f M onte Alban or
Mitla was repeated with modifications many cen
turies later at Chichn Itza in the profiles o f the
Toltec-Maya Chacmool Tem ple as well as at the
Castillo.
The Southern Maya lowlands repertory of profiles is m ore difficult to define and interpret. A
recognizable group from Uaxactun and Tikal
through to the end o f the Initial Series period in
the ninth century arises from a coherent system of
design. In this group the exterior profile o f the
vaulted building echoes the profile o f the platform
terrace on which it is built (fig. III-197). This profile, shared by platform and building alike, can be
called aslanted and undercut apron m olding. An
architect would cali it a chamfered bevel. The
chamfer, or recessed molding, acts as a talus, and
the strong shadow it casts separates the terraces,
while the bevel, which carches the light, emphasizes the weight o f the terrace whose shape it de
fines. The chamfer sometimes occurs in the bevel
itself.
N orthern Maya builders separated the vaulted
building from its platform by endowing it with a
profile more characteristic o f buildings than of
platforms (fig. III-198). This profile resembles the
binder with which a thatched roof is gathered or
cinched together at the eaves and at the peak. It is
an imitation in stone o f the construction with wat-

355

tle and slender saplings used in Mayan houses.

This binder molding, as it is called, appears in
Chenes, Puuc, and East Coast buildings. At the
Caracol in Chichn Itza the binder molding at impost level has five members, expressive o f the
structural requirem ents o f this annulated vault
system.
The outsloping profiles used at Tajn (fig.
III-195e) and Xochicalco (fig. III- 195f) may be
regarded as a regional variety of the binder mold
ing, rising upon a talus and sloping out above the
construction. H ere the binder is like a paneled
strip, containing niches o r geom etric frets. Its
early history is unclear; it is possible that the form
may reflect contact with Maya peoples.
The last o f the M esoamerican terrace molding
appears in Post-Classic Aztec architecture after
1300. Aztec influence throughout ancient Mesoamerica was total, and no regin entirely resisted
its terrible appeal. Its architectural symbol was
probably the stair-balustrade which is distinguished by having two slopes of different pitch
(fig. III-195g). It is a special variant o f the binder
molding; the effect resembles the constriction
when a bag o f earth is abruptly capped by a confining lid. To the climber, the suddenly increased
pitch o f the balustrades makes the stairs seem
steeper. All its victims, w hether ancient or modern, have been intimidated by such visual changes
of angle in their ascent of the painfully narrow
treads (fig. III-199).
These six m ajor M esoamerican profiles correspond to geographic and ethnic groupings, as did
the orders o f classical antiquity codified by Vitruvius, the first-century architect o f the em peror
Augustus. For Vitruvius, the Doric order was the
earliest and was associated with the Peloponnesus
in the reign of the Dorian kings of Achaea. The
Ionic order arse later when the Athenians colonized Asia Minor. T he Corinthian order was invented last by a sculptor wishing to imitate the
growth o f acanthus leaves in the ornam entation of
the capitals. Each order had definite expressive
properties: the Doric was manly; the Ionic, womanly; the Corinthian, o f slighter, more maidenly
proportions. These expressive conventions have
persisted to the present day in architectural theory. D uring the centuries after the Italian Renais-

A ncient Am erica

356

sanee, architecturai fashions continued to re

capitlate the earlier history of the orders. In the
fifteenth century the Corinthian order was the one
preferred by Italian sculptors; sixteenth-century
M annerist architects turned to Doric and the even
cruder Tuscan severity; seventeenth-century Palladianism used Ionic details, and Baroque archi
tects preferred the Corinthian. W hen the orders
were depaganized in the Renaissance, Christ and
the saints, both male and female, were assimilated
into the Vitruvian system. Even the ages o f man
were equated with Vitruvian orders in schemes
comparing the Tuscan column to od age, the D o r
ic to the prim e of m anhood, and so on.
In the past quarter-century, the students ol
American antiquity have become increasingly par
dal to terms borrowed from M editerranean archaeology. The Americanists use of the word
classic to desgnate events older than a . d . 900 is
now part o f a terminology having almost unquestioned acceptance. O ther suggestions o f the clas
sic parallel are the following: the presence of a
canon of propordons based upon numerical ratios;
the prevalence of the use o f local-tone coloring,
without perspective shading, to show shadows;
and the strong anthropom orphism of most M eso
american art. These all reinforce the idea o f the
classic character o f American antiquity before
a . d . 1000. The existence o f an expressive system,
shown in the terrace profiles we have examined,
again confirms the parallels with Greco-Romn
antiquity that have long been apparent. Such a sys
tem o f expressive architecturai forms, as we have
seen, is also necessary to the existence o f a classical
tradition perpetuated by repeated renewals or rediscoveries. M esoamerica shares this pattern with
the M editerranean world. Indeed this study itself
forms a m inute part o f the ongoing renascence of
American antiquity in the tw entieth century.

III
But tradition is m ore than self-renewal: it also contains pauses and lapses and many disappearing
structures. As an example o f the self-cleansing
character o f disjunction, let us turn to the jaguar
theme in M esoamerican art. D uring the Classic
Age, from a . d . 100 to 700 in the Valley of Mx

ico, a dom inant pattern o f ritual and ceremony

centered upon a jaguar-serpent-bird icn. This
cult was known first at Teotihuacan (fig. III-200),
and it reappeared much later at Tula and Chichn
Itz. Jaguar-serpent-bird images appear at Teoti
huacan as quadrupeds and bipeds, as helmeted
busts, as priestly headdresses, as ritual instru
ments, and as frontal icons. U ncom pounded jag
uar images are nonexistent. Every four-footed im
age o f the jaguar is in one way or another
compounded with nonjaguar parts drawn from
other life forms. The eyes are usually round and
rimmed by feathers; the broken-circle spots on the
pelt are treated as flowers or seashells; the tongue
is a bifid serpents tongue, and the entire body is
often covered with a design resembling a fishnet.
For example, the most easily recognizable jaguar
from Teotihuacan is the thirteen-inch-long onyx
(tecali) figure in the British Museum shown lying
belly-down with its paws extended (fig. 111-92).
On each front paw is a glyphlike cipher which resembles the serpents m outh and is associated with
rain. The eyes are the round eyes o f a bird, and the
legs are shown fringed with serrated forms like the
paw-wing o f avian derivation in Olmec art before
the fifth century b . c .
Similar instances of com pounded jaguar forms
abound in the muris. In the mural of the m ytho
logical animals at Teotihuacan (assigned by Clara
Milln to an early Stage Two in the history o f wall
painting at the site)9 various four-footed jaguars
appear among the waves. O ne has a winged figure
halfway down its throat (fig. III-201). O ne squarejawed feline, swimming with an overhand stroke
and spitdng vigorously, wears a floral pelt like that
of the Tetitla cat. A nother lacks some claws, hav
ing perhaps been flayed (fig. III-202). These early
jaguars all have round birds eyes.
By far the largest class o f jaguar images consists
of human beings wearing jaguar costumes. Sometimes the costum e is an entire pelt, with head and
claws and tail, but m ore often it is only a jaguar
headdress. W hether in full garm ent or headdress
alone, the jaguar traits are always compounded
with others drawn from bird and serpent images.
The kneeling jaguar-man from the wall paintings at Tetitla (fig. 111-85) wears an overall netted
costume with the meshes extending to the muzzle

R enascence and D isjunction in M esoam erican Art

and ears o f the jaguar suit. The w earers human
identity is revealed by the shield and staff he holds
in his hands. H e kneels upon a pathway leading to
a temple decorated with floral spots like those of a
pregnant she-jaguar. These spots surround the
doorway and fill the crenellations below a band of
netting similar to the netting worn by the jaguarman. They probably signify the dedication o f the
temple to a spirit or forc represented by these
markings. Elsewhere the netted figure is associated only with the jaguar-serpent-bird figure in
muris and pottery decoration.
If we suppose that frontal figures are more likely to represent objects o f worship than were the
profile figures serving or accompanying the front
al figure, it is plausible to maintain that large and
isolated frontal figures are cult objects or icons.
Cylindrical vessels on tripods, of Teotihuacan
III date, show crouching jaguar-serpent-birds in
frontal aspect, both alone and attended by priests
(fig. III-203). In one liturgical scene, the intended
representation o f the priests headdress is un
known, but the icn he approaches is the familiar
figure with birds eye, serpents tongue, and jaguars m outh in double profile. This figure appears
frontally with a large pecten shell on its chest and
serpent scales on its limbs. In the background are
conch shells, feathered eyes, and flames. The offering borne by the priest resembles a bird with
beak and round eye, reminding us o f the mural of
the mythological animals at Teotihuacan, where a
jaguar is seen swallowing a bird (fig. III-201), suggesting that birds were offerings welcome to the
jaguar-serpent-bird.
This scheme reappears w ithout the offerings in
a num ber o f molded pottery fragments. The feath
ered birds eyes surm ount a jaguars mouth, and
the bifid serpents tongue hangs between the legs
and claws of the crouching jaguar. This them e ap
pears again in abbreviated form at Tula and at
Chichn Itza, and we shall examine these forms
later.
A human form, shown only to the waist, is frequently represented in muris, pottery designs,
and clay figurines as a pyramidal bust, crowned by
the jaguar-serpent-bird helmet. In a mural found
at Zacuala, this figure carries a shield on his left
arm and, in his right hand, a feathered jaguar-mask

357

held like a vessel or censer. The helm et is in the

iconic form o f a jaguars m outh seen in doubleprofile, and it has feathered birds eyes and netted
panels. The censerlike mask repeats these forms in
single profile and is surm ounted by a netted jag
uars paw and a bifid serpents tongue. Among the
feathers on the mask are drops o f water and cuspated lines like those of the waves in the fresco of
the mythological animals at Teotihuacan (fig.
III-201). A helm eted b u st10 is also of this type, but
the eyes behind the goggles are closed, suggesting
death, and the plumage seems to be that o f a butterfly, often associated with burials.
The significance of the helmeted-bust form is
suggested by a double figurine o f baked clay in
the Diego Rivera M useum in Mxico City (fig.
III-204). A standing human with the head of a
jaguar carries just such a helm eted bust in his outstretched arms. The helm et is o f the Oaxacan
type, with the parted upper lip of a jaguar. The
scenic and narrative context o f the jaguar-headed
man holding the helm eted bust allows a provi
sional interpretation of the figure as a heraldic
bearer. The helm eted bust may have been
intended as the portrait likeness o f a dynastic personage, whose clan or family membership it
signified.
Cats and dogs do not usually go about together,
but at A tetelco jaguars and coyotes appear in
peaceful procession on a panel inside a border
where a coyotes body is intertwined with netted
jaguars, suggesting the interchangeable and complementary character of the two forms (fig.
111-79). The netted jaguars have feathered birds
eyes and bifid serpents tongues. U nder the
mouths of both creatures appear trilobed watersigns like those common in Oaxaca. O n the righthand border, the water-sign is augmented by an
eye signifying the brightness o f running water.
The water them e reappears in the upper border,
where goggled rain-faces appear among the twining parts of the bodies of netted jaguars and
coyotes. The scene invites speculation upon the
unin of cat and dog and the relation o f this unin
to water, running water, and rainwater.
Although neither jaguar or coyote was un
known in the Valley o f Mxico, the coyote was
more common in the dry northern plateaus and

358

A ncient Am erica

the jaguar in the humid lowlands o f Veracruz and

Tabasco. It is possible that their encounter at T eo
tihuacan may have signified some resolution of
opposites in the cult o f Teotihuacan, such as, for
example, the unifying o f unlike peoples in a com
mon ritual.
The association o f intertw ining coyote and jag
uar is both com plementary and reciprocating;
each seems to supply what the other lacks, and
their apparent affection for one another is m utu
al. It is also a unique association, for the jaguarserpent-bird is found with no other land animal at
Teotihuacan except man, as in the jaguar holding
a human bust (fig. III-204). Furtherm ore, there is
no example at Teotihuacan in any m dium of
the association between jaguar and eagle that became common at Tula and o f central im portance
in the warrior cult of Aztec religin at Tenochtitlan. These various instances o f the associa
tion of jaguar and eagle in Aztec iconography all
have in common the idea of darkness as expressed by night, caves, eclipse, and any other
disappearance of the sun. N o such expression of
darkness is manifest in the repertory o f representations o f the jaguar at Teotihuacan. Further
more, warriors at Teotihuacan identified themselves with a weapon-bearing owl rather than
with a jaguar. The com pounding of properties
suggested by jaguar, serpent, and bird elements,
as emblems o f earth, water, and air, respective!}',
points to transcendent powers o f a metaphysical
nature rather than to a cult o f war.
For these reasons, it is doubtful that either the
eagle or the jaguar images at Teotihuacan carried
the meaning assigned to them by the later Aztecs.11 By the same token, doubts arise about the
validity of any interpretation o f the religious significance of jaguars throughout the entire T eoti
huacan em pire which are colored by the beliefs of
the Aztecs of six centuries later when a disjunc
tion, occurring between new and od mythologies,
had altered the jaguars form and thereby his
meaning.
Let us now contrast the felines o f seventh-century
Teotihuacan with those of the thirteenth-century
Toltec and fifteenth-century Aztec peoples. Both
these later peoples brought about a new era of

political expansin, using od symbolic forms for

the worship o f the new gods brought into the Valley o f Mxico by wandering tribes from the North,
who came as hunters and nomads after the collapse
of the polity and the faith at Teotihuacan.
W hen the new Post-Classic peoples began to
use the jaguar-serpent-bird form, it was already
two thousand years older than they, and it had
changed meaning from a miscegenated Olmec
were-jaguar before 400 b . c . to a transcendent
spirit com pounded o f various animal powers. The
new folk in due time used the variants as they saw
fit and transformed the jaguar and the eagle into a
Symbol o f warfare conveyed by the complemen
tary images of these creatures. The earlier compounds o f the jaguar were converted by the pro
cess o f disjunction to other purposes. For ex
ample, the od jaguar-serpent-bird acquired a new
Toltec meaning as a Symbol o f the underworld by
being placed in a new context at Chichn Itza and
Tula (fig. III-205).
At Chichn Itza the jaguar-serpent-bird appears
on the bases of square piers (fig. III-205a), as
Tozzer has noted, almost 500 times in connection with the top figure o f the sun disk, the atlantean bacab, or a mask. 12 Tozzer believed that the
jaguar-serpent-bird referred to the underworld
when conjoined with the sun disk or with a skybearer. At Tula the same figure (fig. III-205b) is
shown between pairs o f eagles and vultures, and
on column bases in what was thought o f as the
underworld position beneath the feet o f a warrior
or priest.
At the same time, however, jaguars and eagles
devouring hearts became the emblems o f the new
warrior societies at Chichn Itza (fig. III-206) and
at Tula with the result that jaguars acquired new
meanings in two ways, as underworld figures when
represented as jaguar-serpent-birds and as em
blems for warriors when shown as seated or
walking animals. N either o f these meanings is evident from jaguar contexts in the usage of Teoti
huacan, or in the usage o f Pre-Classic Olmec or
Oaxacan peoples. It seems certain, on the evi
dence of their representations, that jaguars meant
different things before and after the end of
Teotihuacan.
With the em ergence of the Aztecs, the symbol

Renascence and D isjunction in M esoamerican Art

of the jaguar-serpent-bird waned and vanished,
perhaps because it was replaced by the eagle-andjaguar warrior cult and because the cult of
Tezcatlipoca as a jaguar was disjoined from the
cult of Quetzalcoatl as a feathered serpent. This
disjunction probably had the effect, among other
things, o f factoring out jaguar traits for separate
treatment.
It seems clear, then, that these newer meanings
for jaguar and eagle were no older than the Toltec
era, which had begun in the Mexican highlands at
Tula, after the fall o f Teotihuacn in about a . d .
700. The people o f Teotihuacn never connected
jaguar and eagle in the fashion o f their successors.
Their respects to the jaguar were addressed to the
jaguar-serpent-bird, first in an iconic, later in a
dynastic aspect, whose representation ceased after
the fall of Tula and before the rise o f Tenochtitlan.
A still unrealized humanistic task for archaeologists everywhere is to distinguish among funda
mental historie forms in the produets of excavation. Such historie forms are the renascent and
disjunctive classes o f objeets we have been consid
ering. T he renascent expressions are repetitions of
a past tradition made in order to assure its perpetuation. The disjunctive expressions, on the
contrary, infuse od forms with new meanings and
clothe od meanings in new forms. Artists and artisans at all times face this choice about the forms
of the past: either the past is viable, deserving to be
continued, or it is irrelevant and condem ned to
discard, at least for a time. O ften the choice im
poses a separation or disjunction between form
and meaning. O ne is renewed, and the other is
replaced.
For the archaeologist to distinguish renascent
from disjunctive classes among his artifaets may
require the training of a humanist. Since the appearance o f Panofskys book in 1960 it has become much clearer that archaeology, in order to

359

discover meaning, needs once again to embrace

humanistic learning.
The relation between Americanist studies and
art historical scholarship is a two-way circuit. Art
historical studies can assist the Americanists in
their efforts to discover an archaeological history.
In the other direction, conclusions drawn from
American antiquity can perhaps lead to areevaluation o f ideas about the history o f art: as by confirming the generality o f renascence and disjunction as
processes o f making and undoing tradition.
NOTES
1 Erwin Panofsky, Renaissance and Renascences in Western Art
(Stockholm: Almquist and Wicksell, 1960).
2 Jane Richardson and Alfred L. K roeber, T hree Centuries
of W om ens D ress Fashions: A Q uantitative Analysis,
Antbropological Records. V, no. 2 (1940).
3 Erwin Panofsky, Renaissance and Renascences, Kenyon
Review, VI (1944), pp. 20136.
4 See A dolf Goldschm idt, D ie Bedeutung der Formenspaltung in der K unstentw icklung, in Independence, Convergence, and Borr&wing in Institutions, Thought and Art
(Cambridge: H arvard University Press, 1937); and Paul
Frankl, T he Crazy Vaults o f Lincoln C athedral, A rt Bulletin, X X X V (1953), p. 105.
5 H enri Focillon, La Vie des formes (Pars: Alean, 1939).
6 Panofsky, Renaissance and Renascences, p. 220.
7 Panofsky, Renaissance and Renascences, p. 8.
8 Ignacio M arquina, Arquitectura prebisptiica (Mxico: In
stituto Nacional de Antropologa e Historia, 1951).
9 Clara H all Milln, A Chronological Study of the Mural
Art o f Teotihuacn (Dissertation: University of Califor
nia, Berkeley, 1962).
10 For illustration, see Miguel Covarrubias, indian A rt of
Mxico and Central America (N ew York: Alfred A. Knopf,
1957), p. 129.
11 These latter-day beliefs o f the Aztecs were summarized
on textual and archaeological evidence by Eduard Seler in
D ie Teotihuacn-K ultur des H ochlands von Mxico,
Gesammelte Abhandlungen zur Amerikanischen Sprach- und
Altertumskunde (Berlin, 1915), V, 484506.
12 Alfred M. Tozzer, Cbichn Itz and its Cenote of Sacrifice,
Peabody M useum o f American Archaeology and Ethnology, H arvard University, M emoirs, X I- X II (Cam
bridge, Mass., 1957).

Bibliography for Part III

[Bibliographic citations for publications on Precolumbian are and archaeology frequently vary in format.
Some works published as parts o f series are listed by
authortitle, while others are given by series title. Furthermore, serial publications often are like periodicals
with thematically diverse articles. Our citations favor
functionality over bibliographic uniformity. B ook titles
are italicized when a work is frequently catalogued under author-title, and series titles are italicized when
catalogued as parts o f larger wholes. A serial publicanon is listed like a periodical when it functions in that
way. The catalogues o f the Peabody M useum at Har
vard, the M iddle American Research Institute at Tulane, and the Benson Latin American C ollecdon at the
University o f Texas were consulted for cataloguing
procedures.}
Acosta, Jorge R. 1944. La tercera temporada de ex
ploraciones arqueolgicas en Tula, Hidalgo, 1942.
Revista M exicana de Estudios Antropolgicos, VI, no. 3.
______ 1945. La cuarta y quinta temporadas de ex
ploraciones arqueolgicas en Tula, Hidalgo, 1943
4 4 . Revista M exicana de Estudios Antropolgicos, V il,
2 3 -2 4 .
______ 1964. E l Palacio de Quetzalpapalotl. Memorias
del Instituto Nacional de Antropologa e Historia,
no. 10. M xico.
______ 1965. Preclassic and Classic Architecture o f
Oaxaca. In Handbook o f M iddle American Indians, ed.
Robert W auchope, III, pt. 2, 8 1 4 36. Austin.
Acosta, Jos de. 1940. H istoria natural y moral de las
Indias [1 5 9 0 ]. M xico.
Adams, Henry P. 1935. The Life a n d W ritings o f G iam battista Vico. London.
Adams, Richard E. W. 1971. The Ceramics o f A lta r de
Sacrificios. Peabody M useum o f Archaeology and
Ethnology, Harvard University, Papers, LXIII, no.
1. Cambridge, Mass.
Agassiz, Louis. 1880. Professor Agassizs Narrative.
Sm ithsonian In stitu tio n Miscellaneous Collections,
XVIII, 3 9 4 - 9 9 . [W ashington]

Agrinier, Pierre. 1960. The Carved H um an Femurs from

Tomb 1, C hiapa de Corzo, Chiapas, Mxico. Papers o f
the N e w World Archaeological Foundation, no. 6.
Orinda, Calif.
Ahlfeld, Fr., and R. N . Wagner. 1931. ber die
H erkunft der im Bereich altperuanischer Kulturen
gefundenen Schmuckstcke aus Sodalith. Zeitschrift f r Ethnologie, XLI1I, 2 8 8 96. [Berlin]
Alcocer, Ignacio. 1935. Apuntes sobre la antigua MxicoTenochtitln. Instituto Panamericano de Geografa e
Historia, M xico, Publicacin 14. Tacubaya.
Andrews, E. W yllys IV. 1942. Yucatn: architecture.
Carnegie Institution o f W ashington, Year Book, XLI,
2 5 7 -6 3 .
--------- 1965a. Archaeology and Prehistory in the
Northern Maya Lowlands. In Handbook o f M iddle
American Indians, ed. R. W auchope, II, 2 8 8 330.
Austin.
______ 1965b. Progress Report on the 19601964
Field Seasons . . . Dzibilchaltn Pro ject. \n Archae
ological Investigations on the Yucatn Pennsula, pp.
2 327. M iddle American Research Institute, Tulane
University, Publication 31. N ew Orleans.
______ 1968. Dzibilchaltn: A N orthern Maya M e
trpolis. Archaeology, X X I, no. 1, 3647.
Anonym ous. 1885. Relacin de la ciudad de Sant Mi
guel de Piura. In Relaciones geogrficas de Indias, II.
Madrid.
--------- 1933 . A r t des Incas; catalogue de lexposition de la
collection J . L. au Palais du Trocadro (JuinOctobre,
1933). Paris.
--------- 194043. A utos que se comenzaron por man
dado del Excmo. Seor Duque de la Palata, Virrey de
estos reynos, para trasladar la villa de Pisco a un paraxe ms seguro y alexado de la mar Ao de 1688.
Revista del Archivo N acional del Per, X III, 9 44,
1 2 5 -7 3 ; X IV , 3 - 3 3 , 1 2 7 -6 4 ; X V , 2 5 - 2 6 , 1 5 5 -8 7 ;
and X V I, 2 7 - 5 7 , 1 4 3 -7 4 .
--------- n.d. Ausstellung altamerikanischer Kunst, Berlin,
1 9 3 1 -1 9 3 2 . N .p.
Armillas, Pedro. 1944. Exploraciones recientes en Te-

361

362

A ncient Am erica

otihuacn, M xico. Cuadernos Americanos, III, no. 4,

1 2 1 -3 6 . [M xico]
______ 1945. Los dioses de Teotihuacn. Anales del
Instituto de Etnografa Americana, U niversidad N a
cional del Cuyo, Mendoza, Argentina, VI, 3561.
Arnauld, Antoine. 177581. Oeuvres philosophiq/tes.
Paris and Lausanne.
Asia and N orth America. 1953. A sia a n d N orth Amer
ica: Transpacific Contacts, comp. Marian W. Smith.
Society for American Archaeology, M emoirs, no. 9.
Salt Lake City, Utah.
Baessler, Arthur. 190203. A ncient Peruvian A rt. 4
vols. Berlin and N ew York.
______ 1906. Altperuanische Metallgerte. Berlin.
Ball, Joseph, and E. W yllys Andrews V. 1975. The
Polychrome Pottery o f Dzibilchaltun. In Arcbaeological Investigations on the Yucatan Pennsula, pp.
22747. M iddle American Research Institute, Tulane University, Publication 31. N ew Orleans.
Bandmann, Gnter. 1951. M ittelalterliche A rchitektur
ais Bedeutungstrger. Berlin.
Barthel, Thomas S. 1963. D ie Stele 31 von Tikal.
Tribus, X II, 159214. [Linden-M useum fr Vlkerkunde, Stuttgart]
______ 1965. Gedanken zu einer bemalten Schale aus
Uaxactun. Baessler-Archiv, N .F ., X III, H eft 1, 131
170. [Verlag von Dietrich Reimer, Berlin]
______ 1966. M esoamerikanische Fledermausdmonen . Tribus, X V , 1 0 1 -2 4 .
______ 1968. H istorisches in den klassischen MayaInschriften . Zeitschrift frE thnologie, XCIII, H eft 1
und
2,
11956.
[Verlag Albert Limbach,
Braunschweig]
Batres, Leopoldo. 1886. N ouvelles fouilles T eoti
huacan. Revue d'ethnographie, V.
______ 1906. Teotihuacn o la ciudad sagrada de los
Toltecas. M xico.
______ 1908. Les fouilles operes Totihuacan. In
XV Congrs international des Amricanistes, Quebec
1906, Acta, II, 2 7 7 - 8 2 . Q uebec.
Bennett, W endell C. 1946. T he Archaeology o f the
Central A ndes. Bulletin o f the Burean o f American
Ethnology, II, no. 143, 6 1 147.
Berger, R., J. A. Graham, and R. F. H eizer. 1967. A
Reconsideration o f the La Venta Site. Contributions
o f the University o f California Archaeological Research
Eacility, III, 124. [B erkeley]
Berlin, Heinrich. 1951. El T em plo de las Inscrip
ciones V I de Tikal. Antropologa e H istoria de
Guatemala, III, no. 1, 3354.
______ 1951. La Inscripcin del T em plo de las Leyes
en Palenque. Anales de la Sociedad de Geografa e H istoria de Guatemala, X X V , 12029-

______ 1958. El glifo em blem a en las inscripciones

mayas. Journal de la Socit des Amricanistes, n.s.,
XLVII, 11119. [M use de lH om m e, Paris]
______ 1959. G lifos nominales en el sarcfago de Pal
enque. Humanidades, II, no. 10, 18. [Guatemala]
______ 1963. The Palenque Triad."Journalde la Socit
des Amricanistes, n.s., LII, 9199.
______ 1965. The Inscription o f the Tem ple o f the
Cross at Palenque. American A n tiq u ity, X X X , no.
3, 3 3 0 -4 2 .
______ 1968a. The Tablet o f the 96 Glyphs at Palen
que, Chiapas, M xico. M iddle American Research Institute, Tulane University, Publication 26, pp. 135
49. N ew Orleans.
______ 1968b. Estudios epigrficos II. Antropologa e
Historia de Guatemala, X X , 1324.
Bernal, Ignacio. 1958. Mxico: Pinturas prehispnicas.
Coleccin U N E SC O de Arte Mundial. N .p.
______ 1971. T he Olm ec R egin Oaxaca. In Observations on the Emergence ofC ivilization in Mesoamerica,
Wenner-Gren Foundation fo r Anthropological Research
Symposium, Burg Wartenstein, 1970, ed. R. F. Heizer
and J. A. Graham, pp. 2 9 50. Contributions o f the
University o f California Archaeological Research Facility, XI. Berkeley.
Beuchat, H. 1912. M anuel d archologie amricaine:
Amrique
prhistorique Civilisations
disparues.
Paris.
Beyer, Hermann, 1921. E l llamado calendario Azteca:
Mxico.
______ 1922. R elaciones entre la civilizacin teotihuacana y la azteca. In Manuel Gamio, La Poblacin del
Valle de Teotihuacn, I, 2 7 3 93. M xico.
______ 1933. Emendations o fth e Serpent Num bers
o f the D resden Maya C odex. Anthropos, XXVIII,
17. [Vienna]
______ 1939. Elucidation o f a Secondary Series on
Lintel 2 at Piedras Negras. E l Mxico Antiguo, IV,
nos. 9 12, 2 8 9 92. [Sociedad Alemana de Mexicanistas, M xico]
Bird, Junius B. 1943. Excavations in N orthern Chile.
American M useum o f N a tu ra l History, Anthropological
Papers, X X X V III, 1 7 1 -3 1 6 . [N e w York]
Blom, Frans. 1924. Report on Uaxactun. MS. Carnegie Institution o f Washington. Washington. [See
also Report on the Preliminary Work at Uaxactun.1
Carnegie Institution o f W ashington, YearBook, XXIII
(1924), 2 1 7 - 1 9 e d .]
______ 1932. The Maya Ball-game Pok-ta-pok (called
T lachtli by the A ztecs). In M iddle American Research
Series, Tulane University, Publication 4, no. 13, pp.
4 8 7 530. N ew Orleans.
Blom, Frans, and O. LaFarge. 1928. Tribes an d Temples.

Bibliography
2 vols. Middle American Research Series, Tulane
University, Publication 1. N ew Orleans.
Boln, Heinrich. 1874. Peruanischen G uanogtzen.
Zeitscbrift f r Ethnologie, VI, 9 3 94, [Berlin}
Bollaert, William. 1860. A n tiq u a ria n , Ethnological, a nd
Other Researches in N ew Granada, Ecuador, Per a nd
C hili, w ith Observations on the Pre-lncarial a n d other
Monuments o f Peruvian Nations. London.
Brggemann, J., and M .-A. Hers. 1970. Explora
ciones arqueolgicas en San Lorenzo Tenochtitln.
Boletn del Instituto Nacional de Antropologa e H isto
ria, X X X IX , 1 8 -2 3 . [M xico]
Bry, T. de. 1590. Historiae Americae Scriptores. Frankfurt.
Calancha, A ntonio de la. 1638. Coronica moralizada del
orden de San A u g u stin en el Per. Barcelona.
Caas, Francisco. 1854. Exploracin de las islas de C h in
cha con tres planos. Lima.
Carneiro, R. L. 1962. Scale Analysis as an Instrument
for the Study o f Cultural Evolution. Southwestern
Journal o f Anthropology, X V III, 14969.
Carr, R. F., and J. E. Hazard. 1961. Map o f the Ruins
o f Tikal, El Petn, Guatemala. In T ik a l Reports, no.
11. Museum M onographs, T he University Museum,
University o f Pennsylvania. Philadelphia.
Caso, Alfonso. 1935. Las exploraciones en Monte Albn,
temporada 1 9 3 4 35. Instituto Panamericano de G e
ografa e Historia, Publicacin 18. M xico.
______ 1942. El paraso terrenal en Teotihuacn.
Cuadernos Americanos, I, no. 6, 12736. [M xico]
--------- 1949- El Mapa de T eozacoalco. Cuadernos
Americanos, VIII, no. 5, 14581.
______ 195859. G lifos teotihuacanos. Revista mex
icana de estudios antropolgicos, X V , 5170.
______ 1962. Calendario y escritura en X ochicalco.
Revista mexicana de estudios antropolgicos, X V III,
4 9 -7 9 .
______ 1963. U na urna con el dios mariposa. Boletn
del Instituto Nacional de Antropologa e Historia. X V I,
7 7 -9 7 . [M xico]
______ 1965. Sculpture and Mural Painting o f Oaxaca. In Handbook o f M iddle American Indians, ed. R.
W auchope and G. W illey, III, 8 4 9 70. Austin.
Chassignon de M onistrol, J. 1580. De gigantibus eorumque reliquis. . . Basel.
Cieza de Len, Pedro de. 1553. Parte primera de la
Chronica del Per. Antwerp.
______ 1864. T h e T rapis o f Pedro de Cieza de Len, a .d .
153250, contained in the First Part o f his Chronicles
of Per. . . , Trans. and ed. Clem ents R. Markham.
The Hakluyt Society. London.
Clancy, Flora. 1971. Mayan G eom etric Stones. MS.
Yale University. N e w Haven.

363

--------- 1976. Maya Pedestal Stones. New Mxico

Studies in the Fine Arts, I, 1019.
Clewlow, C. W., Jr. 1974. A Stylistic a n d Chronological
Study o f Olmec M onum ental Sculpture. Contributions
o f the University o f California Archaeological R e
search Facility, X IX . Berkeley.
Clewlow, C. W ., Jr., R. A. Cowan, J. F. O Connell, and
C. Benemann. 1967. Colossal Heads o f the Olmec
Culture. Contributions o f the University o f Califor
nia Archaeological Research Facility, IV. Berkeley.
Cobo, Bernab. 1653. Historia del N u evo M undo.
MS. Muoz C ollection in the Academy o f History.
Madrid.
______ 189093. H istoria del Nuevo M undo. 4 vols.
Seville.
C oe, Michael D. 1956. "The Funerary Tem ple Am ong
the Classic Maya. Southwestern Journal o f Anthropol
ogy, X II, no. 4, 3 8 7 94. [Albuquerque]
______ 1962. Radiocarbon Dates from Teotihuacn.
Mimeographed. Yale University. N ew Haven.
______ 1966. Preliminary R eporton the First Seasons
Work at San Lorenzo Tenochtitln, Veracruz. Mim
eographed. Yale University. N ew Haven.
______ 1967. La segunda temporada en San Lorenzo
Tenochtitln, Veracruz. Boletn del Instituto N a
cional de Antropologa e Historia, X X V III, 110.
[M xico]
______ 1968. San Lorenzo and the Olm ec Civilizadon. In Dumbarton Oaks Conference on the Olmec, ed.
E. P. Benson, pp. 4 1 78. Washington.
--------- 1970. The Archaeological Sequence at San
Lorenzo Tenochtitln, Veracruz, M xico. C ontribu
tions o f the University o f California Archaeological Re
search Facility, VIII, 2134. [Berkeley]
______ 1973. The M aya Scribe an d H is World. The
Grolier Club. N ew York.
Coe, Michael D ., and Elizabeth P. Benson. 1966. Three
M aya Relief Panels a t Dumbarton Oaks. Studies in PreColumbian Art and Archaeology, no. 2. Dumbarton
Oaks. Washington.
Coe, Michael D ., R. A. D iehl, and M. Stuiver. 1967.
Olm ec Civilization, Veracruz, Mxico: Dating o f
the San Lorenzo Phase. Science, CLV, 13991401.
Coe, William R. 1965. Tikal: Ten Years o f Study o f a
Maya Ruin in the Lowlands o f Guatemala. Expedition, VIII, no. 1, 5 -5 6 . [Philadelphia]
______ 1970. Two Incised Obsidians from Cache 161
at Tikal. MS. Philadelphia.
Coe, William R., Edwin M. Shook, and Linton Satterthwaite. 1961. T h e Carved W ooden Lintels o f Tikal.
In T ik a l Reports, no. 6. M useum Monographs, The
University M useum, University o f Pennsylvania.
Philadelphia.

364

A ncient Am erica

Coggins, Clem ency Chase. 1975. Painting and Draw

ing Styles at Tikal: An Historical and Iconographic
Reconstruction. Dissertation. Harvard University.
Cambridge, Mass.
Covarrubias, M iguel. 1934. T he Eagle, the Jaguar an d
the Serpent. N e w York.
______ 1946. Mxico South. N e w York.
______ 1957. Indian A r t o f Mxico a n d Central America.
Alfred A. Knopf. N e w York.
Delfortrie, E. 1871. Les G ites de chaux phosphate
dans le dpartem ent du Lot. Actes de la Socit Linnenne de Bordeaux, X X V III, 5 0 5 618. [Bordeaux]
D elougaz, Pinhas. 1940. T he Temple Oval at K hafajah.
Oriental Institute Publications, LIII. Chicago.
D ibble, Charles. 1942. Cdice en Cruz. M xico.
Dietschy, Hans. 194041 . M ensch und G ott bei den
mexikanischen Indianern. Anthropos, X X X V
X X X V I, 3 2 6 -4 0 . [Vienna]
Drerup, H einrich. 1959- Bildraum und Realraum in
der rmischen Architektur. Rmische M itteilungen,
LXVI, 1 4 7 -7 4 .
Drucker, Philip, Robert F. H eizer, and Robert J.
Squier. 1959. Excavations a t La Venta, Tabasco, 1955.
Smithsonian Institution, Bureau o f American Ethnology, Bulletin no. 170. Washington.
Dulignon, D esgrange. 1884. D ivers objets dethnographie pruvienne. B ulletin de la Socit d'A nthropologie de Bordeaux et du S u d Ouest, I, 15758.
[Bordeaux]
Dumbarton Oaks C ollections. 1963. Handbook o f the
Robert Woods Bliss Collection ofPre-Columbian A rt. ed.
Elizabeth P. Benson and Michael D . Coe. Dumbar
ton Oaks. W ashington.
Dtting, Dieter. 1965. Das Knoten-Graphem bei den
Maya. Z e itsc h riftf r Ethnologie, X C , 6 6 103. [Berlin]
______ 1972. H ieroglyphische M iscellanea. Z e it
schrift f r Ethnologie, X C V II, no. 2, 2 2 0 56.
Easby, Elizabeth K ennedy, and John F. Scott. 1970.
Before Corts: Sculpture o f M iddle America. The M etro
politan M useum o f Art. N ew York.
Ekholm, Gordon. 1953. A Possible Focus o f Asiatic
Influence in the Late Classic Cultures o f M esoamerica. Society fo r American Archaeology, Men/oirs,
no. 9, 7289- [Salt Lake City]
Ekholm, G ordon, and R. H eine-G eldern. 1951. Significant Parallels in the Sym bolic Arts o f Southern
Asia and M iddle Am erica. In The C ivilizations of
A ncient America: Selected Papers o f the X X I X t h Inter
national Congress o f Americanists, ed. Sol Tax, I, 2 9 9
309. Chicago.
Ekholm, Susanna M. 1969- M ound 3 0 a a n d the Early

Preclassic Ceramic Sequence o f Izapa, Chiapas, Mxico.

Provo, Utah.
Fe jos, Paul. 1944. Archaeological Explorations in the Cor
dillera Vilcamba, Southeastern Per. Viking Fund Pub
lications in Anthropology. N e w York.
Fernndez de O viedo y Valds, Gonzalo. 1547. Histo
ria natural y general de las Indias. Salamanca.
Focillon, Henri. 1934. La Vie de formes. Paris.
______ 1939. La Vie des formes. 2d enl. ed. Alean. Paris.
______ 1942. The Life o f Forms in A rt. N ew Haven.
Follett, Prescott H . F. 1932. War and W eapons o f the
Maya. In M iddle American Papers, M iddle American
Research Series, T ulane University Publication 4, pp.
373410. N ew Orleans.
Forssmann, Eric. 1956. Saule u n d Ornament. Stockholm.
Franco, Jos Luis. 1961. Representaciones de las mar
iposas en M esoamrica. E l Mxico Antiguo, IX,
1 9 5 -2 4 4 . [M xico]
Frankl, Paul. 1953. T he Crazy Vaults o f Lincoln Cathedral. A r t Bulletin, X X X V , 9 5 -1 0 7 .
Frzier, A m ede Franc;ois. 1717. Relation du voyage de la
merdu S u d a u x cotes du C hili, du Prou, et du B rsilfa it
pendant les annes 1712, 1713 an d 1714- Amsterdam.
Fuente, Beatriz de la. 1965. La Escultura de Palenque.
Estudios y fuentes del arte en M xico, Instituto de
Investigaciones Estticas, X X . M xico.
______ 1974. La cabeza colosal de Cobata. Anales de!
Instituto de Investigaciones Estticas, X II, no. 43, 4 9 56. [M xico]
Gamarra Dulanto, Luis. 1942. Apuntes sobre el guano
del Per y la antigedad del hombre en Amrica.
X X V I I International Congress o f Americanists, Lima,
1939, Proceedings, I, 12327. Lima.
Gamio, Manuel. 1922. La Poblacin del Valle de Teoti
huacn. 3 vols. M xico.
Gann, Thomas. 1900. M ounds in N orthern Hon
duras. In Sm ithsonian Institution. Bureau o f Ameri
can Ethnology, I9 th A n n u a l Report (1 8 9 7 98), pt. 2,
pp. 6 6 5 92. Washington.
Garca Carraffa, Alberto, and Arturo Garca Carraffa.
1 9 0 2 -3 6 . Enciclopedia herldica y genealgica his
panoamericana. 59 vols. Madrid.
Garcilaso de la Vega, El Inca. 1609. Primera parte de los
commentarios reales. Lisboa.
______ 186971. First Part ofthe Royal Commentaries of
the Yncas, trans. C. R. Markham. 2 vols. London.
______ 1945. Commentarios reales de los Incas. 2 vols.
B uenos Aires.
Garibay, Angel Mara. 1937. La poesa lrica Azteca.
Mxico.
Gary, Margaret, Robert M cA fee, Jr., and Carol L

Biblography
W olf, eds. 1972. Glossary o f Geology. American G eological Institute. W ashington.
Gates, W. E. 1932. Eras o f the Thirteen G ods and the
N ine G ods. T he M aya Society Quarterly, I, 7892.
Gem elli Careri, Giovanni Francesco. 1699. Giro del
Mondo. G iuseppe R oselli. N apoli.
Gladwin, Harold S. 1937. Excavations at Snaketown.
Globe.
Gobernantes del Per. 192126. Cartas y papeles: siglo
X V I. 14 vols. Madrid.
Goldschmidt, Adolf. 1937. D ie B edeutung der Formenspaltung in der Kunstentwicklung. In Independence, Convergence, a n d Borrowing in Institutions,
Thought a n d A rt. Harvard University Press. Cam
bridge, Mass.
Goldwater, R ob en . 1938. P rim itivism in Modern P a in t
ing. N e w York.
Gombrich, Ernst. 1960. A r t a n d Illusion: A Study in the
Psychology o f Pictorial Representation. N ew York.
Gonzlez de la Rosa, Manuel. 1908. Estudio de las
antigedades peruanas halladas bajo el huano. Re
vista histrica, III, 3 9 45. [Lima]
G oodenough, W. H . 1963. Som e Applications o f
Guttman Scale Analysis to Ethnography and Culture
Theory. Southwestern Journal o f Anthropology, X IX ,
2 3 5 -3 9 .
Goodman, J. T. 1897. Appendix: The Archaic Maya
Inscriptions. In A. P. Maudslay. Biologa CentraliAmericana: Archaeology, I. R. H . Porter & Dulau &
Co. London {1 8 8 9 1902], [A lso reprint ed. Milpatron Publishing Corp., distributed by Arte Primi
tivo Inc. N ew York. 1974.]
Graham, Ian. 1970. The Ruins o f La Florida, Petn,
Guatemala. In Monographs a n d Papers in M aya A r
chaeology, ed. William R. Bullard, Jr., pp. 4 2 5 - 5 5 .
Peabody M useum o f Archaeology and Ethnology,
Harvard University, Papers, LXI, Cambridge, Mass.
Graham, John A. 1971. A Maya Hieroglyph Incised
on Shell. Contributions o f the U niversity o f California
Archaeological Research Facility, X III, 15559.
[Berkeley]
Graves, T. D ., N . B. Graves, and M. J. Kobrin. 1969Historical Inferences from Guttman Scales: The
Return o f Age-A rea Magic? C urrent Anthropology,
X , 338.
Greene Robertson, Merle. 1974. The Quadripartite
Badge A Badge o f Rulership. In Primera Mesa Re
donda de Palenque, Part I: A Conference on the A rt,
Iconography, a n d D ynastic History o f Palenque; Palen
que, Chiapas, Mxico, December 1422 , 1973, ed.
Merle G reene Robertson, pp. 7 7 - 9 3 . The Robert
Louis Stevenson School. Pebble Beach, Calif.

365

Guamn Poma de Ayala, Felipe. 1936. Nueva cornica y

buen gobierno (Codex pruvien illustre). Paris.
Guttman, L. 1944. A Basis for Scaling Qualitative
Data. American Soctological Review, IX , 13950.
Hamy, E. T. 1897. Galerie amricaine du Muse d Ethnographie du Trocadro. Paris.
Hariot, Thomas. 1588. A Briefe an d True Report o f the
N ew Found L and o f Virginia. London.
Hartung, Horst. 1970. N o tes on the OaxacaTablero.
Boletn de Estudios Oaxaqueos, no. 27, 28. [Mitla,
Oaxaca]
H eine-G eldern, R. 1959- Representations o f the Asia ticT igerin . . . Chavn . . . (I, 3 2 1 26); Chnese
Influences in M xico . . .the Tajn Style (I, 195
206), C hinese Influences on the Pottery o f M x
ico . . . (I, 2 0 7 20). In X X X /// International Congress o f Americanists, San Jos, Costa Rica, 1958. Proceedings. San Jos, Costa Rica.
Helfrich, Klaus, 1973. Menschenopfer u n d Totungsrituale im K ult der M aya. Ibero-Amerikanisches Institut Preussischer Kulturbesitz, M onum enta Ameri
cana IX. Gebr. Mann Verlag. Berlin.
Hellmuth, N . 1973. Teotihuacn Influenceon the Art
o f Tikal. MS. Yale University. N ew Haven.
H entze, C. 1932. M ythes et symboles lunaires. Antwerp.
______ 1936. Rituels, croyances . . . de la C hine antique
et de lAmrique. Antwerp.
Herrera y Tordesillas, Antonio de. 1601. Descripcin de
las Indias Occidentales. Madrid.
______ 172627. H istoria general de los hechos de los
castellanos, en las islas i tierra firm e del mar ocano. 4
vols. Madrid.
H eyden, Doris. 1975. An Interpretation o f the Cave
Underneath the Pyramid o f the Sun in Teotihuacn,
M xico. American A n tiq u ity , XL, no. 2, 13147.
H olm es, W. H . 189597. Archaeological Studies among
the Ancient Cities o f Mxico. 2 vols. Chicago.
H olton, G. 1970. The R oots o f Complementarity.
Daedalus, X C IX , 636ff.
Hutchinson, G. E. 1948. Survey o f Contemporary
K now ledge o f Biochem istry, III: the Biogeochem istry o f Vertebrate Excretion. Bulletin o f the American
M useum o f N a tu ra l History [X C V I (1950) Ed .]
Hutchinson, Thom as J. 1873. Tw o Years in Per, w ith
Exploration o f its A ntiquities. 2 vols. London.
--------- 1874. Explorations amongst Ancient Burial
Grounds (Chiefly on the Sea-Coast Valleys) o f Pe
r. Journal o f the Roy a l Anthropological Institute of
Great B rita in a n d Ireland, III, 31 1 26.
Hvidtfeldt, Arild. 1958. Teotl and Ixiptatli: Some Cen
tral Conceptions in A ncient American Religin. Copenhagen.

366

A ncient Am erica

Jackson, C. A. 1971. M ultidimensional Scaling and

the Seriation o f the T w elve O lm ec Colossal H eads.
MS. University o f California. Berkeley.
Jim nez de la Espada, Marcos. 1881. Relaciones geo
grficas de Indias: Per, I. Madrid.
Jones, Christopher. 1964. Maya M iscellaneous Texts
on Minor O bjects. M. A. thesis. University o f Penn
sylvania. Philadelphia.
______ 1970. Inaugural Dates in the H ieroglyphic Inscriptions at Tikal. MS.
______ 1977. Inauguration D ates o fT h ree Late Clas
sic Rulers o f Tikal, Guatemala. American A n tiq u ity,
XLII, no. 1, 2 8 - 6 0 .
Joyce, Thomas A. 1 9 12a. A Short G uide to the American
A ntiq u ities in the B ritish M useum. Oxford.
______ 1912b. South American Archaeology. London and
Boston.
Kelem en, Pl. 1943. M edieval American A rt: A Survey in
Two Volumes. The Macmillan Company. N ew York.
Kelley, David H. 1962a. Glyphic Evidence for a D y nastic Sequence at Quirigu, Guatemala. American
A n tiq u ity, X X V II, no. 3, 3 2 3 -3 5 .
______ 1962b. A History o f the D ecipherm ent o f
Maya Script. Anthropological Linguistics, IV, 148.
[Bloom ington, Indiana}
______ 1965. T he Birth o f the G ods at Palenque.
Estudios de C ultura M aya, V, 9 3 134. [M xico}
Kidder, Alfred V. 1940. Looking Backward. Proceedings o f the American Philosophical Society, LXXXII1,
5 2 7 -3 7 .
Kidder, Alfred V ., Jesse D. Jennings, and Edwin M.
Shook. 1946. Excavations a t K am inaljuy, Guatem a
la. Carnegie Institution o f W ashington, Publication
561. W ashington.
Kirchhoff, Paul. 1950. T heL eagueofM ayapanand Its
Thirteen Calendars: An Historical-Geographical Interpretation. M imeographed. Abstract o f lecture
presented at W enner-Gren Foundation Supper Conference on N ovem ber 10.
Kosok, Paul, and M ariaReiche. 1947. The Mysterious
Markings o f N azca. N a tu ra l History, LVI, no. 5,
2 0 0 - 0 7 and 2 3 7 - 3 8 .
Kruskal,J. B. 1964. M ultidimensional Scaling by Optimizing G oodness o f Fit to a N onm etric H ypothe s is '' Psychometrika, X X IX , 11529.
Kubler, G eorge. [S ee A Bibliography o f Works by
G eorge Kubler (to 1982), below, 4 3 1 -3 6 .}
Laet, Joannes de. 1633. Novus orbis. Lige.
Lafitau, Joseph-Franqois. 1724. Moeurs des sauvages
ameriquains compares a u x moeurs des premiers temps. 2
vols. Paris.
Lake, A. K. 1937. The Origin o f the Romn H ou se.
American Journal o f Archaeology, XL1, 5 0 9 601.

Landucci, Sergio. 1972. I filosofi e i selvaggi, 1 5 8 0 1780. Bari.

Larco H oyle, Rafael. 1 9 3 8 -3 9 . Los Mochicas. 2 vols.
Lima.
______ 1941. Los Cupisniques. Lima.
______ 1944. C ultura Salinar: sntesis monogrfica.
Trujillo.
Lehmann, Karl. 1945. T he D om e o f H eaven. A rt
Bulletin, X X IX , 2 2 5 -4 8 .
Lehmann, Walter. 1933. A us den Pyramidenstadten in
Alt-M exiko. Berlin.
Leroi-Gourhan, Andr. 1965. Prhistoire de l art occiden
tal. Paris.
Lescarbot de Vervins, M. 1618. Histoire de la Nouvelle
Franee. Paris.
Linn, Sigvald. 1934. Archaeological Researches a t Teoti
huacn, Mxico. The Ethnographical M useum of
Sweden, n.s., Publication 1. Stockholm.
______ 1942. Mexican H ighland Cultures: Archaeological
Researches a t Teotihuacn, Calpulalpan, a n d Chalchicomula in 193435- The Ethnographical Museum
o f Sweden, n.s., Publication 7. Stockholm.
______ 1956. Radiocarbon Dates in Teotihuacn.
Ethnos, X X I, 1 8 0 -9 3 . [Stockholm }
Lipsius, Justus. 1637. Ivsti L ipsi . . . Opera omnia, postremvm ab ipso aveta et recensita. 4 vols. Antwerp.
Lizardi Ramos, Csar. 1947. Copn y el jeroglfico de
los sacrificios humanos. Honduras M aya, ao II, nos.
2 and 3, 4 0 - 4 3 . [Sociedad de Antropologa y Ar
queologa de Honduras, Tegucigalpa]
Lizrraga, Reginaldo de. 1946. Description de las Indias.
Lima.
Long, Richard C. E. 1923. Maya High N um bers.
M an, X X X IV , 6 6 -6 9 Lpez de Velasco, Juan. 1894. Geografa y descripcin
universal de las Indias. Madrid.
Lothrop, S. K., W. F. Foshag, and Joy Mahler. 1957.
Robert Woods Bliss Collection: Pre-Columbian Art.
Phaidon Publishers Inc. N ew York. [2d rev. ed. The
Phaidon Press. London. 1959.}
Ludendorff, H. 1935. D ie astronomische Inschrift aus
dem Tem pel des Kreuzes in Palenque. Sitzungsberichte der preussischen Akademie der Wissenschaften
(phys.-math. Kl.), X V II, 3 0 1 -3 0 .
Lundell, Cyrus L. 1934. R uins ofPolol an d other Archae
ological Discoveries in the Department o f the Peten,
Guatemala. Carnegie Institution o f Washington,
Contributions to American Archaeology, II, no. 8.
Washington.
Maler, Teobert. 190103. Researches in the CentralPortion o f the Jsumatsintla Valley. Peabody M useum of
Archaeology and Ethnology, Harvard University,
M emoirs, II, nos. 1 and 2. Cambridge, Mass.

Bibliography
______ 1908a. Explorations o f the Upper U sum atsintla
a n d Adjacent Regin: A lta r de Sacrificios; Seibal; Itsimt-Scluk; Cankuen. Peabody M useum o f Ameri
can Archaeology and Ethnology, Harvard U niver
sity, M em oirs, IV, no. 1. Cambridge, Mass.
______ 1908b. Explorations in the Department o f Petn,
Guatemala, a n d Adjacent Regin: Topoxte; Yaxh;
Benque Viejo; N aranjo. Peabody M useum o f Ameri
can Archaeology and Ethnology, Harvard U niver
sity, M emoirs, IV, no. 2. Cambridge, Mass.
______ 1971. Bauten der M aya: Aufgenommen in den
Jahren 1886 bis 1905 u n d beschrieben von. . . . Monumenta Americana, IV. Berlin.
Malraux, Andr. 194950. The Psychology o f A rt, trans.
S. Gilbert. 3 vols. N e w York.
Marcus, Joyce. 1971. N aranjo. MS. Cambridge,
Mass.
Marquina, Ignacio. 1922. Arquitectura y Escultura.
In La Poblacin del Valle de Teotihuacn, Mxico, ed.
Manuel G am io, I, pt. 1, 9 5 164. M xico.
______ 1928. Estudio arquitectnico comparativo de los
monumentos arqueolgicos de Mxico: Contribucin de
Mxico a l X X I I I Congreso de Americanistas. M xico.
______ 1951. Arquitectura prehispnica. Instituto N a
cional de Antropologa e Historia, Memorias, 1.
M xico. [2d rev. ed. 1968]
Mathews, Peter. 1976. The Identification o fT h em es
in the Sculpture o f Naranjo, Departm ent o f Petn,
Guatemala. MS. N e w Haven.
Maudslay, A. P. 18891902. Biologia Centrali-Americana: Archaeology. 1 vol. text, 4 vols. plates. R. H.
Porter and Dulau & Co. London. [Reprinted. Milpatron Publishing Corp., distributed by Arte Primitivo
Inc. N ew York. 1974.]
Miller, Arthur. 1966. Los pjaros de Q uetzalcatl. In
Teotihuacn: X I Mesa Redonda [ i 966]: Sociedad M ex
icana de Antropologa, II, 8 7 101. M xico. 1972.
Milln, Clara S. H . 1962. A Chronological Study o f
the Mural Art o f Teotihuacn. Dissertation. U ni
versity o f California. Berkeley.
______ 1966. T he H istory o f Mural Art at Teotihua
cn. M imeographed. Paper presented at XI Mesa
Redonda o f the Sociedad Mexicana de Antropologa,
August 813 , 1966. M xico. [Published in Teotihua
cn: X I Mesa Redonda (1966): Sociedad M exicana de
Antropologa, II, 1 -1 6 . M xico. 1972.]
Milln, Ren. 1960. The Beginnings ofT eotihuacn.
American A n tiq u ity , X X V I, no. 1, 110.
______ 1966a. Cronologa y periodificacin: datos estratigrficos sobre perodos cermicos y sus rela
ciones con la pintura mural. In Teotihuacn: X I Mesa
Redonda [ i 966]: Sociedad M exicana de Antropologa, I,
118. M xico.

367

______ 1966b. Extensin y poblacin de la ciudad de

Teotihuacn en sus diferentes perodos. In Teoti
huacn: X I Mesa Redonda {7966]: Sociedad Mexicana
de Antropologa, I, 5758. M xico.
______ 1970. Teotihuacn: Com pletion o f Map o f
Giant Ancient City in the Valley o f M xico. Science,
CLXX, 1 0 7 7 -8 2 .
M illn, Ren, and J. A. Bennyhoff. 1961. A Long
Architecturai Sequence at Teotihuacn. American
A n tiq u ity , X X V I, 5 1 6 -2 3 .
Miyawaki, T. 1970. M ultidimensional Scaling o f the
Olm ec H eads. MS. Yale University. N ew Haven.
M ontesinos, Fernando. 1906. Anales del Per. 2 vols.
Madrid.
M ontoto, Santiago. 1928. Nobiliario hispano-americano del siglo X V I. In Coleccin de documentos ind
itos para la historia de Hispano-Amrica, III. Madrid.
Morley, Sylvanus Griswold. 1920. The Inscriptions at
Copn. Carnegie Institution o f W ashington, Publication 219. Washington.
--------- 193738. T he Inscriptions o f Petn. 5 vols. Car
negie Institution o f Washington, Publication 437.
Washington.
______ 1946. The A ncient M aya. Stanford University
Press and Oxford University Press. Stanford and
London.
Muoz Camargo, D iego. 1947. Historia de Tlaxcala. 2d
ed. M xico.
Murphy, Robert Cushman. 1936. Oceanic Birds o f South
America. 2 vols. N ew York.
N icolini, Fausto. 1949. Commento storicoaliaseconda Scienza nuova. 2 vols. Banco di Napoli. Roma.
Paddock, John, ed. 1966. Mesoamerican Notes, VI I
VIII. {Departm ent o f Anthropology o f the U niver
sity o f the Americas. M xico.]
Palacios, Enrique Juan. 1936. Inscripcin reciente
mente descubierta en Palenque. M aya Research, III,
no. 1, 217. {N ew Grleans]
Panofsky, Erwin. 1944. Renaissance and Renas
cences. Kenyon Review, VI, 2 0 1 36.
______ 1960. Renaissance a n d Renascences in Western
A rt. Stockholm.
Parsons, Lee. 1971. M esoamerican Trade and its Role
in the Emergence o f Civilization. In Observations on
the Emergence o f C ivilization in Mesoamerica, WennerGren Foundation fo r Anthropological Research Symposium ,B urgW artenstein, 1970, ed. R. F. H eizerand
J. A. Graham, pp. 16995. Contributions o f the U ni
versity o f California Archaeological Research Facility, XI. Berkeley.
Paz Soldn, Mariano Felipe. 186263. Geografa del
Per. Paris.
Peabody M useum. 1874. Seventh A n n u a l Report of the

368

A ncient Am erica

T rus tees o f the Peabody M useum o f American Archae

ology a n d Ethnology, Presented to the President and Fellows o f H arvard College, M ay, 1874. Cambridge,
Mass.
Peafiel, A ntonio. 1900. Teotihuacn: estudio histrico y
arqueolgico. M xico.
Prez de Torres, Simn. 1749. Discurso de mi viage.
In Historiadores prim itivos de las Indias Occidentales,
ed. Andrs Gonzlez Barcia, III, 143. Madrid.
Peterson, Frederick A. 1952. Tlaloc en soportes de
vasijas teotihuacanas. Tlatoani, I, 1314. [M xico J
Pickett,J. C. 1848. LettersandD issertations uponSundry
Subjects. W ashington.
Pia Chan, Ramn. 1964. Algunas consideraciones
sobre las pinturas de Mul-Chic, Yucatn. Estudios de
cultura M aya, IV, 6 3 78.
Prem, Hanns J. 1971. Calendrics and Writing in
M esoam erica. In Observations on the Emergence of
C ivilization in Mesoamerica, Wenner-Gren Foundation
fo r Anthropological Research Symposium, B urg Wartenstein, 1970, ed. R. F. H eizer and J. A. Graham, pp.
11232. Contributions o f the U niversity o f Califor
nia Archaeological Research Facilty, X I. Berkeley.
Proskouriakoff, Tatiana. 1946. A n A lbum o f M aya A rchitecture. Carnegie Institution o f W ashington, Pub
lication 558. Washington.
______ 1950. A Study o f Classic M aya Sculpture. Car
negie Institution o f W ashington, Publication 593.
W ashington.
______ 1954. Varieties o f Classic Central Veracruz
Sculpture. Carnegie Institution o f W ashington, Pub
lication 606: Contributions to American Anthropology
a n d History, no. 58, 61 121. [W ashington]
______ 1960. H istorical Implications o f a Pattern o f
Dates at Piedras Negras, Guatemala. American A n
tiquity, X X V , no. 4, 4 5 4 - 7 5 .
______ 1961. Portraits o f W om en in Maya Art. In
Essays in Pre-Columbian A r t a n d Archaeology, ed. Sam
uel K. Lothrop et al., pp. 8 1 99- Harvard University
Press. Cambridge, Mass.
______ 1963. H istorical Data in the Inscriptions o f
Yaxchiln, Pare I Estudios de cultura M aya, III, 149
67. [M xico]
______ 1964. Historical Data in the Inscriptions o f
Yaxchiln, Part II. Estudios de cultura M aya, IV,
1 7 7 -2 0 1 .
______ 1968. The Jog and the Jaguar Signs in Maya
W riting. American A n tiq u ity , X X X III, no. 2, 2 4 7 51.
______ 1971. Early Architecture and Sculpture in
M esoamerica. In Observations on the Emergence of
C ivilization in Mesoamerica, Wenner-Gren Foundation
for Anthropological Research Symposium, Burg Warten-

stein, 1970, ed. R. F. H eizer and J. A. Graham, pp.

14155. Contributions o f the University o f Califor
nia Archaeological Research Facility, X I, Berkeley.
______ 1974. The Hand-grasping-fish and Associated
Glyphs on Classic Maya M onum ents. In Meso
american W riting Systems: A Conference at Dumbarton
Oaks, October 3 0 th a n d 31st, 1971, ed. Elizabeth P.
Benson, pp. 16578. Dumbarton Oaks Research Library and Collections. W ashington.
Ramusio, Giovanni Battista. 1606. Delle navigationi et
viaggi raccolte. 3 vols. Venice.
Rands, RobertL. 1955. Som e M anifestations o f Water
in Mesoamerican Art. Sm ithsonian Institution, Bureau o f American Ethnology, Bulletin, CLVII, 2 6 5 393. [W ashington]
Rathje, W. L. 1971. T he Origin and D evelopm ent o f
Lowland Classic Maya Civilization. American A n tiq
uity, X X X V I, no. 3, 2 7 5 -8 5 .
Reiche, Maria. 1949. Los dibujos gigantescos en el suelo de
las pampas de N azca y Palpa. Editorial Mdica Per
uano. Lima.
Richardson, Jane, and AlfredL. Kroeber. 1940. Three
Centuries o f W om ens D ress Fashions: A Quantitative Analysis. Anthropological Records, V, no. 2.
Ricketson, O. G ., Jr., and E. B. Ricketson. 1937. Uaxactn, Guatemala, Group E, 19261 9 3 1 - Carnegie
Institution o f Washington, Publication 477. Wash
ington.
Riese, Berthold. 1971. Grundlagen zu r Entzifferungder
Mayahieroglyphen, dargestellt an den Inschriften von
Copan. M useum fr V lkerkunde und Vorgeschichte, Beitrge zur mittelamerikanischen Vl
kerkunde, XI. Hamburg.
Robelo, C ecilio A. 190508. Diccionario de mitologa
nahoa. M xico.
Robertson, Donald. 1959. Mexican M anuscript Paint
ing o f the Early Colonial Period: The Metropolitan
Schools. N e w Haven.
Robicsek, Francis. 1972. Copan: Home o f the M ayan
Gods. The M useum o f the American Indian, Heye
Foundation. N ew York.
Rowe, John H . 1946. Inca Culture at the Tim e o f the
Spanish C onquest. Carnegie Institution o f Wash
ington, Bureau o f American Ethnology, Bulletin, II, no.
143, 1 8 3 -3 3 0 . [W ashington]
Roys, Lawrence. 1945. M oon Age Tables. Carnegie
Institution o f W ashington, Notes on M iddle American
Archaeology a n d Ethnology, II, no. 50, 1 5 9 -6 9 .
[W ashington]
Roys, Ralph L. 1933- The Book o f C hilam Balam of
Chumayel. Carnegie Institution o f Washington, Pub
lication 438. Washington.
Rubn de la Borbolla, D . F. 1953. Mxico: monumentos

Bibliography
histricos y arqueolgicos. Instituto Panamericano de
Geografa e Historia, Publicacin 145. M xico.
Ruppert, Karl. 1935. The Caracol o f Chichn Itz,
Yucatn, Mxico. Carnegie Institution o f W ashing
ton, Publication 454. Washington.
--------- 1943. T he Mercado, Chichn Itz, Yucatn.
Carnegie In stitu tio n o f W ashington, Publication 546:
Contributions to American Anthropology a n d History,
VIII, no. 43 , 2 2 3 - 6 0 . [W ashington]
Ruppert, Karl, and John H . D enison, Jr. 1943. Archae
ological Reconnaissance in Campeche, Q uintana Roo,
a n d Petn. Carnegie Institution o f W ashington, Pub
lication 543. W ashington.
Ruppert, Karl, J. Eric S. T hom pson, and Tatiana P oskouriakoff. 1955. Bonampak, Chiapas, Mxico. Car
negie Institution o f W ashington, Publication 602.
W ashington.
Ruz Lhuillier, Alberto. 1945. Campeche en la arqueologa
maya. Acta antropolgica, I, nos. 2 3. M xico.
______ 1952. Exploraciones en Palenque, 1951.
Anales del Instituto Nacional de Antropologa e Historia,
V, no. 33, 4 7 66. [M xico]
--------- 1955a. Exploraciones en Palenque, 1952.
Anales del Instituto Nacional de Antropologa e H isto
ria, VI, 79110. [M xico]
--------- 1955b. Uxmal: Temporada de trabajos, 1951
1952. Anales del Instituto N acional de Antropologa e
Historia, V I, no. 34, 4 9 67. [M xico]
______ 1958. Exploraciones arqueolgicas en Palen
que, 19535 6 . Anales del Instituto N acional de
Antropologa e Historia, X , 6 9 299. [M xico]
--------- 1968. Costumbres funerarias de los antiguos
mayas. M xico.
______ 1973. E l Templo de las Inscripciones, Palenque.
Instituto Nacional de Antropologa e Historia, Col
eccin Cientfica: Arqueologa VII. M xico.
______ 1973. Ms datos histricos en las inscripciones
de Palenque. Estudios de cultura M aya, IX , 9 3 117.
[M xico]
Sahagn, Fr. Bernardino de. 1938. H istoria General de
las Cosas de N ueva Espaa. 5 vols. M xico.
San G eronim o, Juan de. 1845. M em orias. In Colec
cin de documentos inditos para la historia de Espaa,
VII, 5 - 4 4 2 . Madrid.
Satterthwaite, Linton, Jr. 1935. Palace Structures J-2
and J-6 . Piedras Negras Preliminary Papers, III. Mimeographed. 5 vols. Philadelphia.
______ 1943. Piedras Negras Archaeology: Architecture,
Part 1, no. 1. University M useum, University o f
Pennsylvania. Philadelphia.
______ 1944. Piedras Negras Archaeology: Architecture,
Part IV : B a ll Courts. University M useum, University
o f Pennsylvania. Philadelphia.

369

______ 1958. Five N ew ly D iscovered Carved Monuments at Tikal and N ew Data on Four O thers. In
T ik a l Reports no. 4. M useum M onographs, The U ni
versity o f Pennsylvania. Philadelphia.
______ 1961. Maya Long C ount. E l Mxico Antiguo,
IX, 1 2 5 -3 5 . [M xico]
______ 1964. Long Count Positions o f Maya Dates in
the Dresden C odex. X X X V International Congress
o f Americanists, Mxico, 1962, Proceedings, II, 4 1 61.
M xico.
______ 1971. D ivisions o f M ulti-Date Maya Inscriptions. MS. Philadelphia.
Satterthwaite, Linton, Jr., and E. K. Ralph. 1960. N ew
Radiocarbon D ates and the Maya Correlation Probl e m American A n tiq u ity, X X V I, 16584.
Saville, Marshall. 1928. Bibliographic N otes on Pal
enque. Indian Notes a n d Monographs, VI, no. 5,
11980. [M useum o f the American Indian, N ew
York]
--------- 1929- The Aztecan G od X ipe T otee. Indian
Notes a n d Monographs, V, no. 2, 15174. [M useum
o f the American Indian, N ew York]
Schele, Linda. 1974. Observations on the Cross M otif
at Palenque. In Primera Mesa Redonda de Palenque,
Part I: A Conference on the A rt, Iconography, a n d Dynastic History o f Palenque; Palenque, Chiapas, Mxico,
December 1422, 1973, ed. M erle G reene Robertson, pp. 41 61. The Robert Louis Stevenson School.
Pebble Beach.
Schmidt, Max. 1929- K unst u n d K u ltu r von Per. Berlin.
Schroeder, Albert H . 1955. Ball Courts and Ball
Games o f M iddle America and Arizona. Archae
ology, VIII, 1 5 6 -6 1 .
Scranton, R. 1949. Group D esign in G reek Architec
ture. A r t Bulletin, X X X I, 2 4 7 -6 8 .
Sjourn, Laurette. 1956. B urning Water: Thought and
Religin in Ancient Mxico. N ew York. [2 d ed . 1960.]
--------- 1958. Los sacrificios humanos: religin o pol
tica. Cuadernos Americanos, X V II, no. 1, 12749.
[M xico]
______ 1959. U n palacio en la ciudad de los dioses
[Teotihuacn]. Instituto Nacional de Antropologa e
Historia. M xico.
--------- 1962. E l universo de Quetzalcatl. Fondo de
Cultura Econmica. M xico.
______ 1963. La cermica de Teotihuacn. Cuadernos
Americanos, X X II, no. 3, 13381.
--------- 1964. La simblica del fuego. Cuadernos
Americanos, X X III, no. 4, 1 4 9 -7 8 .
--------- 1965- El Quetzalcatl en Teotihuacn. Cuad
ernos Americanos, X X IV , no. 1, 13156.
______ 1966a. Arqueologa de Teotihuacn: la cermica.
Fondo de Cultura Econmica. Mxico.

370

A ncient Am erica

______ 1966b. E l lenguaje de las formas en Teotihuacn.

M xico.
______ 1966c. Arquitectura y p in tu ra en Teotihuacn.
M xico.
Seler, Eduard. 1893. Peruanische Altertmer, Berln.
______ 190102. Codex Fjrvry-Mayer: Eine altmexikanische Bilderhandschrift. . . . Berlin-London.
______ 190223. Gesammelte Abhandlungen zu r amerikanischenSprach- undA ltertum skunde. 5 vols. Berln.
[Reprinted. Akadem ische Druck- und Verlagsanstalt, Graz. 1961.]
______ 1904. Altm exikanischer Schmuck und soziale
und militrische Rangabzeichen. In Gesammelte A b
handlungen z u r amerikanischen Sprach- u n d Alter
tumskunde, II, 5 0 9 619- Berln.
______ 1915. D ie Teotihuacan-Kultur des Hochlands
von M xiko. In Gesammelte Abhandlungen zu r amer
ikanischen Sprach- u n d Altertum skunde, V, 4 0 5 585.
Berlin.
______ 1917. D ie Ruinen von U xm al. Abhandlungen
der konigl. preussischen Akademie der Wissenschaften
(phil.-hist. Klasse), III. [Berlin]
______ 1927. Einige K apitel aus dem Geschichtswerk des
Fray Bernardino de Sahagn aus dem Aztekischen bersetz. Stuttgart.
Shepard, Anna O. 1948. The Symmetry o f Abstract
D esign, with special R eference to Ceramic Decoration. Carnegie Institution o f W ashington: Contributions to American Anthropology a n d History, IX, 2 0 9 93.
Shook, E. 1951. T he Present State o f Research on the
Pre-Classic H orizons in Guatemala. X X I X Interna
tional Congress o f Americanists, N ew York, 1949, Proceedings, pp. 9 3 100.
Smith, A. Ledyard. 1932. T w o Recent Ceramic Finds
at Uaxactun. Carnegie Institu tio n o f Washington,
Publication 4 3 6 : Contributions to American Archae
ology, II, no. 5, 1 - 2 5 . [W ashington]
Smith, E. Baldwin. 1956. Architecturai Symbolism o f Im
perial Rome a n d the M iddle Ages. Princeton.
Smith, Robert E. 1937. A Study o f Structure A-I
C om plex at Uaxactn, Petn, Guatemala. Carnegie
Institution o f W ashington, Publication 456: Contribu
tions to American Archaeology, III, no. 19, 189231.
[W ashington]
______ 1955. Ceramic Sequence a t Uaxactun, Guatemala.
2 vols. M iddle American Research Institute, Tulane
University, Publication 20. N ew Orleans.
Spinden, H erbert J. 1913- A Study o f M aya A rt. Pea
body M useum o f American Archaeology and Eth
nology, Harvard University, Papers, VI. Cambridge,
Mass.

______ 1924. T h e Reduction o f M ayan Dates. Peabody

M useum o f American Archaeology and Ethnology,
Harvard University, Papers, VI, no. 4. Cambridge,
Mass.
Squier, E. G. 187172. Antiquities from the Huanu
Islands o f Per. Journal o f the Anthropological Institute o f N ew York, I, 4 7 56.
Stevenson, William Bennet. 182529. A Historical and
Descriptive N arrative o f Tw enty Years Residence in
South America. 2 vols. London.
Stirling, Matthew, W. 1940. Great Stone Faces o f the
Mexican Jungle. N a tio n a l Geographic Magazine,
L X X V III, no. 3, 3 0 9 -3 4 .
______ 1943. Stone M onuments o f Southern Mxico.
Smithsonian Institution, Bureau o f American Eth
nology, Bulletin 138. Washington.
Stromsvik, Gustav. 1947. G uide Book o f the R uins of
Copan. W ashington.
Strong, William Duncan. 1947. Finding the Tom b ofa
W arrior-God. N a tio n a l Geographic M agazine, XCI,
4 5 3 -8 2 .
Tello, Julio C. 1938. Arte antiguo peruano. Inca, II.
[Lima]
Thom pson, J. Eric S. 1932. The Solar Year o f the Mayas
at Q uirigu, Guatemala. Field Museum o f Natural
History, Publication 315, Anthropological Series,
X V II, no. 4. Chicago.
______ 1936. The Dates o f the Tem ple o f the Cross,
Palenque. M aya Research, III, 2 8 7 93. [N ew Or
leans]
______ 1948. A n Archaeological Reconnaissance in the
Cotzumalhuapa Regin, Escuintla, Guatemala. Car
negie Institution o f W ashington, Publication 574:
Contributions to American Anthropology and His
tory, IX , no. 44. Washington.
______ 1950. M aya Hieroglyphic W riting: A n Introduction. Carnegie Institution o f W ashington, Publica
tion 589. Washington.
______ 1952. La inscripcin jeroglfica del tablero de
El Palacio, Palenque. Anales del Instituto Nacional de
Antropologa e Historia, IV, 6 1 68. [M xico]
______ 1960. M aya Hieroglyphic W riting: A n Introduction. 2d ed. University o f Oklahoma Press. Norman.
______ 1962. A C atalogof M aya Hieroglyphs. University
o f Oklahoma Press. Norman.
______ 1965. Maya Hieroglyphic W riting. In Handbook o f M iddle American Indians, ed. R. Wauchope,
III, pt. 2, 6 3 2 59. Austin.
______ 1970. M aya History an d Religin. University of
Oklahoma Press. Norman.
______ 1972. A Commentary on the Dresden Codex: A

Bibliography
Alaya Hieroglyphic Book. Memoirs o f the American
Philosophical Society, XCIII. Philadelphia.
_____ 1973. Maya Rulers o f the Classic Period and
the D ivine Right o f Kings. In The Iconography of
Middle American Sculpture, pp. 5 2 71. The M etro
politan M useum o f Art. N e w York.
Thrush, Paul W ., et al. 1968. A Dictionary o f AI ining,
Mineral, a n d Related Terms. U.S. Bureau o f Mines.
Washington.
Tobriner, Stephen. 1972. T he Fertile Mountain: An
Investigation o f Cerro G ordos Importance to the
Town Plan and Iconography o f Teotihuacn. In Teo
tihuacn: X I Mesa Redonda [1966]: Sociedad M ex
icana de Antropologa, II, 10315. M xico.
Tolstoy, P. 1972. R eview o f Olmec: A n Early A r t Style
of Pre-Columbian Mxico by Charles R. W icke. Amer
ican Journal o f Archaeology, LX X V I, 4 5 7 59.
Torquemada, Juan. 1723. M onarcha indiana. Madrid.
Toscano, Salvador. 1944. A rte precolombino de Mxico y
Amrica Central. M xico.
Tozzer, Alfred M. 1911. A Preliminary Study o f the Prehistoric R uins o f T ika l, Guatemala. Peabody Museum
o f American Archaeology and Ethnology, Harvard
University, M em oirs, V, no. 2. Cambridge, Mass.
______1913. A Preliminary Study o f the Prehistoric
Ruins o f N akum , Guatemala. Peabody M useum o f
American Archaeology and Ethnology, Harvard
University, M emoirs, V, no. 3. Cambridge, Mass.
______ 1957. Chichn Itz a n d its Cenote o f Sacrifice, 2
vols. Peabody M useum o f American Archaeology
and Ethnology, Harvard University, M emoirs, X I
XII. Cambridge, Mass.
Tschudi, Johann Jakob von. 1847. Travels in Per, d u r
ing the Years 1 8 3 8 1842, trans. T. Ross. London.
______ 186669. Reisen durch Sudamerika. 5 vols.
Leipzig.
Tuggle, H. Davis. 1966. Cultural Inferences from the
Art o f ElTajn, M xico. M.A. thesis. University o f
Arizona. Tucson.
Ureta y Peralta, Pedro. 1792. Manifestacin de la
pureza de voces y conceptos de la Descripcin de
Arica y su Provincia contra las dudas y reparos del
ignorante, contenidos en el Mercurio nmero 193.
Mercurio peruano, VI, 2 1 3 20. [Lima]
Vaillant, G eorge C. 1938. A Correlation o f Archae
ological and Historical Sequences in the Valley o f
M xico ' American Anthropologist, n.s., XL, no. 4, pt.
1, 5 3 5 -7 3 .
______ 1939. Indian A rts o f N orth America. N ew York.
______ 1942. T he Aztecs o f Mxico. N ew York.
Valcrcel, Luis E. 1933. Esculturas de Pikillajta. Re
vista del 1Museo Nacional, II, 2 1 48. [Lima]

371

Vico, Giovanni Battista. 1725. P rincipj d i una scienza

nuova. Naples.
______ 1928. La Scienza nuova seconda giusta l'edizione
del 1744, con le varianti, dell edizione del 1730 e d i due
redazioni intermedie inedite, ed. F. N icolini. 2 vols.
Bari.
______ 1931. La Scienza nuova prim a, ed. F. Nicolini.
Bari.
______ 1944. The Autobiography o f G iam battista Vico,
trans. and ed. M. H. Fisch and T. G. Bergin. Ithaca.
Villa R., Alfonso. 1934. The Yaxuna-Cob Causeway.
Carnegie Institution o f W ashington, Publication
436: Contributions to American Archaeology, IX.
Washington.
Villagra Caleti, Agustn. 1947. Las pinturas de B o
nampak. Cuadernos Americanos, VI, 15168.
--------- 1954. Trabajos realizados en Teotihuacn, 19 5 2 .
Anales del Instituto N acional de Antropologa e H isto
ria, VI, 6 9 - 7 8 . [M xico]
Virchow, R. 1873. H olzgotzen von den Guano-Ins e l n Zeitschrift f r Ethnologie, V, 15355. [Berlin]
Vitruvius, 1926. Ten Books on Architecture. trans. M. H.
Morgan. Cambridge, Mass.
von Winning, Hasso. 1947a. Representations o f Tem
ple Buildings as Decorative Patterns on Teotihuacn
Pottery and Figurines. Carnegie Institution o f W ash
ington: Notes on M iddle American Archaeology an d E th
nology, III, no. 83, 17077. [Cambridge, Mass.]
______ 1947b. A Symbol for Dripping Water in the
Teotihuacn Culture. E l Mxico Antiguo, VI, nos. 9
12, 3 3 3 -4 1 . [M xico]
______ 1948. The Teotihuacn OwI-and-Weapon
Symbol and its Association with Serpent Head X at
Kaminaljuy. American A n tiq u ity , X IV , no. 2,
129 - 32 .
______ 1949. Shell D esigns on Teotihuacn Pottery.
El Mxico Antiguo, VII, 12653.
______ 1958. Figurines with M ovable Limbs from An
cient M xico. Ethnos, X X III, no. 1, 160. [Stockholm]
______ 1961. Teotihuacn Symbols: The R eptiles Eye
G lyph. Ethnos, X X V I, no. 3, 1 2 1 -6 6 .
______ 1963. Una vasija de alabastro con decoracin
en relieve. Estudios de cultura M aya, III, 1 1 3 -1 8 .
______
1968.
D er Netzjaguar in Teotihua
cn . . . eine ikonographische Untersuchung. E th
nos, X V I, 3 1 -4 6 .
Weitzmann, Kurt. 1947. Illustrations in Roll a n d Codex:
A Study o f the Origin an d M ethod o fT e x t lllustration.
Princeton.
Whitaker, Arthur Preston. 1941. The Huancavelica
Mercury M ine. Cambridge.

372

A ncient Am erica

W icke, Charles R. 1956. Los murales de Tepantitla y

el arte cam pesino. Anales del Instituto Nacional de
Antropologa e Historia, VIII, 11722. [M xico!
______ 1965. Olmec: An Early Art Style o f Pre-Columbian M xico. Dissertation. University o f Arizona. Tucson.
______ 1971. Olmec: A n Early A r t Style of Pre-Columbian
Mxico. Tucson.
W iener, Charles, 1880. Prou et Bolivie. Paris.
W illey, Gordon R. 1971. Com m entary. In Observations o f the Emergence o f C ivilization in Mesoamerica,
Wenner-Gren Foundation fo r Anthropological Research
Symposium, B urgW artenstein, 1970, ed. R. F. H eizer
and J. A. Graham. Contributions o f the University o f
California Archaeological Research Facility, XI.
Berkeley.
______ 1972. T he A rtifacts o f A lta r de Sacrificios. Pea
body M useum o f Archaeology and Ethnology, Har
vard University, Papers, LXIV, no. 1. Cambridge,
Mass.

______ 1974. T he Classic Maya Hiatus: A Rehearsal

for the Collapse? In Mesoamerican Archaeology: New
Approaches, ed. Norm an Ham m ond, pp. 4 1 7 30.
University o f Texas Press. Austin. [Proceedings o f a
Symposium on M esoamerican Archaeology held by
the University o f Cambridge Centre o f Latin Ameri
can Studies, August 1972]
Williams, H ., and R. F. H eizer. 1965. Sources of
Rocks U sed in O lm ec M onum ents. Contributions of
the U niversity o f C alifornia A rchaeological Research Fa
cility, I. [B erkeley]
W illson, R. W. 1924. Astronomical Notes on the Maya
Cdices. Peabody Museum o f Archaeology and Eth
nology, Harvard University, Papers, VI, no 3. Cam
bridge, Mass.
W ittkower, R. 193839. Eagle and Serpent: A Study
in the Migration o f S y m h i s Journal o f the Warburg
and Courtauld Institutes, II, 2 9 3 325.
Zavala, Silvio A. 1935. La encomienda indiana. Madrid.

III-l Basalt head of a dead man from Tenochtidn, Mxico, c. 1500 (Mxico Ciry, Mu
seo Nacional de Antropologa).

III-2 Stone head of Quetzalcatl, c. 1500 (Mxico City, Museo

Nacional de Antropologa).

III-3 Basalt figure of a kneeling woman from the Valley of

Mxico {?), c. 1500 (New
York, American Museum of
Natural History).

III-4 Stone figure of a standing man, fifteenth century

(Mexico City, Museo Nacional de Antropologa).

. ii

;ih

' i 1i

III-5 Diorite head of Coyolxauhqui, probably from Tenochtitln, Mxico, fifteenth century (Mxico City, Museo Nacional de Antropologa),

II1-6 Basalt death figures of Coatlicue, fifteenth century (Mxico City,

Museo Nacional de Antropologa).

I I 1-7 Stone face panel representing Xipe Totee,

probably from Tenochtitlan, Mxico, c. 1500 (London, British Museum).

III-8 Stone figure of standing Xipe Totee

from Tepepan, Mxico, c. 1500 (New
York, Museum of the American Indian).

I I 1-9 Diorite coiled serpent, fifteenth

cencury (Mxico City, Museo Nacional de
Antropologa).

III-11 Stone jaguar-vessel for blood sacrifices, from Tenochtitlan, Mxico, c.

1500 (Mxico City, Museo Nacional de Antropologa).

III-13 Carneolite grasshopper, fifteenth century (Mxico City,

Museo Nacional de Antropologa).

I II - 10 Stone Quetzalcoatl, fif

teenth century (Vienna, Naturhistorisches Museum).

111*12 Andesite feline figure, fifteenth century

(Mxico City, Museo Nacional de Antropologa).

III - 14 Porphyry calabash, fifteenth century (Mxico City,

Museo Nacional de Antropologa).

II 1-15 B. de Sahagn, Codex Florentino,

after 1569 (Florence, Biblioteca Laurenziana). Scene of heart sacrifice. From
Paso y Troncoso, 1907.

I II-16 Map of the Peruvian Guano Islands. From Murphy,

1936.

III-17 Photograph of the guano stack at N orth Chincha Island, Per, c. 1860 (New
York, American Museum of Natural History).

I
111-19 Wooden staff found
before 1861 on the Chincha
Islands, Per, colonial period
(Salem, Mass., Peabody Museum).

III-20 Wooden staff with

helmeted female figure
found before 1873 on the
Chincha Islands, Per,
fourteenth or fifteenth
century
(?)
(present
whereabouts unknown).
From Hutchinson, 1873.

III-18 Armorial slab of stone

found in 1847 by W. Bollaert,
from North Chincha Island,
Per, 1575-1600 (?) (present
whereabouts unknown). From
London lllustrated Times, 1859.

III-2 1 Silver fishes collected by E. G. Squier and found before

1867 on the Chincha Islands, Per, fourteenth or fifteenth cen
tury (?) (New York, American Museum o f Natural History).

111-22 Wooden figure found on

the Chincha Islands, Per,
Tiahuanacoid period (present
whereabouts unknown). From
Hutchinson, 1873-

111-23 Five objects found at the 62-foot level on the Chincha Islands, Per, Mochica period (present whereabouts unknown). From
Hutchinson, 1873.

I I 1-24 Wooden staff found be

fare 1870 on South Guanape Island, Per, Inca period (?) (London, British Museum).
111-25 Penguin resting on cloth, found between 1867 and 1873 on the
Guaape Islands, Per (present whereabouts unknown). From Hutchinson,
1873.

111-26 Three wooden figures found in 1871 on

North Macabi Island, Per, Mochica period (formerly Hamburg, present whereabouts unknown).
From Virchow, 1873.

I I 1-27 Map of Paracas Bay in Per, 1713. From Frzier, 1732.

111-28 Wood and pottery objects

found in 1870 on North Macab Island,
Per, Mochica period (London, British
Museum). From joyce, 1912a.

111-29 Wooden male figure found in

1870 on North Macab Island, Per,
Mochica period (London, British Museum). Fromjoyce, 1912a.

111-31 Wooden female figure found by J. P. Davis in

1863-64, from South Guaape Island, Per (present
whereabouts unknown). From Squier, 187172.

111-30 Vessel showing visit of boats to

guano rocks, Mochica period (Chicln,
Per, Museum of Hacienda Chicln).
Courtesy of R. Larco Hoyle.

I I 1-32 Sheet gold figure drawn and exhibited by T.

Ewbank in New York in 1859, from the Chincha Islands,
Per (present whereabouts unknown). From Squier,
1871-72.

II 1-33 Gold female figure acquired by

a Lima merchant between 1870 and
1900, from the Chincha Islands, Per
(Berlin, Museum fr Vlkerkunde).
From Schmidt, 1929.

"a ry b a llo s *

111-34 Wooden, helmeted,

cup-holding figure from the
Guanape Islands, Per, fourteenth or fifieenth century (?)
(Berlin, Museum fr Volkerkunde). From Schmidt,
1929.

I I 1-35 Domestic artifacts found in

1868 in a survey of the Macabi Islands,
Per, Mochica period (Bordeaux,
France, Muse Prhistorique et dEthnographie; unlocated 1947). From Dulignon, 1884.

blackware

111-36 Pottery found in 1860 by Commander

Jones on the Chincha Islands, Per, Inca period
(Mystic, Conn., Marine Museum).

111-37 Wooden staffwith throne

and seated figure found in 1870
on N orth Macabi Island, Per,
Mochica period (London, British
Museum). From joyce, 1912a.

111-38 Wooden
mace
with
throne and seated figure found in
1870 on N orth Macabi Island,
Per, Mochica period (London,
British Museum). From Joyce,
1912a.

111-41 Ceramic seated figure

found between 1875 and 1877 by
C. Wiener on Lobos Island,
Per [probably Macabi}, Mochica
period (present whereabouts unknown). From Wiener, 1880.

111-39 Figures found on the Chincha

Islands, Per (present whereabouts unknown). From Hutchinson, 1873.

IH-42 Wooden head of mummy

bundle found between 1875 and
1877 by C. Wiener on Lobos Island, Per [probably Macab],
Mochica period (present where
abouts unknown). From Wiener,
1880.

111-40 Wooden figures found between

1875 and 1877 by C. Wiener on Lobos
Island, Per [probably Macab],
Mochica period (present whereabouts
unknown) From Wiener, 1880.

111-43 Wooden staff with throne, canopy,

and seated figure found between 1875 and
1877 by C. Wiener on Lobos Island, Per
[probably Macabi], Mochica period (pre
sent whereabouts unknown).
From
Wiener, 1880.

111-44 Wooden figures from Caete Valley, Per, Mochica period (present where
abouts unknown). From Wiener, 1880.

III-46 Teotihuacan, Mxico, Perspective dniwing of site, c. 600. Courtesy of R. Milln.

111-47 Monte Albn, Mxico, Plan of site, before 900. From Marquina, 1951.

I I 1-49 Xochicalco, Mxico, Plan of southwest group with ball-court, bcfore

900. From Marquina, 1951.

III
48 Mida, Mxico, Plan of site, as in the
tenth century (?). From Holmes, 189597.

L **

TODITO -

111-50 Chichn Itz, Mxico, Plan and elevation of the Mercado, before
1200. From Ruppert, 1943.

o
O
111-51 Tula, Mxico, Plan of the south group, before 1200. From Acosta,
1943-44.

20METRF.S
6 0 FEET

111-52 Atetelco (Teotihuacan), Mxico, Elevation

and plan o a dwellinggroup, c. 500. From Marquina
and Armillas, 1935.

111-53 Mapa de Quinatzn, sixteenth century

(Paris, Bibliochque Nationale). Drawing of the
palace court at Texcoco, Mxico.

111-54 Teotihuacan, Mxico, Wall painting from the Temple of Agriculture (lost ordestroyed), Period II, before 300 ( *). From Gamio, 1922.

I I 1-55 Uaxactn, Guatemala, Wall painting in Structure B-XIII, before 600.

111-56 Monte Alban, Mxico, Wall paintings

in Tomb 104, c. 600. Drawing by Barry Legg,
based on M. Covarrubias, 1957. Courtesy of
Alfred A. Knopf, Inc.

I I 1-57 Bonampak, Mxico, Wall paintings in Room 2, c. 800. Wall 5 (arraignment

of victims) and wall 6 (battle). Courtesy of the Instituto Nacional de Antropologa e
Historia.

111-60 Chichen Itz, Mxico, Wall painting

in the Temple of" the Jaguars (inner room,
south wall), before 1200 (?). Siege operations
during a battle. Courtesy of the Peabody Mu
seum, Harvard University.
111-58 Bonampak, Mxico, Wall paintings in Room 2, c. 800. Walls 7 and 8 (battle).
Courtesy of the Instituto Nacional de Antropologa e Historia.

111-59 Chichn Itz, Mxico, Wall pjinting in the Temple of the

Warriors, before 1050 (?). A seacoast village. From Morris, Charlot, and Morris, 1931.

111-61 Bonampak, Mxico, Wall paintings in Room 1, c. 800. Right-hand panel: wall 1 (a robing
ceremony) and wall 2 (procession in lower register). Courtesy of the Instituto Nacional de
Antropologa e Historia.
111-62 Bonampak, Mxico, Wall paintings in Room L, c. 800. Left-hand panel: wall 3 (presentation
of an infant) and wall 4 (procession in lower register). Courtesy of the Instituto Nacional de
Antropologa e Historia.

II 1-63 Bonampak, Mxico, Wall painting in Room 3, c. 800. Right-hand panel: wall 11 (dancers); and left-hand panel: wall 12
(procession of an image). Courtesy of the Instituto Nacional de Antropologa e Historia.

30 Fkfc'f

111-64 Bonampak, Mxico, Room 1, c. 800, Perspective view, elevation, and plan. Courtesy of the Carnegie Institution of
Washington.

111-65 Santa Rita, British Honduras, Wall painting with gods and calendar glyphs, tenth century (?). From
Gann, 1900.

111-66 Tulum, Mxico, Wall paintings in the west passage of the Temple of
the Frescoes, eleventh-twelfth centuries (?). From Lothrop, 1952.

111-67 Codex Perestanus, after 1000 (?)

(Paris, Bibliothque Nationale). Calendrical pages. From L. de Rosny, 1887.

111-68 Tizatlan, Mxico, Altar paintings showing Tezcatlipoca, after 1000. Courtesy of L.
Aveleyra.

111-70 Mida, Mxico, Lintel paintings from the Arroyo and Church groups, after 1000 (?). (A)
Arroyo north, (B) Church east, (C) Church north, and (D) Church west. From E. Seler, 1902-23.

111-71 Zacuala (Teotihuacan), Mxico, Wall painting of agoggled celebrant bearing acorn plant and
censer, before 550 (?). Courtesy of A. Miller.

111-72 Tepantitia(Teotihuacn), Mxico, Wall painting, Period III, before 550 (?). Detall of amural
border with twined serpents and of agoggled raingod bearinggoggled raingod scepters. Courtesy of
A. Miller.

111-73 Tepantitla (Teotihuacn), Mxico, Wall painting (lower register: 2.40 x 1.10 m.), Period III,
before 550 (?). Restoration of Paradise mural with a twined serpent border (Mxico City, Museo
Nacional de Antropologa). Courtesy of the Instituto Nacional de Antropologa e Historia.

111-74 Tepancitla(Teotihuacan), Mxico, Wall painting, Period III, before 550 (?). Restoration of the fragmented
upper register above the Paradise mural, showing a frontal female figure lanked by celebrants (Mxico City,
Museo Nacional de Antropologa). Courtesy of the Instituto Nacional de Antropologa e Historia.

111-75 Tepantitla (Teotihuacn), Mxico, Wall painting, Period III, before 550 (?). Drawing of female figure in
figure III -74. Courtesy of the Instituto Nacional de Antropologa e Historia.

111-76 Tepantitla (Teotihuacan), Mxico, Paradise mural, Period III, before 550 (?). Detail of
frontal human figure from upper right quadrant height oi head: 0.04 m.). Courtesy of A. Miller.

111-77 Tetitla (Teotihuacan), Mxico, Wall painting, before 550 (?). Detail of a frontal female
head (height: 0.33 m.). Courtesy of A. Miller.

111-78 Atetelco (Teotihuacan), Mxico, Wall paintings, c.

500 (?). Reconstruction of repeated figure of profile celebrant. Courtesy of A. Miller.

111-79 Atetelco (Teotihuacan), Mxico, Wall painting in East Strucrure (lower register: 0.74 x 1.98 m.; border width: 0.20 m.), c.
500 (?). Upper register: celebrants carrying conch shelis which have water-dripping speech scrolls; lower register: netted jaguar
{left) and coyote (right) within a frame of twined serpent bodies with clawed feet. Courtesy of A. Miller.

..

M&9 s
111-80 Atetelco (Teotihuacan), Mxico, Wall painting in East Structure, c. 500 (?). Celebrant figure
framed by twined serpent bodies (height of figure: 0.42 m.). Courtesy of A. Miller.

M M fe fe

111-81 Teotihuacan, Mxico, Wall painting of Room 19 m

Zone 5a (0.80 x 0.75 m.), before 550 (?). Detail of a celebrant
dressed as a bird and carryingan aspergillum and an incense bag.
Courtesy of A. Miller.

111-82 Teotihuacan, Mxico, Wall painting of Room 19 in

Zone 5a (2.20 X l.36m .), before 550(?). Reconstruction of the
celebrant in figure 111-81 and of the serpent-bird border with
wings, radiant discs, conch shell sections, and footprints. Alter
L. Sjourn, 1966c.

111-83 Atetelco (Teotihuacan), Mxico, Wall painting, c. 500 (?). Lower register with armed figures at dance platform or court.
Courtesy of A. Miller.
=*==_______ - _______ _________________

111-84 Atetelco (Teotihuacan), Mxico, Wall paint

ing (figure: 0.50 x 0.75 m .), c. 500 (?). Drawing of
personage shown in fig. 111-83. After L. Sjourn,
1966c.

111-85 Tetitla (Teotihuacan), Mxico, Wall painting of jaguar-man approaching a temple (0.73 x 2.12 m.), before 550 (?) (Wash
ington, D.C., Dumbarton Oaks).

111-86 Teotihuacan, Mxico, Talus painting ol Palacio de los Jaguares (0.82 x 1.69 m.), before 550
(?). Frameofdiademsalternaring with starfish surrounds jaguar blowing aconch shell. Courtesy of A.
Miller.

111-87 Teotihuacan, Mxico, Wall painting (0.83 x 1.17 m.),

before 55 (?) (Cleveland, The Cleveland Museum, Purchase
from thej. H. Wade Fund). Fragmentofasingingpriestflanked
by maguey leaves.

111-89 Zacuala (Teotihuacan), Mxico, Wall painting

of a jaguar-warrior with a feathered shield (height:
1.06 m.), before 550 (?). Courtesy of A. Miller.

111-88 Teotihuacan, Mxico, Wall painting from the Temple

of Agriculture (lose or destroyed), Period II, before 300 <?).
From Gamio, 1922.

111-90 Tetitla (Teotihuacan), Mxico, Lava brazier

with old-man effgy (height: 0.60 m.), before 550 (?).
Courtesy of A. Miller.

IU-92 Stone jaguar-vessel from the foot of the Pyramid of the Sun at Teotihuacn,
Mxico, Period II, bt'fure 300 (?) (London, British Museum).

I I 1-94 Zacuala (Teotihuacan), Mxico, Wall painting, before

550 (?). Fragment of a butterfly figure with a curl (RE) glyph
pectoral. After Sjourn, 1959.

111-93 Tetirla (Teotihuacan), Mxico, Wall painting, before

550 (?). Detail with a charismatic hand. Courtesy of A. Miller.

111-95 Teotihuacan, Mxico, Wall painting from the Temple

of Agriculture, Period II, before 300 (?). Fragment with compound jaguar-bird-serpent forms shown leaping in waves.
Courtesy of A. Miller.

111-96 Teotihuacn, Mxico, Wall painting from the Temple

of Agriculture, Period II, before 30 (?). Replica of part of fig.
II1-95. Courtesy of the Instituto Nacional de Antropologa e
Historia.

111-97 Teotihuacn, Mxico, Wall painting from the Temple

of Agriculture, Period II, before 30 (?). Replica ofafragment
with a feathered serpent and a dog-jaguar. Courtesy of the In
stituto Nacional de Antropologa e Historia.

I I 1-99 Zacuala (Teotihuacan), Mxico, Wall painting, before

550 (?). Drawing of a detail showing a human figure carrying a
censer and shield and wearing a netted jaguar helmet with feathered quetzal eyes. From Sjourn, 1966b.

111-98 Painted pottery brazier with four directional symboJs

from Burial 10 at Zacuala (Teotihuacan), Mxico, Period IV,
C-14 date (Y1264): a .d . 330-80. From Sjourn, 1959.

III-100 Xochicalco, Mxico, Stone relief carvings of curl (RE)glyphs, before 900. Courtesy
of A. Miller.

111-101 Relief tripod vessel from Burial 2 at Zacuala (Teoti

huacan), Mxico, after 300 (?). Drawingofcurl (RE)glyph alternating with a flower set upon a butterfly wing. From Sjourn,
1959-

I II-102 Incised tripod vessel from Teotihuacan, Mxico, after

300 (?). Drawing of alternating tripled mountains and face panels incorporating the curl (RE) glyph as the mouth. From von
Winning, 1961.

III-104 Relief tripod vessel Irom Burial 2 at Zacuala (Teotihuacan), Mxico, after 300 (?).
Drawing of a design representing a goggled rain figure alternating with a flayed-skin face.
From Sjourn, 1959.

I II-103 Molded clay figurine

from Teotihuacan, Mxico, after
550 (?). Drawing offat man with a
curl (RE) glyph stamped on his
forehead. From von Winning,
1961.

from Teotihuacan, Mxico, Peri

od IV, after 550 (?). Drawing of a
warrior wearing a headdress with
a quetzal emblem. From S
journ, 1966b.

I I I - 105 Incised tripod vessel from Teotihua

can, Mxico, after 300 (?). Drawing of a frag
ment with four-element (lightning-fire-rainwater) pattern. From von Winning, 1958.

I II-106 Stamped clay figurine from Teoti

huacan, Mxico, Period IV, after 550 (?).
Drawing of designs on headdresses. From
von Winning, 1958.

111-108 Incised clay vessels from Teotihuacan, Mxico, after 300 (?). Drawings of mouth
(RE) and curl (RE) glyphs with pendant water-drops. From von Winning, 1961.

k A -

I II-109 Teotihuacan, Mxico, Stone piers

surrounding court ofQuetzalpapalotl building,
before 700. Left: west range with frontal owlserpent form, originally painted green and red,
with obsidian insets in eye friezes. Right: East,
north, and south ranges with profile quetzal
bird with fame borders top and bottom. From
Acosta, 1964.

i * * A A A

o)o loto o io jo io jcM o ;

I
r

\\V \S ^

I I I - 110 Tetitla (Teotihuacan), Mxico, Mural panel of an owl with spread wings (owl: 0.61-1.10 m.), before
550 (?). Courtesy of A. Miller.

I I I - l l l Tetitla (Teotihuacn), Mxico, Wall painting, before

550 (?). Drawing of heads of owls with water dripping from
their mouths and a border of raingod masks. After Sjourn,
1966c.

after 550 (?). Drawing of a warrior in quilted cotton

armor, wearing owl and spear panoply, with a human
hand decorating the shield. From Sjourn, 1966b.

III- 113 Molded orangeware bowl from Las Colinas (near Calpulalpan), Mxico, Period
III, before 550 (?). Drawing. From Linn, 1942.

4BLSLfSLSLJBLJL&
I II - 114 Relief tripod vessel from Teotihuacn, Mxico, after
300 (?). Drawing of the design of a radiant flower on a butterfly
wing alternating with a frontal butterfly in goggles and of the
lower border of blossoms. From Seler, 1915.

III-l 15 Molded and painted clay adornos of butterflies for braziers from Teotihuacan, Mxico,
after 550 (?). Drawings. From Sjourn, 1959.

III-116 Relief tripod vessel from Teotihuacan, Mxico, after 300 (?). Drawing of feathered eye on a crossroads beneath a mountain and over a curl (RE) glyph, alternating with
dripping eyes, and of a lower border of conch sections. From von Winning, 1961.

%
z

*
*

III-l 17 Teopancaxco (Teotihuacan), Mxico, Wall painting in Casa de los barrios, before 550 (?). Replica of a mural depicting a cultemblem flanked by celebrants wearing bird-jaguar headdresses decorated with starfish. Courtesy of the Instituto Nacional de
Antropologa e Historia.

III-l 19 Teotihuacan, Mxico, Ciudadela, inner face of

the central pyramid, before 300. Courtesy of the In
stituto Nacional de Antropologa e Historia.
III-l 18 Teotihuacan, Mxico, Structure 5 bordering the Moon
Plaza, after 450. Sections of tableros. From Acosta, 1964.

I B

I I I - l20 Tripod vessels from Teotihuacan, Mxico, after 300

(?). Drawings of slab feet. From Sjourn, 1966.

III-122 Frescoed tripod vessel

from Teotihuacan, Mxico,
after 300 (?) (Mxico City, Kurt
Stavenhagen Collection). Draw
ing. From Sjourn, 1960.

I I I - l21 Frescoed tripod vessel from Teo

tihuacn, Mxico, after 300 (?). (Mxico
City, Diego Rivera Museum). From Artes de
Mxico, 1960.

III-124 Painted red-onocre vessel from Zacuala

(Teotihuacn),
Mxico,
after 300 (?). Mountain
III-123 Plano-relief tripod vessel purchased at Santiago Ahuzotla, form capped by tableroMxico, after 300 (?) (Hollywood, Cal., Hasso von Winning Collec shape. From Sjourn,
tion). Drawing. From von Winning, 1947.
1959.

III-125 Tikal, Guatemala, Stela 10, ded.* 9.5.5.0.0 (539), style 9.8.0.0.02
(59340). Drawings of east (A, B) and west faces (C, D), Courtesy of the
Tikal Project, University Museum, Pennsylvania. [The dates for chis figure
and for figures 111-126 through III-186 have been provided by the editor.
There are generally three kinds of notations before the date to indicate the
nature of the evidence. Ded.* refers to a date described as a dedicatory date
in Linda Schele, Maya Glypbs: The Verbs (Austin, 1982); style indicates a
stylistic date from T. Proskouriakoff, A Study of Classic Maya Sculpture
(Washington, 1950); and text records means that the date is mentioned in the
text of the inscription.}

I II-126 Tikal, Guatemala, Stela 10, ded.*

9.5.5.0.0 (539), style 9.8.0.0.02 (59340).
Drawing of the rear. Courtesy of the Tikal Pro
ject, University Museum, Pennsylvania.

a-

cL

&

ft'

I I I - l27 Quirigua, Guatemala, StelaC,ded.* 9.17.5.0.0(77j),style9.18.0.0.02(79040). Drawings of case

(A, B) and west faces (C, D). From Maudslay, 1889-1902.

I I I - 128 Q u iriR u a , G u a te m a la , S te la F, d e d .* 9-1 7 .0 .0 .0 (7 7 1). s ty le 9. 17.().<).()2 (7 7 1 i 4 0 ). D raw in fS o f w e s t (A , B ) a n d e a s t fa c e s (C , D ).

III-129 Quirigua, Guatemala, Stela D, ded.* 9.16.5.0.0 (756), style 9.17.0.0.02 (77140). Drawings of west face (A, B). From
Maudslay, 1889-1902.

I I I - l30 Quirigu, Guatemala, Stela D, ded.* 9-16.5.0.0 (756), style 9.17.0.0.02 (77140). Drawings of east face (C, D). From
Maudslay, 18891902.

III-131 Copan, Honduras, Stela N, ded.* 9-16.10.0.0 (761), style 9 .16.10.0.02 (761 40). Drawings of east (A) and west sides (B).
From Maudslay, 1889-1902.

III-132 Palenque, Mxico, Sancruary tablet in the Temple of the Cross, ded."
9.13.0.0.0(692). Drawing. From Maudslay, 18891902.

III-133 Palenque, Mxico, Sanctuary tablet in the Temple of the Foliated Cross, ded.* 9.13.0.0.0(692). Drawing. From
Maudslay, 1889-1902.

111-134 Palenque, Mxico, Sanctuary tabletin the Temple of the Sun, ded.* 9-13.0.0.0(692). Drawing.
From Maudslay, 1889-1902.

III-l 36 Palenque, Mxico, Stairway tablet of House

C in the Palace, middle cycle 9; text records
9-8.9.13.13 (603). Drawing. From Maudslay, 1889
1902 .

I I I - 137 P a le n q u e , M x ic o , E ast p a n e l o f th e T e m p le o f th e In s c rip c io n s, c. 9 .1 3 .0 .0 .0 ( c. 6 9 2 ) . D ra w in g . F ro m M a u d s la y , 1 8 8 9 - 1 9 0 2 .

III-13H Palenque, Mxico, Central panel of the Temple of the Inscriptions, c. 9.13-0.0.0 (c. 692). Drawing. From Maudslay, 18891902.

III-l 38 Palenque, Mxico, Central panel of the Temple of the Inscriptions, c. 9.13.0.0.0 (c. 692). Drawing. From Maudslay, 18891902.

I I I - 140
Palenque,
Mxico, South door
jam bof Temple XVIII
middle to late cycle 9, text records
9.12.6.5.8
(977).
Drawing. From Senz,
1956.

III-l 39 Palenque, Mxico, West panel of the Temple of the lnscriptions, c. 9.13.0.0.0 (c. 692). Drawing. From Maudslay, 18891902.

III-141 Yaxchiln, Mxico, Lintels 35 and 37 of

Structure 12, text records 9.4.11.8.16 (525).
Drawings. From Maudslay, 1889-1902.

III-142 Tikal, Guatemala, Stela 31, ded.* 9.0.10.0.0 (445).

Drawing of the rear face. Courtesy of the Tikal Project,
University Museum, Pennsylvania.

III-143 Palenque, Mxico, Tablet of the 96 Glyphs, ded.* 9.17.15.0.0 (?85). Drawing. From Palacios, 1936.

III-144 Palenque, Mxico, Tablet of the 96 Glyphs, ded.* 9.17.15.0.0 (785). Drawing. From Berlin, 1968.

G 1

III-145 Palenque, Mxico, Tablee of the 96 Glyphs, ded.* 9 17.15-0.0 (785). Kublers segmentation {right) contrasted with H.
Berlins in 1968 (left). Diagrams by G. McDonogh. (For ease of reference, I have kept the Maya reading order by pairsof columns, read
from left to right and top to bottom. The inscription is also diagrammed in continuous order from top to bottom and from beginning to
end. The vertical axis retains the left and right columns of the original, instead of stringing the inscription in single horizontal lines, as
preferred in the studies of Berln and Proskouriakoff. Here and in plates 111-147,111-149, and 111-151 cross-hatching equals Calendar
Round expressions, vertical hatching equals Distance Numbers, and the horizontal hatching in pate III-151 stands for Period
Endings.)

III-146 Relief panel 1 from El Cayo, Mxico (?), d e d / 9.17.15.0.0 (785), style
9.17.10.0.02 (78040) (Washington, D.C., Dumbarton Oaks Collections). Drawing by M.
Coe.

I I I - 147 Relief panel 1 from El Cayo,

Mxico (?), ded.* 9.17.15.0.0 (785),
style 9.17.10.0.0+2 (78040) (Wash
ington, D.C., Dumbarton Oaks Collec
tions). G. Kublers segmentation. Diagram by G. McDonogh. (Right-slanted
hatching equals Initial Series; leftslanted hatching equals Secondary
Seriesj.

I II - 148 Lintel 25 from Yaxchilan, Mxico, ded.* c. 9.14.15.0.0 (c. 726) (London, British Museum). Drawing of the front edge. From
Maudslay, 1889-1902.

III-150 Wall panel (Lintel 2) from Piedras Negras, Guatemala, ded.* 9.11.15.0.0 (667), style
9.13.10.0.02 (70240) (Cambridge, Mass., Peabody Museum). Courtesy of the University
Museum, Pennsylvania.

III-149 Lintel 25 from Yaxchilan, Mxico, ded.* c.

9.14.15.0.0 (c. 726) (London,
British Museum). G. Kublers
segmentation. Diagram by G.
McDonogh-

III-151 Wall panel (Lintel 2) from Piedras N e

gras, Guatemala, ded.* 9 11-15.0.0 (667), style
9 .13.10.0.02 (70240) (Cambridge, Mass., Pea
body Museum). G. Kublers segmentation. Diagram
by G. McDonogh.

III-152 Tikal, Guatemala, Wooden lintel 2 of Tem

ple I, c. 780. Drawing. Record ofanevent occurringc.
700. Courtesy of W. R. Coe.

III-l 53 Wooden lintel 3 of Temple I of

Tikal, Guatemala, c. 751 (Basel, Museum fr Vlkerkunde). Drawing. Re
cord of an event occurringc. 700. Cour
tesy of W. R. Coe.

III-l 54 Wooden lintel 2 of Temple IV

of Tikal, Guatemala, c. 751 (Basel, Museum fr Vlkerkunde). Drawing. Re
cord of an event occurringc. 751. Cour
tesy of W. R. Coe.

III-l 55 Wooden lintel 3ofTem ple IV ofTikal, Guatemala, c. 7 5 1 (Basel, Museum fr Vlkerkunde). Drawing. Record
of an event occurring c. 751.

I I I - l 56 Tikal, Guatemala, Stela 31,

ded.* 9.0.10.0.0 (445). Left side.
Courtesy of W. R. Coe.

I I I - l 57 Onyx marble bowl from southwestern Maya regin

(height: 11.6 cms.; diameter: 15.8 cms.), ded.* 9.17.11.5.19 (782)
(Washington, D.C., Dumbarton Oaks Collections, Access, no.
B-147.MAS). Woman at A.

III - 161 Palenque, Mxico, Pier C of House D at the

Palace, style c. 9.12.0.0.0 (c. 67240). Drawing of a
detail. From Maudslay, 18891902.

III-162 Palenque, Mxico, Sarcophagus from the Tem

ple of the Inscriptions, ded.* c. 9.13-0.0.0 (c. 692).
Drawings of serpent heads by L. Schele.

oooo

III-163 7- and 9-heads in chronological sequence (see table 111-12). Drawings by E. Jovce.

III-164 Copn, Honduras, Altar T ', text records

9.17.12.5.17 (782). Drawing of adetail. From Morley, 1920.

III-166 Yaxha, Guatemala, Stela 4, style 9-l lO.O.O2

(465 40). West side. From Maler, 1908.

111-165 Tikal, Guatemala, Inscribed obsidians from Cache

161, c. 914.0.0.0 (c. 711). Drawing by W. R. Coe.

I II - 167 Yaxha, Guatemala, Stela 2, style

(53440). West side. From Maler, 1908.

9-5.0.0.02

I II-168 Yaxh, Guatemala, Plan of the east side of Plaza C, Early Classic. From Maler, 1908.

lp

(o ri)

III-169 Tikal, Guatemala, Stela 2, style 9-3.10.0.0 2 (50440). Front. Drawing by W. R. Coe. Cour
tesy of the University Museum, Pennsylvania.

III-170 Copan, Honduras, Stela D,

rext records 9.15.5.0.0 (736), style
9.15.10.0.02 (74140). Drawing of
south face. From Maudslay, 18891902

H I-171 Copan, Honduras, Stela D, text records 9.15.5.0.0 (736),

style 9-15.10.0.0 2 (74140). Drawngs of east and west sides.
From Maudslay, 1889-1902.

111-172 Skeletal serpent heads of the T I 035 type from the Tablets of the Sun, Slaves, and
Palace at Palenque, Mxico. Drawngs. l:rom Schele, 1974.

I I I - l73 Carved bowl from Burial 132 atTikal, Guatemala,

before 554 (Tikal, Guatemala, Morley Museum). Drawing
by W. R. Coe. Courtesy of the University Museum,
Pennsylvania.

I I I - l74 Cancuen, Guatemala, Stela 1, style 9.17.1 .0.()2

(78040). East and west faces. From Maler, 1908a,

I I 1-175 Aguas Calientes, Guatemala, Stela 1, stvle 9.17. lO.O.Oir 2 (78040). Front and rear faces. From Morley, 1937
38.

III-176 Stela 40 at Piedras Negras, Guatemala, style

9-15.0.0.0 (731 40) (Guatemala, Guatemala, Museo
Nacional). From Morley, 1937-38.

I I I - l77 Copn, Honduras, Altar T, text records

9.17.12.5.17 (782). Drawngs of top and of north (right)
and south sides (below). From Maudslay, 18891902.

III-178 Onyx marble bowl from Jaina Island, Mxico (height:

11.3 cms.;diameter: 14.7 cms.), 9.16.0.0.09.19.0.0.0 (c. 751
810) (Washington, D.C., Dumbarton Oaks Collections, Access. no. B-553.MAS).

I I 1-180 Naranjo, Guatemala, Stela 30, ded.* 9.14.3.0.0

(714), style 9 .14.0.0.02 (711 40). From Proskouriakoff, 1950.

III-181 Naranjo, Guatemala,

Stela 33, style 9.16.0.0.0 2
(751 40). From Proskouriakoff, 1950.

I I I - l82 Tikal, Guatemala, Altar 5, ded.* c. 9.14.0.0.0 (c. 711). Courtesy of the University
Museum, Pennsylvania.

I I I - l83 Tikal, Guatemala, Wooden lintel 2 in Temple III, style

9.19.0.0.0 2 (81040). Drawing. From Coe, Shook, and Satterthwaite, 1961.

III-l 84 Yaxchilan, Mxico, Upper step of the middle doorway

of Structure 44, c. 9.15.0.0.0 (c. 731). From Morley, 1937-38.

III-185 Bonampak, Mxico, Wall paintings in Ruom 1 of

Structure 1, wall 4, c. 9.18.10.0.0 (c. 800). From Ruppert,
Thompson, and Proskouriakoff, 1955.

I I I - l86 La Florida, Guatemala, Stela 9, style 9 .16.0.0.02

(751 40). Drawing. From Graham, 1970.

III-l 87 Codex Mendoza, sixteenth cenrury

(Oxford, Bodleian Library). Nobles wearing
mantles. From Seler, 1904.

I I I - l88 Teotihuacn, Mxico, Terra-cotta figurines wearing jaguar helmets,

after 550 (?). Drawings. From Sjourn, 1962.

-i i-:

L
V.

rsza

I II - 189 Uaxactun, Guatemala, Burial crypts in Pyramid E of

Structure A -l, Tepeu I and lateTepeu II. Drawing. From R. E.
Smith, 1937.

III-190 Uaxactiin, Guatemala, Pyramid E of Structure A-l,

Classic period. Plan, elevations, and sections. From R. E. Smith,
1937.

I II-192 Teotihuacan, Mxico, Incised slate

fragment from the cave beneath the Pyramid of
the Sun, before 550 (?). Drawing of a man wearing a jaguar helmet. From Heyden, 1975. Cour
tesy of the Instituto Nacional de Antropologa e
Historia.

III-191 Tripod polychrome pate from Uaxactn, Guatemala. Tepeu II (Guate

mala, Guatemala, Museo Nacional). From A. L. Smith, 1932. (The pate was
discovered at 2 in figure III-189.)

I II-193 Teotihuacan, Mxico, Northern (Moon) pyramid and plaza, before 600.

III-194 Teotihuacan, Mxico, Ciudadela court, before 600. From che south.

e and f

III-195 Terrace profles in Mesoamerica: (a) the tablero

and talus, Teotihuacan; (b) the dentated profile of Monte
Alban; (c) the slanted and undercut profile of the South
ern Maya lowlands; (d) the binder molding of the northern Maya provinces; (e, 0 the outsloping cornices of
Tajn (e) and Xochicalco (f); (g) the double-sloped balustrade of Aztec architecture. Drawings adapted by E. G.
A. Kubler from Marquina, 1951.

III-197 Tikal, Guatemala, Platform on the N orth Acr

polis, Early Classic period. Reconstruction of an early
building.

111-196 Stone model of a temple from Monte Albn, after 300 (?)
(Mxico City, Museo Nacional de Antropologa). Courtesy of the
Instituto Nacional de Antropologa e Historia.

I II - 198 Edzn, Mxico, Pyramidal platform, Classic period. View of binder

molding in veneer facing over concrete rubble core.

I I I - 199 Tlatelolco, Mxico, Pyramidal plarform with double-pitched stairway,

after 1300.

111-201 Teotihuacan, Mxico, Wall painting

of the Mythological Animis, before 550 (?).
Detail of a jaguar eating a winged figure in the
waves. Courtesy of A. Miller.

111-202 Teotihuacan, Mxico, Wall painting

of the Mythological Animis, before 550 (?).
Detail of a jaguar lacking some claws.

III-200 Cylindrical tripod vessel with

appliqu molded relief of the jaguarserpent-bird icn, before 700 (?) (Teoti
huacan, . Mxico, Museo Regional).
Courtesy C. Reyes Valerio.

III-203 Teotihuacan, Mxico, Incised cylindrical, brown-sJipped vessei, after 300- Drawing of celebrant bringing an oflering to the
jaguar-serpent-bird icn. From von Winning, 1949.

Mxico, before 700 (?) (Mxico City, D.F.,

Museo Anahuacalli). Drawing of a man-jaguar holding a bust with a jaguar-serpentbird helmet. From Sjourn, 1962.

111-205 Drawings of jaguar-serpent-bird icons from Chichn Itz

(a) and Tula (b), Mxico. From
Tozzer, 1957.

III-206 Chichen Itza, Mxico, Tablero reliefs with warriors, eagles, and jaguars, before 1200. Drawing. From Tozzer, 1957.

III-207 La Venta, Mxico, Olmec head no. 1. From Clewlow et al.

III-208 San Lorenzo, Mxico, Olmec head no. 4. From Clewlow et al.