The Effects of Aztec Conquest on Provincial Commoner Households
at Calixtlahuaca, Mexico
by
Angela Huster
A Dissertation Presented in Partial Fulfillment
of the Requirements for the Degree
Doctor of Philosophy
Approved April 2016 by the
Graduate Supervisory Committee:
Michael Smith, Chair
Barbara Stark
Emily Umberger
Katherine Spielmann
ARIZONA STATE UNIVERSITY
May 2016
Pro Q ue st Num b e r: 10108079
All rig hts re se rve d
INFO RMATIO N TO ALL USERS
The q ua lity o f this re p ro d uc tio n is d e p e nd e nt up o n the q ua lity o f the c o p y sub m itte d .
In the unlike ly e ve nt tha t the a utho r d id no t se nd a c o m p le te m a nusc rip t
a nd the re a re m issing p a g e s, the se will b e no te d . Also , if m a te ria l ha d to b e re m o ve d ,
a no te will ind ic a te the d e le tio n.
Pro Q ue st 10108079
Pub lishe d b y Pro Q ue st LLC (2016). Co p yrig ht o f the Disse rta tio n is he ld b y the Autho r.
All rig hts re se rve d .
This wo rk is p ro te c te d a g a inst una utho rize d c o p ying und e r Title 17, Unite d Sta te s Co d e
Mic ro fo rm Ed itio n © Pro Q ue st LLC.
Pro Q ue st LLC.
789 Ea st Eise nho we r Pa rkwa y
P.O . Bo x 1346
Ann Arb o r, MI 48106 - 1346
ABSTRACT
This archaeological study analyses households at the Postclassic site of
Calixtlahuaca (State of Mexico, Mexico), to evaluate the directness and collectiveness of
local and imperial Aztec rule based on their effects on the commoner population.
Scholars are divided as to whether Aztec rule was generally positive (due to opportunities
for economic and cultural interaction) or negative (due to taxation and loss of autonomy).
Contexts at Calixtlahuaca date to three periods, the Dongu (AD 1130-1370), Ninupi
(1370-1450), and Yata (1450-1530) phases. The first two phases show the pre-Aztec
trajectory, which is compared to the final period under Aztec rule to disentangle general
trends toward regional integration from Aztec effects. Each phase includes six excavated
households.
I assess economic changes on three dimensions: foreign trade, local craft
production, and household wealth. Trade is evaluated for obsidian and ceramics (INAA,
petrography, type classification) and local crafting is evaluated for ceramic, lithic, textile,
and molded ceramic items. Wealth is measured using all excavated artifacts, with the
relative values of artifact classes based on Colonial Nahuatl wills. Prior to Aztec rule,
trade was increasing and diversifying, but craft production was low. Under Aztec rule,
trade reoriented toward the Basin of Mexico, craft production remained low, and
household wealth stabilized. Pre-Aztec inter-household variation for all dimensions is
low, before increasing during the Yata phase.
Cultural changes are evaluated for ritual activities and foodways. I evaluate the
degree of interhousehold variability, the overall similarity to other parts of Central
Mexico, the degree of change under Aztec rule, and immigration versus emulation as
i
potential explanations for that change. Evaluation is based on the distinction between
high and low visibility objects and practices. The Dongu and Ninupi phase households at
Calixtlahuaca were culturally homogeneous and regionally distinctive. During the Yata
phase, the site became moderately more Aztec, but this change was unevenly distributed
among households.
Together, the economic and cultural patterns at Calixtlahuaca indicate that the
pre-Aztec local organization of power was relatively collective, but that this was partially
overlaid by relatively indirect and non-collective Aztec imperial rule, with mildly
negative effects.
ii
DEDICATION
To my parents
iii
ACKNOWLEDGMENTS
First, I wish to thank my chair, Mike Smith. From the footnote on his call telling
me that I’d been accepted into the graduate program at ASU, and by the way, would I
want to come work for his new field project that summer, to his comments on the final
draft of my dissertation, Mike has provided me with research opportunities and
challenged me to think about what I’m really saying. I also thank the other members of
my committee – Barbara Stark, Kate Spielmann, and Emily Umberger – for their ongoing
support and diverse opinions.
I also owe a debt to all those who have participated in the Calixtlahuaca
Archaeological Project. This dissertation could be as broad as it is precisely because I
could draw on so many people’s fieldwork and analyses. I thank all of those
archaeologists who worked on the excavations – Alex, Isabel, Julie, Tim, Melissa,
Caitlin, Jeff, Maëlle, Miriam, Marieke, Rikki, and Andrew – all those who rotated
through the lab in subsequent seasons, especially Brad Andrews, and all of the student
volunteers for the project at ASU, especially Kea Warren. I also wish to thank the
inhabitants of San Francisco Calixtlahuaca, especially those who worked for the project
in the field or lab. There is no way we could have completed the artifact analysis without
the help of the project “tepalcateras” – Judith, Janeth, Delfina, Julia, Azucena, and
Graciela.
Logistically, my research at Calixtlahuaca was supported by the National Science
Foundation, both as the grants funding the larger CAP project, and in the form of my
Doctoral Dissertation Improvement Grant. I also received financial support from the
iv
School of Human Evolution and Social Change, and the ASU Graduate and Professional
Students Association. ASU and El Colegio Mexiquense, Zinacantapec, Mexico provided
infrastructure support for my research
I have received ongoing professional encouragement and support from my fellow
archaeologists. I thank Guy Marden and Mary Oman for their support of a high school
student, who had to be told yet again, that “That’s just a rock, not a flake.” At the
University of Oregon, Doug Kennett supported my interest in Mesoamerica. At Arizona
State University, I have been challenged and encouraged by both the faculty and my
fellow graduate students, especially my fellow Calixtlahuaquista Juliana Novic.
In a broader sense, I owe my family a debt for being the sort of crazy folks that
encouraged me to pursue my dreams, however unconventional. These include my parents
and siblings for our many family trips to Mexico and beyond. After I learned how to
hitchhike in Baja when I was ten and climbed the pyramids at Teotihuacan when I was
eleven, they wonder why I became an archaeologist! I thank my Kraemer grandparents
for taking me to museums, and my Huster grandparents for taking me on my first PIT
dig. I owe my boyfriend Jerry for proofreading and unconditional support. Finally, Tavithe-dog get a big thank you for giving me an excuse to get out of the house and go for a
walk on the many occasions when I needed a break from writing.
v
TABLE OF CONTENTS
Page
LIST OF TABLES………………………………………………………………………xiv
LIST OF FIGURES……………………………………………………………………...xx
CHAPTER
1. THE PROBLEM OF SEEING AZTEC IMPERIALISM IN THE
ARCHAEOLOGICAL RECORD………………………………………………...1
Interaction in Postclassic Mesoamerica…………………………………...2
Models of Aztec Imperialism……………………………………………...6
Hegemonic vs. Territorial Control………………………………...7
Degree of Collective Organization………………………………..8
The Site of Calixtlahuaca………………………………………………...10
Research Foci…………………………………………………………….11
Economic Changes……………………………………………….12
Archaeological Markers of Economic Change…………………..15
Cultural Changes…………………………………………………18
Archaeological Markers of Cultural Change…………………….21
Summary of Research Goals……………………………………………..23
2. THE AZTEC EMPIRE………………………………………………………..24
Imperialism………………………………………………………………24
Theoretical Models………………………………………………26
Classifications of Imperial Variation…………………………………….36
Direct and Indirect Rule………………………………………….37
vi
CHAPTER
Page
More and Less Collective Rule…………………………………..42
Economic Effects of Rulership Strategies……………………………….47
Cultural Effects of Rulership Strategies…………………………………51
The Aztec Empire………………………………………………………..59
Imperial and Local Strategies……………………………………60
Previous Studies of Aztec Imperialism…………………………………..62
The Basin of Mexico……………………………………………..62
The Provinces…………………………………………………….66
Case Study Regional and Site Background……………………………...71
3. SITE BACKGROUND AND PROJECT METHODS………………………..77
Geology and Environment……………………………………………….77
Site Description…………………………………………………………..80
Archaeological Work…………………………………………………….84
Field Methods……………………………………………………………87
Chronology………………………………………………………………89
Samples…………………………………………………………………..90
Analyses………………………………………………………………….93
General Ceramic Classification………………………………….94
Attribute Analysis………………………………………………..96
Instrumental Neutron Activation Analysis………………………99
Petrography……………………………………………………..102
Lithic Analyses…………………………………………………103
vii
CHAPTER
Page
Other Analyses………………………………………………….105
Collections and Data Management……………………………………..106
4. “EVERY KIND OF MERCHANDISE SUCH AS MAY BE MET WITH IN
THE LAND”: REGIONAL EXCHANGE SYSTEMS……....………………...107
The Study of Commercial Economies in Mesoamerica………………..109
Patterns of Exchange in Postclassic Central Mexico…………………...111
The Basin of Mexico……………………………………………111
Central Mexico Beyond the Basin……………………………...114
Trade and Empire at Calixtlahuaca……………………………………..115
Hypotheses About Trade at Calixtlahuaca……………………………...116
Ceramic Importation……………………………………………………118
INAA……………………………………………………………122
Petrography………………………...…………………………...131
Type Based Analysis……………………………………………140
Ceramic Exchange: Results…………………………………………….150
Import Quantities and Sources………………………………….155
Inter-Household Variation in Ceramic Exchange………………159
Obsidian Exchange……………………………………………………..162
Sampling Calibration…………………………………………...163
Obsidian Exchange: Results……………………………………………166
Discussion………………………………………………………………172
Conclusions……………………………………………………………..175
viii
CHAPTER
Page
5. “THE GRIDDLE MAKER IS ONE WHO MOISTENS CLAY”: LOCAL
CRAFT PRODUCTION………………………………………………………..178
Empire and Economic Development…………………………………...179
The Economics of the Aztec Empire…………………………………...184
Mesoamerican Craft Production………………………………………..186
Ceramic Vessel Production……………………………………………..188
Direct Evidence for Ceramic Production……………………….188
Indirect Evidence for Ceramic Production……………………..190
Molded Ceramic Items………………………………………………….196
Lithic Production……………………………………………………….199
Evidence for Lithic Production at Calixtlahuaca……………….202
Cloth Production………………………………………………………..209
Cotton Textile Production………………………………………214
Maguey Textile Production……………………………………..217
Discussion………………………………………………………………219
Possible Alternatives……………………………………………223
Conclusions……………………………………………………………..225
6 .“MECÍA TEICUH BOUGHT AN OLD JÍCARA FOR 15 CACAO BEANS”:
MEASURING WEALTH………………………………………………………227
Quality of Life, Wealth, and Status…………………………………….229
Quality of Life………………………………………………….230
Wealth ………………………………………………………….232
ix
CHAPTER
Page
Status……………………………………………………………234
Wealth Variation in Postclassic Mesoamerica………………………….236
Analyses of Wealth at Calixtlahuaca…………………………………...238
Ceramics………………………………………………………..239
Lithics…………………………………………………………..247
Rare Items………………………………………………………251
Summary of Individual Index Results………………………………….254
Master Wealth Index……………………………………………………256
Historical Values of Domestic Goods in Mesoamerica………...258
Price Standardization…………………………………………...263
Calculation of Value Ratios…………………………………….264
Scoring Specific Value Categories……………………………………..266
Ceramics………………………………………………………..266
Lithics…………………………………………………………..276
Ground Stone…………………………………………………...277
Rare Items………………………………………………………278
Master Wealth Index: Results …………………………………………..278
Conclusions……………………………………………………………..283
7. “BY THEMSLEVES THEY CELEBRATED THE FEAST DAY”:
CULTURAL CHANGES IN RITUAL PRACTICES………………………….287
Rulership and Cultural Change…………………………………………288
Aztec Policies Toward Acculturation…………………………………..291
x
CHAPTER
Page
Causes of “Aztec-ization”: Travelling Pots, People or Ideas…………..293
Migration………………………………………………………..294
Local Acculturation…………………………………………….300
Ethnohistoric Evidence for Migration in the Toluca Valley……………302
Differentiating Migration and Emulation………………………………304
Questions for Calixtlahuaca…………………………………………….309
Public and Private Ritual at Calixtlahuaca……………………………...311
Figurines………………………………………………………………..313
Interhousehold Variation in Figurine Use over Time …………..316
Intersite Comparisons of Figurines……………………………..322
Local and Imported Figurines at Calixtlahuaca………………...327
Figurine Analyses: Results……………………………………..331
Censers………………………………………………………………….332
Interhousehold Variation in Censer Use over Time……………334
Intersite Comparisons of Censers………………………………338
Local and Imported Censers at Calixtlahuaca………………….344
Censer Analyses: Results……………………………………….349
Conclusions……………………………………………………………..350
8. “THE PRINCIPAL FOODS OF THESE WERE TAMALES, BEANS”:
CULTURAL CHANGES IN FOODWAYS…………………………………...355
Food and Identity……………………………………………………….357
Mesoamerican Foodways……………………………………………….359
xi
CHAPTER
Page
Questions for Calixtlahuaca…………………………………………….362
Maize Preparation Practices…………………………………………….363
Ground Stone…………………………………………………...367
Cooking Ceramics………………………………………………377
Maize Preparation Practices: Results…………………………...388
Food Serving Practices…………………………………………………391
Serving Vessel Form Variability at Calixtlahuaca……………..392
Regional Variation in Serving Vessel Frequencies…………….395
Serving Vessel Form Analysis: Results………………………...400
Analysis of Decorative Groups…………………………………401
Analysis of Decorative Groups: Results………………………..409
Conclusions……………………………………………………………..410
9. CONCLUSIONS…………………………………………………………….417
Economic Hypotheses and Expectations……………………………….420
Economic Findings……………………………………………………..422
Cultural Hypotheses and Expectations…………………………………427
Cultural Findings……………………………………………………….431
Methodological Implications…………………………………………...435
Implications for Understanding the Aztec Empire……………………..438
Future Research Directions……………………………………………..440
REFERENCES CITED…………………………………………………………………441
APPENDIX
xii
A. CERAMIC TYPE LIST AND CODING SHEET…………………………..493
B. CERAMIC ATTRIBUTE CODES, CODING SHEETS, AND CONTEXTS
SAMPLED……………………………………………………………………...501
C. INSTRUMENTAL NEUTRON ACTIVATION ANAYLIS (INAA) SAMPLE
LISTS AND LAB REPORT……………………………………………………510
D. PETROGRAPHY SAMPLE LIST AND FABRIC DESCRIPTION
REPORT………………………………………………………………………..579
E. LITHIC TECHNOLOGICAL CLASSIFICATION CODES………………..594
xiii
LIST OF TABLES
Table
Page
1.1 Archaeological Artifact Types Associated with Various Craft Production Activities
Potentially Occurring at Calixtlahuaca…………………………………………………..17
3.1 Excavated Household Components in the DS-1 (core household) and DS-2 (extended
household) Samples with Lot Counts, Volume Excavated, and Sherd Counts……….....86
4.1 INAA Ceramic Sourcing Results by Household Component...…………………….124
4.2 INAA Ceramic Sourcing Group to Project Ceramic Type Correlations…………...129
4.3 Ceramic Petrography Results by Household.……………………………………....132
4.4 Ceramic INAA Group to Petrographic Group Correlations………………………..133
4.5 Ceramic Petrographic Sourcing Group to Project Ceramic Type Correlations…….137
4.6 Number of Samples Sourced by Type, with DS-1 Total and Rim Sherd Counts, and
INAA and Petrography Samples Submitted and Successfully Sourced ……………….144
4.7 Ceramic Petrography Results by Macroregion for Each of Six Sampled Household
Components..…………………………………………………………………………...152
4.8 INAA Ceramic Group Frequencies by Household Component, after Weighting by
Vessel Form Frequencies in the Total Original Assemblages to Account for Stratification
by Form in the Original Sample………………………………...………………………153
4.9 Ceramic Type Classification Based Import Frequencies by Household Component
and Source Macroregion with Rim Sherd and Total Sherd Based Values……………..158
4.10 INAA-Based Obsidian Source Percentages for Excavated Phases and Survey at
Calixtlahuaca……………………………………………………………………………163
xiv
Table
Page
4.11 Grey Obsidian Sourcing Results with Sample Results Divided by Phase and Lithic
Production Technology. Includes both Burke and MURR Sample Results..…………..164
4.12 Primary Obsidian Sources by Phase, Showing Correction Procedure Based on
Production Technology Frequencies by Phase to Adjust for Lithic Production Technology
Related Bias in Original Sample Selection……………………………………………..166
4.13 Comparative Total and Green Obsidian Frequencies for Late Classic to Early
Colonial Sites in Central Mexico……………………………………………………….168
4.14 Green and Total Obsidian Quantities by Household Component, with Obsidian
Weights, Counts, and Standardizations by Rim Sherd and Total Sherd Counts...…......170
4.15 Comparative Frequencies of Imported Ceramics at Postclassic Central Mexican
Sites……………………………………………………………………………………..174
5.1 Excavated Contexts at Calixtlahuaca with Overfired Sherds.……………………...190
5.2 INAA Based Frequencies of Local and Imported Ceramics by Macroregion Showing
Adjustment for Vessel Form Frequencies during the Phase……………………………191
5.3 Standardization Measure Values for Ceramic Bowls, Jars, and Hourglass Censers, by
Phase……………………………………………………………………………………195
5.4 Molds for Small Ceramic Items at Calixtlahuaca by Type and Provenience.……...197
5.5 Total and Production Related Lithic Frequencies for Core-Blade, Biface and Bipolar
Lithic Production Artifacts by House and Phase……………..………………………...204
5.6 Cotton and Maguey Spindle Whorl Frequencies at Calixtlahuaca and other
Comparative Sites………………………………………………………………………212
xv
Table
Page
5.7 Cotton (cotton-weight whorls and spinning bowls) and Maguey (maguey-weight
whorls and tabular basalt scrapers) Textile Production Tool Frequencies per 1000 Total
Sherds by Household Component………………………………………………………216
5.8 Summary of Craft Production Artifact Frequencies by Household Component..….222
6.1 Simple Ceramic-Based Wealth Measures (% serving vessels, % decorated ceramics,
Smith index of decorated and imported ceramics) by Household Component………...241
6.2 Ceramic Production Step Value Scoring…………………………………………...244
6.3 Average Production Step Index Values by Household Component………………..246
6.4 Obsidian-Based Wealth Indices (Total obsidian/1000 sherds, green obsidian/1000
sherds, % green obsidian) by Household Component………………………………….249
6.5 Rare Item Counts by Household Component, with Frequencies per 1000 Sherds…252
6.6 Documentary Values for Domestic Items Recovered in Mesoamerican Archaeological
Contexts………………………………………………………………………………...261
6.7 Ceramic Vessel Forms and Prices in the Codex of the Potters of Cuauhtitlan...…...268
6.8 Excavated Ceramic Assemblage Value Calculations, Based on Historic Price Ratios
………...…..…………………………………………………………………………….275
6.9 Master Wealth Index Values for Ceramics, Lithics, Ground Stone, and Rare Items, by
Excavated Component, Based on Historical Price Ratios……………………………...281
6.10 Household Component Level Pearson's R Correlations and Chi-Squared
Significances for the Master Index and its Component Factors ……………………….283
7.1 Total Figurine Frequencies per 1000 Sherds and Subject Matter Percentages of
Identifiable Pieces, by Household Component and Phase at Calixtlahuaca……………321
xvi
Table
Page
7.2 Figurine Frequencies per 1000 Sherds and/or 100 Rims at Calixtlahuaca and
Comparative Postclassic Central Mexican sites……………………..…………………324
7.3 Percentages of Figurine Subject Categories at Calixtlahuaca and Comparative Sites,
with K-means Clustering Results for 2-6 Groups……………………………..………..325
7.4 Figurine Paste Groups by Household Component…………..……………………...329
7.5 Rim and Total Sherd Based Frequencies of Functional Censer Classes by Household
Component, with Summary Statistics by Phase………………………………………..337
7.6 Functional Censer Form Frequencies out of Total Rim Sherds at Calixtlahuaca and
Comparative Sites, with K-means Cluster Results……………………………………..340
7.7 Functional Censer Form Frequencies out of Total Sherds at Calixtlahuaca and
Comparative Sites, with K-means Cluster Results……………………………….…….342
7.8 INAA Group Results with Regional Affiliations) for Censers and Sahumadors, by
Project Ceramic Classification Type…………………………………………………...345
7.9 Petrography Group Results with Regional Affiliations for Censers and Sahumadors,
by project Ceramic Classification Type………………………………………………...345
7.10 Summary of Results of Analyses of Ritual Practices Showing Shifts in
Interhousehold Variation and Local vs. Basin of Mexico Cultural Patterns Under Local
and Aztec Imperial Rule………………………………………………………………..351
8.1 Mano and Metate Counts and Frequencies per 1000 Sherds by Household Component
at Calixtlahuaca…………………………………………………………………………368
8.2 Grinding Stone Frequencies at Calixtlahuaca and Comparative Central Mexican
Sites……………………………………………………………………………………..370
xvii
Table
Page
8.3 Mano Handle Forms by Household Component at Calixtlahuaca………………….373
8.4 Means, Standard Deviations, and Coefficients of Variation for Mano Thickness and
Mano Width by Phase…………………………………………………………………..374
8.5 Mano Cross-Section Shapes by Household Component at Calixtlahuaca………….376
8.6 Comals as Percentages of per Total Sherds and Rim Sherds by Household Component
at Calixtlahuaca, with Summary Statistics by Phase…………………………………...379
8.7 Comal Frequencies as Percentages of Total Sherds and Rim Sherds at Calixtlahuaca
and Comparative Postclassic Sites in Central Mexico……………………..…………...381
8.8 INAA and Petrography Sourcing Results by Geographic Macroregion for Comals
…………………………………………………………………………………………..383
8.9 Interior-Incised Jars as Percentages of Total Sherds and of Plain/Eroded Jar Sherds by
Household Component at Calixtlahuaca……..…………………………………………386
8.10 Relative serving Vessel Form Frequencies by Rims Sherds and Total Sherds, by
Household Component at Calixtlahuaca..………………………………………………393
8.11 Serving Vessel Form Frequencies from Rim Sherd Counts at Calixtlahuaca and
Comparative Sites, with K-means Clustering Results for 1-5 Clusters……………...…397
8.12 Serving Vessel Form Frequencies from Total Sherd Counts at Calixtlahuaca and
Comparative Sites, with K-means Clustering Results …………………………………400
8.13 Major Ceramic Decorative Groups and Associated Vessel Forms at Calixtlahuaca
……………..……………………………………………………………………………403
8.14 Ceramic Decorative Group Frequencies by Household Component, with K-Means
Clustering Results and Summary Statistics by Phase…………………………………..408
xviii
Table
Page
8.15 Summary of Results of Analyses of Foodways Showing Shifts in Interhousehold
Variation and Local vs. Basin of Mexico Cultural Patterns Under Local and Aztec
Imperial Rule…………………………………………………………………………...411
xix
LIST OF FIGURES
Figure
Page
1.1 Map Showing the Location of Calixtlahuaca within Central Mexico Relative to the
Aztec Capital of Tenochtitlan and General Regional Topography………………………..3
1.2 Postclassic Chronologies at Calixtlahuaca and Comparable Central Mexican Sites.....6
2.1 The Maximum Extent of the Aztec Empire, showing the Imperial Core, Territorial,
and Strategic Provinces……………………………………………………………….….60
2.2 Comparative Postclassic Sites in Central Mexico used for Analysis in this
Dissertation………………………………………………………………………………63
2.3 Comparative Postclassic Sites in Western Mesoamerica with Prior Archaeological
Work Discussed in this Dissertation, with the Territories of the Aztec Empire shown…67
3.1 Map of the Site of Calixtlahuaca showing Site Boundaries, Monumental Architectural
Groups, and Excavation Unit Locations…………………………………………………79
3.2 The Round Pyramid (Group C) at Calixtlahuaca…………………………………….82
3.3 Monumental Group D at Calixtlahuaca……………………………………………...83
3.4 Typical House at Calixtlahuaca (Excavation Unit 309)……………………………..83
3.5 Schematic of the DS-1 (Core Household Sample) to DS-5 (all Excavated Contexts)
Samples of Contexts Excavated by the Calixtlahuaca Archaeological Project.................91
4.1 Geographic Source Macroregions for Ceramics Excavated at Calixtlahuaca...........119
4.2 Selected Ceramic Types Associated with the Basin of Mexico Source Macrogroup,
Based on both INAA and Petrographic Analyses………………………………………121
4.3 Selected Ceramic Types Associated with the S/SW State of Mexico Source
Macrogroup, Based on both INAA and Petrographic Analyses………………………..122
xx
Figure
Page
4.4 Summary of Imported Ceramics by Macroregion by Phase, Based on INAA (A),
Petrography (B), Types, Rim Sherds Only (C), and Types, all Sherds (D)…………….151
4.5 INAA Ceramic Imports by Household Component. Values from Table 4.6…….…160
4.6 Rim Sherd Type-Based Ceramic Imports by Household Component….…………..160
4.7 Obsidian Source Frequencies per 1000 Sherds by Phase…………………………..171
5.1 Examples of Figurine Molds and Spindle Whorl Molds Excavated at
Calixtlahuaca……………………………………………………………………………198
5.2 Histogram of Spindle Whorl Frequencies per 1000 Sherds at Postclassic and Early
Colonial Sites…………………………………………………………………………...213
5.3 Decorated and Plain Cotton Spindle Whorls from Calixtlahuaca, showing the Range
of Variation in Shapes………………….……………………………………………….215
5.4 Examples of Maguey Spindle Whorls from Calixtlahuaca…...…………………….218
5.5 Tabular Basalt Maguey Scrapers (“Desfibradores”) from Calixtlahuaca...………...219
5.6 Relationship Between Component Size (sherd count) and the Number of Craft
Activities Present…………………………………………..............................………...220
5.7 Temporal Trends in Craft Production at Calixtlahuaca, by Craft….……………….221
6.1 Summary of Selected Wealth Indices by Phase…………………………………….255
6.2 Summary of Highest and Lowest Scoring Components for Each Phase for Selected
Wealth Indices…...……………………………………………………………………..256
6.3 Vessel Forms Depicted in the Codex of the Potters of Cuauhtitlan…..……………271
6.4 Master Wealth Index Results by Phase, with Contribution of Each Artifact Class...280
xxi
Figure
Page
7.1 Expected Patterns of High- And-Low Visibility Items in Cases of Migration and
Local Emulation………………………………………………………………………...306
7.2 Selected Figurines from Calixtlahuaca….……………….…………………………315
7.3 Hierarchical Clustering Results for Figurines for Calixtlahuaca by Phase and
Comparative Sites………………………………………………………………………327
7.4 Figurine Frequencies per 1000 Sherds by Household Component, Subdivided by
Figurine Source Group………………………………………………………………….330
7.5 Censer and Sahumador Forms……………………………………………………...333
7.6 Hierarchical Clustering Results for Functional Censer Forms at Calixtlahuaca and
Comparative Sites, Based on Percentages of Rim Sherds………………………...……341
7.7 Hierarchical Clustering Results for Functional Censer Forms at Calixtlahuaca and
Comparative Sites, Based on Percentages of Total Sherd Counts……………………...343
7.8 Frequency of Texcoco Molded-Filleted Sahumadors at Calixtlahuaca as a Percentage
of Total Sherds, by Household Component…………………………………………….348
8.1 Stepwise Processing Decisions for Five Common Maize Foods…………………...365
8.2 Mano Handle and Cross-Section Forms…………….……………………………...372
8.3 Graph of Comals as a Percentage of all Sherds, by Household Component……….378
8.4 Interior-Incised Jar Sherds………………………………………………………….384
8.5 Scatterplot of Comal and Interior-Incised Jar Sherds, as Percentages of all Vessel
Sherds……………………………………………………………………..…………….388
8.6 Rim Sherd Based Hierarchical Clustering Results for Serving Vessel Forms at
Calixtlahuaca and Comparative Sites …………………………………………………..398
xxii
Figure
Page
8.7 Total Sherd Count Based Hierarchical Clustering Results for Serving Vessel Forms at
Calixtlahuaca and Comparative Sites …………………………………………………..399
8.8 Primarily Local Ceramic Decorative Families……………………………………..404
8.9 Primarily Non-Local Ceramic Decorative Families………………………………..405
8.10 Hierarchical Clustering Results for Ceramic Decorative Family Frequencies……406
xxiii
CHAPTER 1
CHAPTER 1. THE PROBLEM OF SEEING AZTEC IMPERIALISM IN THE
ARCHAEOLOGICAL RECORD
The ethnohistoric documentation of the Aztec Empire provides a rich tapestry of
alliances between cities, military victories, treacherous nobles and wise advisors.
Pointedly absent from this picture is the commoner majority of the population, especially
outside of the Basin of Mexico. In an effort to provide a window on this silent majority, I
ask whether political incorporation into the empire resulted in changes to the everyday
lives of commoners living in provincial areas, using the Postclassic (AD 1130-1530) site
of Calixtlahuaca, Mexico as a case study (Figure 1.1). Berdan and Smith (1996:216-217)
identify the effects of Aztec imperialism on commoners as one of the least understood
aspects of the Aztec Empire.
This project focuses on changes in economic interaction and cultural practices,
two domains where previous work (Berdan, et al. 1996; Brumfiel 1998; Overholtzer
2012; Smith 2003c) suggested that Aztec processes of consolidation of control and
ongoing rulership might have affected commoners as well as elites. In order to
distinguish the results of imperial rulership strategies from more generalized regional
cultural change, I compare three time periods (prior to the establishment of the empire,
during the initial imperial formation and expansion into other areas, and after the Aztec
conquest of the study site itself). I assess economic interaction by looking at regional
patterns of trade, local craft production, and household wealth. My examination of
cultural change focuses on foodways and domestic ritual practices. I use these to evaluate
general trajectories of cultural change at the site, as well as the degree of interhousehold
1
variation over time. These analyses are used to compare scenarios for economic and
cultural interaction based on two popular dimensions for interpreting imperial/provincial
interaction: direct vs. indirect rule, and the degree of collectiveness of the state. Prior
considerations of these two dimensions of rule have produced very different
interpretations of the Aztec state, with descriptions (Hassig 1985, 1988; Smith and
Berdan 1996b) of the Aztec Empire as an indirect-rule empire with minor negative
effects on conquered areas contrasting with those that portray it as a relatively collective
state where rulers were accountable to their subjects (Blanton and Fargher 2008; Fargher
and Blanton 2007; Fargher, et al. 2011).
Interaction in Postclassic Mesoamerica
The Postclassic period (AD 900-1521) in Mesoamerica saw two major processes.
The first was a steadily increasing degree of cultural interaction and economic integration
across wide geographic regions, throughout the period. The second was the foundation
and expansion of the Aztec Empire, beginning in the AD 1428 and continuing through
the arrival of the Spanish in 1519. One of the major issues in investigating Aztec
imperialism is differentiating the effects of these two processes, that is, the effect of
Aztec rule from broader background processes of cultural and economic change.
2
Figure 1.1 Map showing the location of Calixtlahuaca within Central Mexico relative to
the Aztec capital of Tenochtitlan and general regional topography.
The first process, of increasing cultural interaction and economic integration, can
be glossed as the development of the Postclassic Mesoamerican world-system. This
process began in the Early Postclassic as Mesoamerican peoples sought to reformulate
networks interrupted by the political collapses of states and their associated economic
systems at the end of the Classic period. It resulted in widely shared cultural practices and
stylistic traits across much of Mesoamerica (Boone and Smith 2003; Ringle, et al. 1998;
Smith and Heath-Smith 1980). It is also closely related to the development of a wellintegrated system of market exchange for both local and long-distance trade. However, at
3
smaller geographic scales, there is substantial variation in the degree to which individual
sites and regions participated in the broader world-system.
The second process, the expansion of the Aztec Empire, took place against the
preexisting backdrop of Postclassic interaction across Mesoamerica, including a city-state
form of political organization, shared elite symbols and culture, and a well-integrated
market system. On a more geographically specific level, the Aztec Empire also followed
other attempts at state-building in the Basin of Mexico, most immediately the Acolhua
and Tepaneca states. As a result, many shared traits used as markers of the Aztec Empire
actually predate its establishment (Sergheraert 2009). The Aztec Empire, or Triple
Alliance1, was founded as an alliance among the Basin of Mexico city-states of
Tenochtitlan, Texcoco, and Tlacopan in AD 1428. The alliance promptly conquered the
remainder of the Basin of Mexico and began expanding outward. By the time the Spanish
arrived in 1519, Tenochtitlan had come to dominate the alliance, which then controlled
territory from one coast of Mesoamerica to the other. The empire was funded by a
regularized system of taxes, paid by conquered provinces, and managed by a combination
of local and imperial officials (Smith 2015a).
These two processes result in issues of equifinality when interpreting cultural or
economic changes during the latter portion of the Postclassic period. Such issues can be
1
There was never a group of people who called themselves the Aztecs nor a political unit that called itself
the Aztec Empire. I follow the regional archaeological and art historical convention of using the term Aztec
to refer to the occupants of Middle and Late Postclassic (1100-1521 CE) Central Mexico and use Triple
Alliance and Aztec Empire interchangeably to refer to the expansionist political unit created by the AD
1428 alliance of the city-states of Tlacopan, Tenochtitlan, and Texcoco. Thus, there were Aztecs before the
founding of the Aztec Empire, and not all members of the empire were Aztecs.
4
minimized by the investigation of sites that meet two criteria: material culture distinct
from that of the Basin of Mexico, and a sufficient rate of cultural change over time to
allow for the development of fine-grained archaeological chronologies. First, if there is a
reasonable degree of preexisting difference between local material culture and that of the
Basin of Mexico, it makes it easier to distinguish between the local adoption of foreign
practices and similarities resulting from parallel development from the same starting
point, such as an earlier immigrant population.
Second, many regions in Mesoamerica have Postclassic chronologies consisting
of 100-200 year phases. Given that the Aztec Empire only existed for around 100 years,
and that in later-conquered provincial areas, the period under Aztec rule may be
substantially shorter than this, chronologies intended to address questions of Aztec rule
need to be able to differentiate 50-70 year phases (Figure 1.2). When this can be done, it
allows for the identification of pre-Aztec trajectories, establishing a baseline and the
direction of change in interregional interaction over time. This baseline can then be
compared to the changes observed under Aztec rule. Unfortunately, cultural differences
and the potential for good chronological control vary inversely with each other. Sites
closer to the Basin of Mexico generally have less cultural differences from the Aztec
heartland, but they were also under Aztec rule for a longer period of time, making it more
likely that a distinct chronological phase for the period under Aztec rule can be identified
but less likely that Aztec material will be distinctive from local traditions. The reverse is
true of sites closer to the edges of the empire.
5
Regional
Period
1550
Toluca Valley
History
Calixtlahuaca Yautepec
F: Spanish
Colonial Period
E: Mexica reign
over T.V.
D: Early Mexica
Late Aztec/
empire
Late
C: Reign of
Postclassic
Tezozomoc
Colonial
1500
1450
1400
Santiago
Yata
(phase 6)
Xaltocan
Isla
Molotla
Tlalli
Ninupi
(phase 4)
Atlan
1350
B: Early Tepanec
empire
1300
1250
1200
Early
Aztec/
Middle
Postclassic
Hai
Dongu
(phase 2)
Pochtla
A: Legendary
period
Dehe
1150
Toltec/
Early
1100 Postclassic
Sanders et Tomaszewski and Huster and
Source
Smith 2015
al. 1979
Smith 2011
Hare and
Overholtzer
Smith 1996 2014
Figure 1.2 Postclassic chronologies at Calixtlahuaca and comparable Central Mexican
sites
Models of Aztec Imperialism
The effects of Aztec imperialism can be best understood in a broader context of
research on the imperialism. Comparative research on empires has identified multiple
6
dimensions of variation in the organization of empires, related to different dimensions of
imperial strategies and imperial/provincial relations. I present a brief introduction to two
models applied to the Aztec Empire here, and these and other related dimensions of
variation are discussed in more detail in the following chapter. The specific implications
of each dimension of variation, for the household level, are discussed later in this chapter,
under economic and cultural scenarios.
Hegemonic vs. Territorial Control
Imperial control may be characterized on a continuum between hegemonic (or
indirect) and territorial (or direct) rule (Hassig 1985; Luttwak 1976). This continuum
applies both within empires, where inner provinces are more likely to be directly
controlled, and among empires, where some states are more territorial or hegemonic
overall. Indirect rule is characterized by rule via client states, where an empire leaves
most of the local power structure and administrative machinery in place. Compliance
with imperial policies is generally enforced through the cooption of local elites and the
threat of military force. By placing the cost of administering a province on local
government, indirect rule provides an inexpensive means of controlling territories, but
one that is subject to regular revolts and one that is limited in how much it can extract
from provincial areas. Indirect rule is likely to produce relatively few cultural changes in
provincial areas and have a neutral to somewhat negative economic effect. Proposed
examples of relatively hegemonic states include first century Rome (Luttwak 1976) and
the Aztec Empire (Hassig 1985, 1988).
7
In contrast, under direct rule, prior local rulers and administrative structures are
replaced by those managed by the empire. Compliance is enforced through a much
greater penetration of imperial bureaucracy into multiple aspects of society, though
military force remains an option for enforcing compliance. This form of rulership is much
costlier for the imperial state, but produces more stable control over provincial areas and
permits a higher overall possible degree of exploitation. It is likely to produce a higher
degree of cultural change at all levels of provincial society. The economic effects of
direct rule are also more pronounced, but may be either positive or negative depending on
the balance between new opportunities for economic development and new taxes.
Degree of Collective Organization
An empire may also be characterized by how collective it is, ranging from noncollective (or network-oriented) to highly collective (or corporate-oriented) states
(Blanton and Fargher 2008; Blanton, et al. 1996; Fargher and Blanton 2007). This
dimension describes the social organization of power within the empire. In relatively
non-collective states, power is based on exclusivity. Elite status, and the right to rule, is
confirmed by the exclusive control of particular classes of goods, especially non-local
goods. Rulers are not broadly obligated to provide public goods to the population, and
those goods they do provide are used to create personal relationships of clientage. There
are few checks on what rulers are permitted to do (up to arbitrarily putting people to
death), but rulers also have relatively little ability to motivate their populace to provide
high levels of taxes to the state. States based on this organizational form are generally
funded by external revenue sources, such as external trade in luxury goods. Most
8
Southeast Asian states, such as Vijayangara, are relatively non-collective (Blanton and
Fargher 2008), and within Mesoamerica, the Classic-period Maya characterize this end of
the spectrum (Blanton, et al. 1996).
In contrast, in relatively collective states, power is based on social integration.
Elites rule based on broad popular support and the ability to successfully provide public
goods to the population. Authority is derived from offices rather than personal authority.
Rulers are accountable to their subject populations, both for the services that they are
culturally expected to provide and for the limits to their authority. Because of the need to
provide public goods, more collective states generally have larger, more complex
bureaucracies. These also allow such states to depend more heavily on internal sources of
revenue, such as taxes on staple crops. More collective states can generally leverage a
higher level of taxation from their populations, but are also expected to provide a higher
level of services in return. The Inka Empire is an example of a relatively collective state
on a cross cultural scale (Blanton and Fargher 2008). Within Mesoamerica, Teotihuacan
is considered to have had more collective rulership (Blanton, et al. 1996).
In a broad sense, these two dimensions of rulership – directness and
collectiveness - should co-vary. More indirect strategies depend more heavily on drawing
local elites into imperial networks, while more direct rule would require a higher degree
of commoner participation, likely requiring somewhat more collective strategies.
Similarly, both direct rule and collective rule usually rely on internal financing, which
should produce a correlation between these two dimensions of imperial variation. As a
result, the differences between Hassig’s (1985, 1988) and Blanton and Fargher’s (Blanton
and Fargher 2008; Fargher and Blanton 2007) conclusions about the nature of the Aztec
9
Empire require further exploration. The two dimensions of imperialism provide a
theoretical framework for predictions about the economic and cultural effects that
different forms of rule are likely to produce.
The Site of Calixtlahuaca
The study site of Calixtlahuaca is located in a geographically intermediate region,
where the requirements for both cultural differences and temporal control can be met. As
a regional capital with a documented Aztec administrative presence, the site provides a
good location for looking at the effects of Aztec rule. Calixtlahuaca is a large city-state
capital located in the Toluca Valley of Central Highland Mexico, occupied between AD
1130 and 1530 (Huster and Smith 2015). The Toluca Valley (sometime called the Upper
Lerma) is located immediately to the west of the Basin of Mexico core of the Aztec
Empire, which places it geographically within the inner provinces of the empire.
However, the Toluca Valley diverges culturally from much of the rest of Central
Highland Mexico during the Early Postclassic. Among other distinctive traits, the region
maintains a predominately red-on-buff decorated pottery assemblage, almost no comals
among the plainwares, a heavy reliance on West Mexican obsidian sources, hilltop or
hillside locations for major cities, and cemetery burials in plaza settings throughout the
Postclassic, all of which differ from the cultural practices of the Basin of Mexico by the
Middle Postclassic.
This degree of cultural difference is, however, accompanied by a relatively long
span of time under Aztec rule, due to the region’s proximity to the Basin of Mexico.
10
Based on ethnohistoric sources, the Toluca Valley was conquered by Axayacatl by the
mid-1470s, though the region also appears in among the conquests of the subsequent
ruler of Tenochtitlan, Tizoc, suggesting some degree of reconquest in the early 1480s
(Hernández Rodríguez 1998; Tomaszewski and Smith 2011). These dates indicate that
the area spent at least 40 years under Aztec rule prior to the arrival of the Spanish in
1519. Archaeological work at this site differentiates three time periods, the Dongu Phase
(AD 1130-1380), Ninupi Phase (1380-1450) and the Yata Phase (1450-1530) (Huster and
Smith 2015). While not exact matches to the historical chronology, these phases provide
reasonable proxies for the periods prior to state formation in the Basin of Mexico, the
periods of the Tepaneca state and the early Triple Alliance when Calixtlahuaca was still
independent, and the period during which Calixtlahuaca was under Aztec rule (Figure
1.2).
Recent work at the site by the Calixtlahuaca Archaeological Project, directed by
M. E. Smith, excavated a wide range of household and terrace contexts at the site. I use
all of the securely associated material from a single household excavation dating to a
single phase as my primary unit of analysis and refer to these units as household
components. There are six household components dating to each phase of the site’s
occupation.
Research Foci
Beyond identifying sites where it is possible to investigate Aztec imperialism, it is
also necessary to identify aspects of commoner life where Aztec rule is likely to have had
11
an influence under different forms of imperialism. Various authors have identified a
general trio of power sources in imperial rule: economic sources, control over knowledge,
and military force (Goldstone and Haldon 2009:4; Yoffee 2005:38-40). As military force
is unlikely to be visible at the household level, I focus on economic and cultural changes.
For both domains, variation in the directness and collectiveness of rule are likely to have
produced differing effects on commoner households. Theoretically based predictions
about economic and cultural change under different forms of rulership can be adjusted for
local conditions based on a combination of the ethnohistoric information available for
Calixtlahuaca and the existing archaeological work both in and outside of the Basin of
Mexico.
Economic Changes
Research on the effects of the growth of the Aztec Empire on conquered citystates suggests a continuum of postconquest economic effects on trade, local production,
and quality of life (Brumfiel 1980; Nichols, et al. 2002; Nichols, et al. 2009; Smith
2003c; Smith and Heath-Smith 1993). Mostly indirect rule and/or network-oriented rule
should produce relatively few changes in regional economic patterns compared to the
period prior to imperial rule. More direct rule and/or collective rule should produce more
noticeable economic changes, generally oriented in more positive directions due to the
potential for economic development among both commoners and elites, due to a
consistent and reliable provisioning of public goods. Each of these theoretical outcomes
is the product of more complex balances among the imperial tax load, the opportunities
for trade provided by the Aztec Empire, and the local responses to these influences. I
12
present two scenarios below, based on relatively indirect, non-collective rule, and on
relatively direct, collective rule. As the rulership variables are continua, intermediate
positions are both possible and probable.
Economic Scenario 1. In this scenario, Aztec rule was relatively indirect, with a
network-oriented rulership strategy focused on co-opting local elites. Under these
conditions, I expect the conquest by the Aztec Empire to have had little effect on existing
economic patterns, as is more commonly seen in the outlying provinces of the empire. I
expect to see the continuation or intensification of pre-conquest patterns of local craft
production and market exchange, and a relatively low frequency of artifacts from the
Basin of Mexico. Overall wealth should remain steady or follow pre-Aztec trajectories.
This pattern is characteristic of sites in strategic provinces or otherwise near the edges of
the empire that had relatively little pre-imperial contact with the Basin of Mexico (i.e.,
Kowalewski, et al. 2010; Venter 2012). Calixtlahuaca might follow this pattern because,
while part of a tributary rather than strategic province, the Toluca Valley still formed part
of the buffer zone between the Aztec and Tarascan empires. In addition, the preliminary
ceramic analysis from Calixtlahuaca suggests an almost complete lack of interaction with
the Basin of Mexico prior to the Late Postclassic, which might have slowed the region’s
integration into Basin-centric market networks after its incorporation into the Aztec
Empire. The presence of relatively indirect, network-based rule by the Aztec Empire is
compatible with a range of degrees of collectivity of local, pre-Aztec governance. I
consider this the more likely of the two economic scenarios presented here.
13
Economic Scenario 2. In this scenario, Aztec rule was more direct, with a more
collective rulership strategy that worked to incorporate both commoners and elites
directly into the imperial system. Under these conditions, I would expect to see more
extensive changes in the local economic system under Aztec rule, as is seen at sites in the
Basin of Mexico. I expect to see a reduction in either the diversity or total quantity of
local craft production and an increased volume of trade with the imperial capital of
Tenochtitlan and/or the Basin of Mexico at the expense of trade with other foreign
regions. This may be described as a change in market systems from a primarily
interlocking central-place system, with smaller centers evenly distributed between
multiple next-order centers and accessing multiple markets, to a primarily dendritic
central-place system, with each center trading only at single higher-order center, and the
highest regional center heavily tied to a geographically distant primate center (Smith
1976). Overall wealth should decrease. Examples toward this end of the continuum
include Morelos (Smith 2010) and portions of the Basin of Mexico (Nichols, et al. 2002).
Calixtlahuaca may show this pattern due to its geographic proximity to the Basin of
Mexico, the post-conquest rearrangement of local power known from the codices
(Chimalpahin 1965 [1606-1631]:105), and the importance of the region in supplying
maize to the Basin of Mexico (Paso y Tronasco 1905-1906:7[2]:6-7). This pattern is more
more likely to have been successful when local governance was already relatively
collective, allowing for the Aztec cooption of preexisting systems.
14
Archaeological Markers of Economic Change
My assessment of these economic scenarios has three components; for each
chronological phase I determined the quantity and diversity of foreign exchange, whether
local craft production intensified or diminished relative to the other phases under
consideration, and whether household wealth increased or decreased.
Analysis of Trade. I address long distance economic connections through three
analyses of ceramics and one of lithics (Chapter 4). Ceramic exchange was evaluated
using INAA (Instrumental Neutron Activation Analysis), petrography, and type
classification, while lithic exchange was evaluated based on XRF (X-Ray Fluorescence)
sourcing. These methods provided an overview of the number of trading regions with
which Calixtlahuaca had connections during each phase, the volume of foreign goods
arriving at the site, and their distribution among households. These show how well
integrated Calixtlahuaca was into broader regional exchange networks over time. The
comparisons among multiple methods of ceramic analysis and multiple artifact classes
provide robust support for the trends that I identify.
First, a stratified random sample of thirty sherds, including both plain and
decorated types, from each household component at the site was chemically characterized
using INAA. As a random sample, the primary use of these sherds was to provide an
independent measure of ceramic variation within and between households, based on the
distribution of source groups. Increased Aztec influence is expected to appear as an
increased frequency of ceramics from Basin of Mexico source groups, and a reduction in
15
the number and/or frequency of other non-local sources as Basin of Mexico ceramics
became more widely available, more socially prestigious, and/or less expensive.
Second, a similar sample of twenty sherds per household was taken from two
household components for each phase for petrography. This sample was used as an
independent check on the trade volume, directionality, and interhousehold diversity
patterns in the INAA sample. In addition, 7-10 sherds from each INAA group were also
submitted for petrography to allow for the direct comparison of the source groups
produced by the two methods.
Third, I used the associations between types and source groups in the preceding
two analyses to assign general geographic associations to all of the basic type-classified
ceramics from all household components. This allowed for a much larger sample size
than the initial samples studied through materials analyses, and an associated
identification of trade that occurred in very low frequencies. All three methods of ceramic
analysis show a high degree of agreement in the directionality and volume of trade over
time.
Lithic exchange was evaluated at the phase, rather than individual household,
level. The lithic assemblage at the site consists primarily of obsidian, which was visually
sorted into grey and green groups in the field. XRF was used to determine the
contribution of various sources to the grey fraction for each phase, and to confirm that the
green fraction came from the Pachuca source.
Identification of Local Craft Production. The intensity and inter-household
variability of local craft production at Calixtlahuaca was assessed directly based on
16
production related artifact frequencies for maguey and cotton textiles, and bifacial, coreblade, and bipolar lithic production (Chapter 5). I assessed ceramic vessel production
primarily indirectly, based on the previously discussed INAA and petrography samples
(see Table 1.1 for artifact types associated with each craft.) Overall, local ceramic
production is measured as the percentage of the INAA samples falling within the local
source groups. In a well-integrated market system, local specialization is expected to
increase, due to an increased reliability of supply of basic necessities from other sources.
As a result, changes in the organization of local craft production can be used as a
secondary line of evidence for participation in regional economic systems over time.
Cloth production may be an exception to general trends of craft production, due to its use
for tax payments.
Craft
Ceramic vessel production
Archaeological Indicators
Waster sherds, concentrations of one type
of sherd
Source
Rice 1987, Stark &
Garraty 2004
Lithic production (Bifacial,
Core-blade, Bipolar)
Lithic artifact types associated with the
production sequence for each technology
Hirth 2006a
Maguey fiber
Large (>10g) spindle whorls, tabular basalt
scrapers
Parsons & Parsons 1990
Cotton fiber
Small (<10g) spindle whorls, spinning bowls Parsons 1972, Smith &
Hirth 1988
Table 1.1 Archaeological artifact types associated with various craft production
activities potentially occurring at Calixtlahuaca
Evaluation of Household Wealth. Household wealth serves as a means of
evaluating the overall effects of economic changes on commoner quality of life.
Household wealth was evaluated for each household component using a variety of single17
artifact class measures, followed by a combined index that incorporated multiple artifact
classes (Chapter 6). Single-class measures of wealth included multiple measures of
ceramic wealth, such as bowl/jar ratios, the percentage of decorated ceramics, bowl and
censer measures used at other Aztec sites (Smith 2006c), and production-step based
values. Other measures include the percentage of green obsidian, the frequency of
obsidian per 1000 sherds, and the frequency of bronze and jewelry items. These singleartifact measures tended to correlate poorly with each other, leading me to develop a
combined index. In the combined index, I assigned relative weights to the values of four
common artifact classes (ceramics of multiple sizes, lithics, ground stone, and
jewelry/bronze) based on values given in Colonial-period Nahuatl documents. This
allows for a single evaluation of overall household wealth over time, despite shifts in the
availability of goods from particular regions over time, which influenced particular single
artifact-class measures.
Cultural Changes
The degree and manner in which cultural practices change under imperial rule are
also affected by forms of rulership. Relatively indirect and non-collective forms of
rulership should have relatively little direct effect on the practices of commoner
households, though commoners may choose to adopt such practices for other reasons.
Relatively direct and relatively collective forms of rulership should both produce a higher
degree of cultural change due to increased contact between imperial and local people.
The effect of Aztec conquest on cultural practices within commoner households is
a topic on which archaeology can provide significant insights. Documentary sources
18
suggest that the Aztec empire promoted the use of shared elite material culture, practices,
and values, but did not similarly target commoners (Berdan, et al. 1996), raising
significant questions about the causes of cultural change following incorporation into the
empire. At the same time, the Aztec Empire did promote stereotypes of particular ethnic
groups and their practices (Berdan 2008). Previous research, based on domestic ritual
assemblages, suggests that state ideology did not have a significant influence on
commoner households (Brumfiel 1998; Klein and Victoria Lona 2009). Several authors,
however, have proposed that changes in food preparation and consumption practices in
Postclassic Mesoamerica may be indicative of broader social changes, presenting an
additional line of evidence that can be used to address imperial conquest related cultural
change (Biskowski 2000; Blanton, et al. 1993). In addition to looking at evidence for the
directness and collectiveness of Aztec rule, I also evaluate local emulation as opposed to
immigration of settlers from the Basin of Mexico as explanations for any observed
cultural changes. Based on Clark (2001), these patterns can be differentiated by the
presence or absence of foreign material culture and foreign style practices in low and
high visibility contexts.
Cultural Scenario 1. Under conditions of relatively indirect and/or non-collective
rule, commoners would not have been actively targeted by Aztec ideology, nor the direct
focus of most imperial policies. As a result, there would have been little to no official
pressure to adopt Aztec-style material culture or practices. In this scenario, Aztec-style
artifacts would be expected to appear occasionally in high-visibility contexts, as
references to one among many distant connections. Aztec-style practices, which require
19
more interaction to learn, should be mostly absent. Neutrality can be contrasted with an
active resistance to Aztec practices, which would be visible as the complete absence of
Aztec style artifacts in certain households when the household’s wealth level and the
overall site frequency for the phase suggest that they should be present.
Cultural Scenario 2. Under conditions of more direct and/or collective rule,
commoners would have had a higher degree of day-to-day interaction with individuals
using Aztec goods and/or practices, and a higher degree of participation in imperiallysponsored events, such as feasts or religious ceremonies. As a result, such forms of
rulership would be likely to produce a higher degree of cultural integration in provincial
areas, even if commoners were not expressly targeted by imperial ideology. Commoners
may have adopted aspects of Basin of Mexico style material culture and social practice in
an effort express a socially advantageous identity, either vis-a-vis local elites, Mexica
migrants, or Aztec officials. This would be visible archaeologically as an appearance or
low-to-moderate increase in the Basin of Mexico style artifacts and/or practices in
household contexts, primarily in high-visibility settings. These proposed changes may be
contrasted with a “site unit intrusion” pattern characteristic of direct migration from the
Basin of Mexico, where a wide variety of forms of material culture, including low
visibility items, replicate those used in the homeland (e.g. Silverstein 2001).
Cultural Scenario 3. In addition to measuring the effects of rule on local people,
governmental strategies can also be examined based on their effects on their broader
populations. More direct rule and/or collective rule is more likely to see the state20
sponsored migration of people between areas, due to these rulership strategies’ greater
penetration into the daily lives of their subjects. There is reasonable ethnohistoric
evidence for the immigration of people from the Basin of Mexico into the Toluca Valley
in general (Carrasco 1950:277-279; Cuauhtitlán 1985 [1606-1631]:57; Zorita 1963
[1566-1585]:22,263ff, 266), though it is unknown whether Calixtlahuaca was one of the
destination settlements for these immigrants. As a result, it is possible that the appearance
of Aztec style objects or practices at the site could be the result of immigration from the
Basin of Mexico. I expect immigrants to appear as households with Aztec-style practices
in low visibility contexts. Depending on the degree to which emphasizing Aztec identity
was socially beneficial, these households may also have higher frequencies of Aztec-style
objects and/or Aztec practices in high visibility contexts.
Archaeological Markers of Cultural Change
More specifically, I looked for cultural changes supporting these three scenarios
in two aspects of household life, domestic ritual (Chapter 7) and culinary practice
(Chapter 8). Carballo (Carballo 2015) identifies both feasting and ritual as strategies of
collective rulership for gaining popular support in Mesoamerica. These two aspects could
have served as arenas for the negotiation of local commoner identities relative to a variety
of others using Aztec-style items, including Aztec elites, local elites, and Aztec
commoner immigrants. For both ritual and culinary assemblages, local and Aztec style
items can be easily distinguished, though the possibility of locally produced versions of
Aztec style items muddies the relationship between style and geographic source.
21
Domestic Ritual Assemblages. The domestic ritual assemblages at Calixtlahuaca
consist of a variety of items, but I focused my analyses on figurines, a relatively low
visibility item, and censers, a relatively high visibility item. In both cases, there are
distinctively local style items, distinctively Aztec style items, and some more widely
distributed Central Mexican styles found in both regions. Censers were included in the
INAA and petrography samples and these source attributions were used to confirm
stylistic attributions of source region. Figurines (n=369) were classified by source and
subject matter, but due to INAH regulations could not be exported for materials analysis
to confirm geographic attributions. Figurines and censers were evaluated both on the
basis of probable origin, and on the basis of probable use, based on subject matter for
figurines and general form for censers.
Foodways. Following the Aztec conquest, the adoption of Mexica methods of
food preparation may have provided an important means of social negotiation, especially
in cases of state-sponsored feasting. I compared evidence for various forms of maize
preparation, a low-visibility activity, with serving vessel form frequencies and ceramic
decorative group frequencies, both of which are higher visibility. Maize preparation
techniques were assessed using variation in mano form, as well as the frequencies of
comals and interior-incised jars which likely represent alternative cooking methods. Food
serving practices were assessed using the frequencies of three categories of serving
vessels – open (bowl/plates/basins), closed (jars/pitchers), and copas. In addition, I
calculated the relative frequencies of major families of decorated ceramics for each
22
household to determine whether there were groupings based on factors other than
chronological phase.
Summary of Research Goals
I seek to contribute to the understanding of the dynamics of Aztec rulership
strategies by looking at commoner households. In looking at commoner households,
rather than elite contexts, I can examine the degree to which Aztec rulership actually
affected the majority of the population. I use a combination of analyses of economic and
cultural change in order to examine the interplay between multiple arenas of interaction.
23
CHAPTER 2
CHAPTER 2. THE AZTEC EMPIRE
The Aztec conquest of Calixtlahuaca did not occur in a historical vacuum, and
neither does this study. This chapter provides a more in-depth background for the
thematic and regional topics covered in this dissertation. It covers current scholarly
perspectives on imperialism and how these relate to the dimensions of rulership under
consideration, an overview of the Aztec Empire, a review of previous case studies of
Aztec imperialism, and an ethnohistorical and archaeological overview of the Toluca
Valley in general and Calixtlahuaca in particular.
Imperialism
This section describes three aspects of empires. First, I present three frameworks
commonly used to discuss imperialism: world-systems analysis, agency based
approaches, and social organization of power models. These three frameworks provide
complementary theoretical viewpoints, with the first looking at the core and the systemic
structure, the second looking at interactions in more peripheral areas and the agency of
multiple actors, and the third seeking to incorporate agency with structural models. Next,
drawing primarily on organization of power models, I explore the range of variation in
empires, with a particular focus on the degree of control over provincial areas. This
section includes a discussion of the larger structural variables that influence why empires
pursue particular strategies of control under particular conditions. This includes a focus
on the pre-conquest organization of provincial areas. Finally, I consider what the cultural
24
and economic effects of different rulership strategies would be, and how this variation
would be visible archaeologically.
I follow Doyle (1986:30) in defining empires on a behavioral basis as, “effective
control, whether formal or informal, of a subordinated society by an imperial society.”
This definition allows a large range of imperial formations to be considered, including
those outside of the range of organizational forms seen in the traditional territorial states
of the Old World. Other definitions of empire mirror this emphasis on control outside the
parent society as the critical characteristic of empires (Hassig 1985; Sinopoli 1994). On a
functional level, imperial control of provinces typically involves (1) military conquest,
(2) construction of an imperial infrastructure, (3) imposition of tribute or taxes, (4)
reorganization of settlement patterns, and (5) imperial cooption of local elites (Smith and
Schreiber 2006:3-4), though the overall intensity of such activities and the balance
between them vary widely among empires.
Following an initial military conquest, states must consolidate their control over
newly acquired territory (Sinopoli 1994). Broadly described, consolidation usually
consists of either co-opting local institutions or undercutting them and instituting new
imperially focused institutions. Consolidation activities are often more focused on
dividing existing local sources of power than creating or promoting a single imperial
identity (Berdan, et al. 1996; Sinopoli 1994; Stark and Chance 2012). Ancient states vary
widely in the degree to which their consolidation and control strategies affected local
populations (Alcock, et al. 2001; Blanton and Fargher 2008). The consolidation process
provides an arena for negotiation among the various groups affected, including multiple
factions of conquerors and locals, elites of various levels, and commoners. While
25
provincial residents, especially commoners, do not usually approach such negotiations
from a position of power, the imperial core has a vested interest in creating a stable,
pacified, taxable population, giving provincials and commoners some degree of leverage
(Fargher and Blanton 2007).
To date, most research on archaeological evidence for imperial control has
focused on elite and/or monumental contexts. Pairing these investigations with a
consideration of commoner contexts allows for the evaluation of how deeply imperial
power reached into the population, as well as how well imperial ideologies of inclusion
or exclusion conformed to the physical reality.
Theoretical Models
A brief historical overview of the models used to interpret imperial-provincial
interaction demonstrates a shift from the primarily core/dominant power centric
perspectives of world systems models, through a reaction in the form of agency-centric
models with a peripheral/provincial focus, to multivariate models seeking to explain the
variability in core-provincial relationships. While these perspectives are sometimes
characterized as opposed to one another, they provide insights on different aspects of
empire/province interaction. My work in this dissertation brings together the insights
provided by each of these theoretical perspectives, with my analysis of the economic
effects of consolidation drawing more heavily on a world-systems background while my
discussion of cultural change includes ideas of resistance and negotiation.
26
World-Systems Models. Historically, the most common theoretical model applied
to ancient empires is one of a number of variants of Wallerstein’s world system model
(Wallerstein 1976). Originally designed to address the 15th century to the present, the
model divides the interacting area (world system) into cores, semi-peripheries, and
peripheries. While the model includes political and social interaction, the primary means
of integration are economic, in the form of staple finance and the regional division of
labor. Core areas import raw materials from peripheries, via semi-peripheries (relatively
developed sub-areas in peripheries), and export manufactured goods back to peripheral
areas in exchange (Frank 1966). This leads to a concentration of wealth in the core at the
expense of the periphery and the technological dependence of the periphery on the core,
creating a positive feedback cycle sustaining and expanding the world system.
Wallerstein also differentiates between world-empires and world-economies. In the
former, an expansive geographic area is controlled by a single political entity, making the
spheres of political and economic interaction essentially conterminous and relatively
homogenous. In the latter, there is substantial differentiation in the level of political and
economic development of cores and peripheries and the spheres of political and economic
control are not necessarily the same.
The applicability of this model to ancient states has been critiqued on multiple
grounds, including the excessive focus of core regions at the expense of peripheries
(Wolf 1982:23), the domination of underdeveloped peripheries by cores (Chase-Dunn
and Hall 1997:36), and the primacy of economic interaction generally and staple finance
in particular (Oka and Kusimba 2008; Schneider 1977). As a result, archaeological and
historical formulations, which often use the more relaxed term of world systems analysis,
27
(Abu-Lughod 1991; Blanton and Peregrine 1997; Peregrine 1996) differ from traditional
world systems theory in three major aspects. First, they remove the emphasis on
economic interaction over social or political interaction in many cases (Hall 1997; Hall
and Chase-Dunn 1996; Smith 2003d). In many ancient cases, there were few
technological differences between relatively core and peripheral areas, meaning that there
was little potential for differential economic development. However, at the same time,
core areas might control access to other resources, such as religious authority or political
legitimation.
The second way archaeological applications differ from traditional world-systems
theory is that when economic integration is considered, the stress on staple finance may
be removed (Feinman 1997). Due to the difficulties of transporting staple goods over
long distances in antiquity, economic exploitation of peripheral areas was more likely to
focus on relatively high-value goods. Third, significant work has gone into the discussion
of multi-core systems and more complex core/non-core interactions (Chase-Dunn and
Hall 1997; Feinman and Nicholas 1991; Kohl 1987; Schortman and Urban 1992; Smith
and Berdan 2003a). It has also been recognized that integration into a world-system may
occur at various levels. Under weaker conditions of integration, the core is likely to
influence peripheral areas far more than the reverse, while under stronger conditions of
integration, influence is more bidirectional (Hall 1998).
The combination of responses to these three criticisms has demonstrated that
ancient world systems often exhibited more balanced negotiations between cores and
peripheries, especially when neither could enforce monopolies on particular goods or on
the use of force. What remains central to the world-system approach is the examination
28
of past interaction across political or cultural boundaries and the potential for unequal
relations in such interaction. Unfortunately, there is an inverse relationship between the
degree of relaxation in the original model and its explanatory power in archaeological
cases. As world systems theory, as a set of testable cause-and-effect predictions, shifted
toward world systems analysis, or a general paradigm of looking at interaction and
inequality, it became applicable to a wider range of ancient cases, but provided less
specifically testable predictions about any particular case (Hall, et al. 2011).
Postclassic Mesoamerica has been constructively described as a world-system, in
the more relaxed sense of the term (Smith and Berdan 2003b). Over the Postclassic, the
level of interregional interaction increased, both economically and culturally.
Economically, both higher-value staples (obsidian, salt) and luxury goods across
Mesoamerica were traded widely (Golitko and Feinman 2015; Nichols, et al. 2002; Smith
1990). Due to the multi-core nature of the system and the uneven distribution of raw
materials, this increase in economic interaction was not inherently exploitative of
peripheral regions. Culturally, the Postclassic international symbol set and Postclassic
international art style (sometimes collectively called the Mixteca-Puebla style) spread
across Mesoamerica, occurring on both portable goods such as ceramics, and in fixed
locations, such as murals and stone carvings (Boone and Smith 2003; Smith and HeathSmith 1980).
During the Early and Middle Postclassic, primary centers (i.e. Tula, Cholula,
Chichen Itza) tended to have greater cultural than economic influence, but both extended
beyond areas directly politically controlled by such states. During the Late Postclassic, a
number of larger expansionist states (the Aztec Empire in Central Mexico, the Tarascan
29
Empire in West Mexico, the Quiche state in western Guatemala) rose to power. Given
that there were neither major changes in the distribution of raw materials nor regionally
specific technological innovations between the Middle and Late Postclassic periods,
world systems analysis would suggest that economic and cultural changes between these
two periods were due to changes in the control of force.
I generally use the perspectives provided by world-systems analysis as a reminder
that ancient states were not internally homogenous and that their variation resulted from
the balance in power between core and peripheral areas. More specifically, I use the
potential variation in the relative balance of power between Calixtlahuaca and the Aztec
Empire in structuring the previously presented cultural and economic scenarios for
Aztec/provincial interaction at Calixtlahuaca. Under relatively traditional world-systems
theory predictions, Calixtlahuaca would become incorporated into the Aztec empire as a
periphery (or possibly semiperiphery), leading to generally negative economic effects,
and an increase in goods imported from Basin of Mexico. In a system where a core can
apply an effective monopoly on force (such as the army) and/or the circulation of
particular goods (such as obsidian from particular sources), the core is in a much stronger
position to dictate economic relationships benefitting itself. Similarly, the core may also
“export” cultural practices, at the expense of those previously found in provincial areas.
Agency-Centered Approaches. The core-centric and highly systemic focus within
world systems approaches has been countered by the development of a variety of
overlapping, agency-focused approaches, including middle ground theory (Malkin 2002;
White 1991), hybridization (Tronchetti and Van Dommelen 2005; Van Dommelen 1997),
30
and resistance theory (Hollander and Einwohner 2004; Scott 1990; Spielmann, et al.
2006). These approaches share a common focus on the manners in which individuals and
groups negotiate their interaction with the dominant group, and the range of possibilities
for negotiation and expression under different degrees of control (Stark and Chance 2012;
Stein 2002b; Van Dommelen 1997, 2012). They also generally structure interaction as
tripartite, involving a homeland, colonists, and natives, and acknowledge that the balance
of power between these three groups can vary widely (Stein 2002c). Many such balances
of power do not fit under the definition of empire, and as such, provide a description of
the alternatives to imperial rule over provincial areas.
Middle ground and hybridization models focus on processes in the interaction
zone between cultural groups. Originally developed to explain North American nativeEuropean colonial interaction (White 1991), the middle ground model proposes that
under conditions of loose imperial control, where neither empire nor subjects can bring
overwhelming force to bear, groups are likely to develop a middle ground of mutually
negotiated, often hybrid, behaviors. Both groups will feel that the meaning of such
behaviors is shared, though their actual understandings of that meaning may differ. If the
balance of power shifts heavily toward one party, they will shift toward dictating the
terms of interaction (White 1991). Over longer time spans, hybrid cultural practices will
develop in such intermediate areas, with the potential to spread outward. Direct
archaeological applications of the middle ground model have been primarily limited to
the Middle East and Mediterranean regions (Malkin 2002, 2005). It is most often applied
to situations of limited contact, especially trading colonies (Malkin 2002; Stein 2002a).
As a broader term covering both attempts at finding cultural common ground and the
31
blended results of this process, hybrid/hybridity/hybridization are used across a wider
range of regions and time periods, though there is still bias toward the ancient
Mediterranean and historical archaeology (Card 2013). Despite this geographic and
temporal bias in past applications, these models have significant potential for
understanding interaction under various balances of power in provincial areas. The
conditions where middle ground theory have been applied strongly parallel the conditions
in Schortman and Urban’s (1994) independently developed argument for situations under
which traditional world systems theory is not likely to apply - namely where none of the
parties involved has a monopoly on exchange, technology, or military force. As a result,
while middle ground and traditional world-systems theory may initially appear
contradictory, it is more useful to characterize them as describing positions at opposing
ends of the spectrum of the balance of power between the ruler and the ruled.
Resistance theory based approaches draw primarily on the work of James Scott
(1990). Interaction within a society can be seen as consisting of multiple transcripts
(viewpoints on society), including a public transcript derived from those in power, which
is at least nominally ascribed to by all groups, and various subordinate, resistant
transcripts produced by one or more groups. The public transcript creates, normalizes and
justifies the existing inequalities within a society. It may include material domination
(such as taxes/appropriation of labor), status domination (from the exclusion from
particular social roles, to overt humiliation), and ideological domination (such as
justifications by ruling groups for existing inequalities (Scott 1990:198)). The mode of
expression of resistance is dependent on the degree of control by the dominant group in
32
various arenas of expression (Scott 1990:95), and may not be intended or recognized by
all groups involved (Hollander and Einwohner 2004:544).
Resistance exists most commonly as an ongoing, background process in spheres
outside of the control of the dominant group, but will sometime reach a flash-point and
break into the public sphere. Economic resistance ranges from intentionally working
slowly (hidden) to strikes and revolts (open). Resistance to status domination can range
from relatively hidden activities such as gossip to the overt destruction of monuments.
Ideological resistance can take forms from folk mythology celebrating countervailing
norms to open rebellion based on counter-ideologies. Within both archaeology and the
larger social sciences, resistance has been applied to such a wide range of behaviors as to
lose much of its utility unless carefully limited (Brown 1996).
Archaeologically, resistance is generally hard to see. In its ongoing, background
aspect it often takes ephemeral, hidden, or deniable forms. When resistance erupts into
open expression, it becomes more visible, but such instances tend to be short-lived; either
a rebellion is quashed, or it succeeds and a new set of public transcripts comes into being.
Archaeological applications of resistance have focused primarily on historical rather than
ancient empires (Brumfiel 1997; Liebmann, et al. 2005; Liebmann and Murphy 2011;
Silliman 2001; Spielmann, et al. 2009; Spielmann, et al. 2006). Attempts to apply the
concept to ancient states have met with mixed results; investigations of commoner
background resistance often fail to find archaeological evidence for their claims, or rely
on the presence of activities likely to have provoked resistance (such as taxation) as proof
of resistance (Brumfiel 1997, 1998; Given 2004). In contrast, investigations of resistance
among multiple elite factions have been more successful due to the higher visibility of
33
actions and higher potential for resource mobilization among such groups (Elson and
Covey 2006; Hutson 2002). While these are useful for investigating elite strategies and
breaking down “elites” as a monolithic category, they do not address resistance by those
with the greatest propensity to engage in it.
I draw on agency-based approaches at two levels. First, in looking at provincial
and commoner responses to Aztec rule, I consider the multilateral and negotiated aspects
of empire in general. While Calixtlahuaca is not a middle ground in the strict sense of
mutually beneficial interaction between two otherwise distinct cultural traditions, it is a
colonial situation falling within the range of cultural contact, accommodation and
negotiation often considered in studies of hybridity. This provides a perspective in which
provincial areas have some level of choice in accepting or rejecting economic and
cultural arrangements proposed by a core area. Given that is highly likely that the Yata
phase occupants of Calixtlahuaca included some Aztec immigrants, the site presents a
classic triad of core-immigrant-local interaction.
I also apply agency based approaches in looking at responses to Aztec rule at the
household level. At this scale, I am able to see intrahousehold-level choices in response
to Aztec rule and examine how these correlate to a household’s standing and choices in
other domains of activity. According to the ethnohistoric record, the Toluca Valley
attempted rebellion against the Aztec Empire at least once, suggesting that there was
sufficient resistance to Aztec rule that it could move from hidden to overt resistance. This
suggests that there was likely a transcript of hidden resistance against the Aztec Empire
during a more extensive period.
34
Collective Action Models. The most recent developments in the understanding of
ancient empires are still emerging, but may be generally described as social-organizationof-power models. Examples include corporate vs. network based leadership strategies
(Blanton, et al. 1996), comparisons of direct and indirect rule (Gerring, et al. 2011), and
collective action models (Blanton 2010; Fargher and Blanton 2007; Fargher, et al. 2011).
While the role of elites in state formation is a longstanding one within anthropology
(Morrison 1994), collective action models in particular approach the issue through the
use of historical cross-cultural comparison to associate particular distributions of power
with a variety of aspects of resulting state organization. Social organization of power
models bridge the systemic focus of world-systems analysis and the individualistic focus
of agency based perspectives.
Collective action perspectives on empires fall within a larger body of study of
collective action in general, both in anthropology and related fields (Carballo, et al. 2014;
Ostrom 1990). This body of work addresses conditions under which people are likely to
cooperate for group benefit, even though it may go against their personal interests. It also
looks at the types of institutions that develop to support collective actions and punish
those who violate collective norms.
In particular, where imperialism is concerned, collective action theory assumes
that states represent a pact between taxpayers and rulers, such that the more highly
dependent a ruler is on internal revenue, the more he will be accountable to taxpayers,
resulting in greater taxpayer voice. At the same time, the resulting bureaucratic structures
must be able to keep tax evasion and administrative graft to low enough levels that they
do not overburden the system. Because these models focus on states as existing along a
35
continuum of balance points between rulers and ruled, they have a high potential utility
for explaining the diversity among and within ancient states, a point which is elaborated
further in the following section on specific dimensions of state variation.
I use collective action models as the basis for generating my second cultural and
economic hypotheses, which argue that the economic and cultural changes following the
Aztec conquest were relatively beneficial to provincial populations. This perspective
encouraged me to look for evidence of the benefits of imperial rule in commoner contexts
where they might not be expected in more traditional formulations of imperial/provincial
and elite/commoner interaction, and to look for systematic patterns in the diversity of
potential strategies adopted by states (Berdan, et al. 1996; Gerring, et al. 2011) and
provincial peoples (Stark and Chance 2012).
Classifications of Imperial Variation
The organization and structure of empires have varied widely over time and
space, leading to a range of classification schemes based on different aspects of imperial
organization. For the purposes of analyzing the interaction between empires and their
provinces, two particularly useful dimensions are the degree of direct control over
imperial territories (territorial states/direct control vs. hegemonic states/indirect control),
and the social organization of power (corporate leadership/highly collective states vs.
network leadership/less collective states). While these classifications share elements,
each focuses on different aspects of imperial organization.
36
Direct and Indirect Rule
One axis of variation in empires is the degree to which the empire directly
controlled administrative functions in conquered areas, as opposed to leaving them under
local control. This axis of variation has been referred to both as the distinction between
territorial and hegemonic empires, and between direct and indirect control (Hassig 1985;
Luttwak 1976). For a wider range of positions along this axis of variation, Rodgers
(Rogers 2005) gives four imperial strategies, from organization imposition (direct rule),
to dual administration and overlay incorporation (intermediate forms of rule), and finally
marginal incorporation (very indirect rule). Under territorial or direct control, conquered
areas are ruled directly by the imperial state, via state organized and controlled
institutions. The state focuses on the control of territory and the defense of this territory
against external threats. This generally requires a standing army located toward the
boundaries of the imperial territory, sufficient political centralization to effectively
provision and direct the armies, and strong internal control to minimize the potential for
revolts.
Militarily, direct-rule states require a sufficient standing army to both protect
frontier areas from further aggression, and to maintain internal order (Hassig 1985:Chap.
5). Administratively, direct-rule states take on responsibility for monitoring compliance
with imperial policies (such as paying taxes), and in pre-modern states geographic
distance alone made monitoring geographically distant provinces a costly proposition
(Hechter 2013:Chap. 4). Direct rule empires can depend on internal (Smith 2004a) or
staple finance (D'Altroy and Earle 1985), due to the degree their penetration into
provincial areas, potentially providing a greater and more stable tax base than would be
37
available from wealth finance alone. However, due to the higher degree of changes to
administrative structures, the higher degree of day-to-day interaction between residents of
the provincial areas and representatives of the empire, and the greater imperial
investment in provincial areas, territorial strategies cause a higher degree of cultural
integration of provincial areas. Due to both military and administrative costs, direct rule
is costlier overall for the state, but in exchange, also tends to be more internally stable.
In contrast, hegemonic states focus on the political control of existing units. They
leave existing power structures in place, as long as such structures can be co-opted to
serve imperial purposes. They use buffer or client states to defend against most ongoing,
small-scale aggressions from outside the empire, reducing the overall need for a standing
army for external defense. As a result, the army can be deployed internally to quell
rebellions, requiring less development of other mechanisms to prevent revolts. They also
rely on local leaders and preexisting administrative structures to enforce state policies,
collect revenue, and provide public goods, shifting the costs of rulership onto provincial
leaders (Hechter 2013:Chap. 4). This leads to the development of less imperial
infrastructure in provincial areas and a heavier dependence on existing local leadership
and bureaucratic structures. Indirect rule is generally associated with a tendency toward
wealth finance (D'Altroy and Earle 1985), as a smaller quantity of higher value goods
require less provincial infrastructure to collect than a larger quantity of staple goods. As a
result, indirect is less costly for the imperial state on a day-to-day basis, but is also more
politically unstable since local leaders maintain control over their traditional power bases.
Gerring et al (2011) argue that empires will pursue more territorial or more
hegemonic strategies of control under specific conditions. The first of these conditions is
38
the relative level of social development of the core and provincial areas. In order for
hegemonic/indirect strategies to function effectively, the provincial area must have a
sufficient preexisting level of technological development and administrative complexity
to effectively function as a proxy for the imperial government. As a result, state (and
some chiefdom) level societies can be effectively controlled using a hegemonic strategy,
while less hierarchical societies generally require a more territorial strategy to construct a
new administrative infrastructure. Historical examples of this principle may be seen in
the variation in British strategies relative to their colonies (Gerring, et al. 2011), the
variation in the Spanish success in establishing control over native populations at various
levels of social complexity (Mahoney 2010), and the differential use of forced labor in
the Americas under European rule (Arias and Girod 2014).
Geographic distance also places constraints on strategies of rule (Feinman 1998).
In premodern states, the cost of projecting power increases sharply as a factor of distance.
As a result, empires must accept that progressively more distant provinces will either (a)
become costlier to rule at the same level of integration, or (b) will have to be ruled more
indirectly. Late Imperial China resolved this problem by making the territories of
officials near the edges of the empire smaller than those at the center, in order to maintain
a consistent degree of control (Skinner 1977). The Aztec Empire took the opposite
solution, ruling many peripheral areas more indirectly via client states, with less
regularized tax collection and less formal organization into provinces (Smith 1987a).
In addition to imperial preferences for direct or indirect rule, political science
predicts that residents of provincial areas will also tend to support one form of rule over
the other in predictable fashions (Siroky, et al. 2013). If indirect and direct rule can
39
provide comparable public goods, people will generally favor indirect rule (Hechter
2013). Local leaders are easier to hold accountable for their actions than foreigners both
due to their higher degree of shared cultural values and their simple geographic
proximity. However, direct rule may be seen as preferable if it can provide greater public
goods, or a more even distribution of the public goods. A more even distribution of
public goods may be grounds for support of direct rule by some local subgroups if local
leaders are perceived as favoring particular segments of the population, such as members
of their own ethnicity or religion. In an empirical test of this prediction, Ferwerda and
Miller (2014) found that during World War II in France Vichy controlled (indirect rule)
areas had significantly fewer incidents of resistance than geographically and socially
comparable areas under direct German control.
In addition to general provincial trends, particular subgroups may differentially
support particular forms of rule. In semi-independent modern states, such as the ethnic
republics of the Russian federation, upper-class individuals are more likely to support
more indirect rule, while members of the middle class are more likely to support more
direct rule (Siroky, et al. 2013). While the particulars of this modern case are not
specifically applicable to prehistory, they do generally demonstrate that provincial
peoples will support the form of rule that most benefits them and that this is different for
different subgroups of a provincial population. Provincial elites are generally likely to
favor indirect rule, as this allows them to keep most of their existing authority and benefit
from providing services to the empire.
Whether commoners would favor direct or indirect rule is likely based on their
prior position relationship with local elites; if this relationship is good, commoners will
40
likely favor indirect rule to maintain the status quo, while if elite/commoner relations are
poor, commoners may support more direct imperial rule as a means of limiting the power
of the local nobility. In the previously mentioned study by Ferwerda and Miller (2014),
French resistance was driven by politically disenfranchised groups – the left wing in
Vichy (French right-wing) territory, but both right- and left- wing groups in areas under
direct German rule.
The Aztec Empire falls firmly toward the hegemonic end of this scale, due to its
use of client states (“strategic provinces”), highly selective use of garrisons in frontier
areas, and general practice of leaving local leaders and administrative organizations in
place (Smith and Berdan 1996b). It is sometimes considered a classic example of the
hegemonic end of the scale (Hassig 1985, 1988). However, while not moving the Triple
Alliance from this general end of the scale, a broader consideration of the activities of the
empire does provide cases of more direct control. By the time of the Spanish conquest,
approximately one third (18/55) of provinces had an imperial military governor in place
of their native ruler in one or more cities (Smith and Berdan 1996a). In addition, there
were probably three levels of hierarchy within the imperial tax collectors assigned to each
province, which indicates at least a basic imperial bureaucratic structure in provincial
areas, rather than a complete reliance on local infrastructure (Smith 2015a:78). There
were also differences in the directness of rule based on the geographic distance from the
Basin of Mexico, with Aztec control becoming less direct with distance. This particular
point is considered in more detail later in this chapter.
41
More and Less Collective Rule
A second axis of variation in imperial organization concerns the social
distribution of power in a state. On a general level, this is described as the degree of
collective action in a given state. As applied to states, collective action refers to the
degree of cooperation among individuals and groups within the state (Olson 1965).
Building a state along more collective lines, rather than ones based on dominance,
requires overcoming human tendencies toward self-serving behavior. Doing so requires
two sets of checks – one on subjects who might seek to freeload and one on leaders who
would seek to exploit their subjects. Checks on commoner subjects may be provided by
either other commoners or by the state, but checks on elites are only effective to the
degree that commoners control some resource desired by elites. As a result, collective
states are most likely to develop in contexts where state revenue is primarily dependent
on internal sources (i.e. citizen taxes/staple finance) rather than external sources (i.e.
long-distance trade/wealth finance). Collective states will also feature relatively high
levels of public goods provided to the population, a well-developed bureaucracy to
manage the collection of resources from the population, the provisioning of public goods
and the limiting of freeloading, and checks on the powers of rulers.
A more specific subset of work on collective action focuses on how leaders use
collective action. Corporate and network strategies (sometime also referred to as
cooperative and individualizing (Carballo 2015)) describe alternative approaches to
legitimizing power structures (Blanton, et al. 1996; Feinman 1995). The terms may be
used to describe either the strategies of leaders, or the forms of social organization
favoring the development of/resulting from those leadership strategies. Corporate
42
strategies draw heavily on collective action. They are socially inclusive and focus on
creating a powerbase by power-sharing. They are likely to feature inclusive language
between commoners and elites (such as “being part of the same lineage”), little emphasis
on markers of elite status, unrestricted monumental architecture, and an emphasis on
offices rather than individuals. In Mesoamerica, most Central Mexican civilizations fall
toward the corporate end of the spectrum, with Teotihuacan as a particularly good
example of a highly corporate state (Blanton, et al. 1996). Teotihuacan is characterized
by a very homogenous provisioning of public goods such as apartment compounds and
neighborhood temples (Smith, et al. 2014). It also has little evidence for individual rulers;
no clear royal palaces or burials have been located to date, and artwork emphasizes
markers of offices rather than named individuals.
In contrast, network strategies are based on social exclusivity and are generally
less collective. Elites seek to gain power based on their control of material (foreign trade,
particular types of luxury goods) or non-material resources (religious authority). Elites
gain support from others elites by sharing such resources among limited groups of
people. Because leadership is not dependent on popular support, network oriented states
generally have less collective action, especially less controls on principals. Network
oriented states tend to have origin stories that emphasize the difference between elites
and commoners (such as separate creations of different classes, or divine justification for
rulership), a strong emphasis on visible markers of status differences, restricted-access
monumental architecture, and an emphasis on leaders as individuals rather than officeholders. In Mesoamerica, the Classic Maya city-states provide a good example of
network-oriented leadership strategies (Blanton, et al. 1996). Luxury goods were
43
produced by specialists attached to elite households (if not by elites themselves), and
distributed as gifts along patronage networks. Monumental architecture often has limited
public access and artistic depictions of rulers emphasize their individuality and
relationships to other individuals.
As with territorial and hegemonic rule, there are certain preconditions under
which more or less cooperative strategies are likely to develop. One of the primary
conditions is the ease of exit from a society. In societies with higher exit costs, either due
to sunk costs in infrastructure or a lack of alternative destinations, commoners are more
likely to remain under less collective regimes, due to a lack of options. It is also easier for
highly collective societies to develop in relatively small states, where information and
sanctioning costs are lower, as is the cost of developing a strong bureaucracy.
Within large, heterogeneous states, such as empires, the degree of collectiveness
of rule may vary, a point which is not taken into consideration in all studies of the topic.
Scheidel (2006, 2015) argues that states organized along non-republican (i.e.
network/non-collective) lines can expand indefinitely with little to no loss of privilege to
those who are already members of the state. In such states, benefits accrue primarily to
elites and are not diluted by the addition of provincial elites to the system. In contrast, in
what Scheidel calls republican states, which generally correspond to collective forms of
rulership, territorial expansion has the potential to result in a dilution of benefits to
current members. As a result, such states may end up with dual (or more) layer
membership, with different groups having differing rights and privileges. In a historical
example, this can be seen in the distinction between Roman subjects and Roman citizens.
44
Different subsets of a population may also pursue different types of corporate or
network strategies, based on which they see as most beneficial to themselves. While
Teotihuacan generally follows a corporate strategy, Manzanilla (2015) argues that the
bottom (similar apartment compounds housing most of the population) and top (few
named leaders, no known royal burials) of society were highly corporate, while
intermediate elites practiced a somewhat more network-oriented strategy based on
restricted control over non-local goods from particular regions. As a general rule,
network-based imperial strategies will focus on integrating provincial elites, while more
corporate strategies will reach a broader section of the provincial population. As a result,
the degree of economic and cultural integration of commoners in provincial areas can
serve as a useful marker of rulership strategies.
In their cross-cultural analysis of collective action, Blanton and Fargher find that
the Aztec Empire was relatively collective, ranking sixth out of thirty state societies for
overall evidence of collective action (Blanton and Fargher 2008:Table 10-11). Rankings
in individual subcategories of the analysis (Public goods provided by the state,
Bureaucratization, and Control of leaders) are generally similar (5-7th place) to the overall
ranking indicating that the overall placement is not driven by a single anomalous factor.
However, within Postclassic Central Mexico, an analysis of relative collectivity (Fargher,
et al. 2011) found that there was substantial variation among subregions and cultural
traditions, with Tlaxcalla as the most collective, the Mixteca Alta and the Puebla Valley
outside of Cholula as the least, and the Basin of Mexico, Morelos and Cholula in an
intermediate position. This variability is visible both in early colonial written descriptions
of the various regions, and in archaeological evidence. Based on written sources, the
45
commoners in the more collective areas generally had stronger rights to land as
freeholders rather than landless laborers, and more opportunities for social advancement
within the government bureaucracy. Archaeologically, the more collective areas
generally have more dispersed urban monumental architecture with open, easily
accessible plazas, and a more even distribution of non-local goods across sites at various
levels of the settlement hierarchy.
The possibilities of regional and class-based differences in collective action raise
an important point for interpretations of the Aztec Empire. The underlying cultural
traditions in Central Mexico were relatively collective (Carballo 2015) and as such Aztec
society can also be considered fairly collective. However, prior to the expansion of the
Aztec empire, collective obligations and benefits were largely organized at the
neighborhood or city-state level. Under Aztec rule, the majority of these collective
obligations continued to be provided by the city-state or neighborhood, rather than by the
empire. As a result, it is misleading to describe the Aztec Empire, at a minimum those
portions of it outside of the Basin of Mexico, as especially collective in and of itself.
Within the Basin of Mexico may be a somewhat different story, as the capitals of the
Triple Alliance both provided more public services to the population in general (dikes on
the Basin Lake system, aqueducts, market administration), and were more accountable to
the local population, especially the advisory councils of major city states.
The two dimensions of variation in imperial rule discussed here are measuring
somewhat different aspects of control. Territorial/hegemonic rule primarily describes the
degree of imperial interference in provincial areas, and can be measured by the degree of
changes following imperial conquest, regardless of the directionality of change. In
46
contrast, the collectiveness of rule measures the particular strategies used by rulers to
consolidate their power, and particular forms of rule are linked to broader directional
changes. These differences notwithstanding, the two dimensions should be reasonably
correlated. Indirect rule provides relatively few public goods, and allows for relatively
little accountability of imperial rulers. Direct rule will require a higher overall level of
collective action, as at least some imperial bureaucracy (for the collection of taxes, if
nothing else), and some public goods (to replace those previously provided by the local
state) will be provided by the empire. However, within these general limits, direct-rule
empires may pursue more or less collective strategies.
Economic Effects of Rulership Strategies
World-systems analysis predicts that cores will generally economically exploit
their peripheries (Wallerstein 1976). However, as Schortman and Urban (1994) point out,
this effect is lessened in cases where more peripheral areas can play multiple cores
against one another, due to any given core lacking a monopoly on force or necessary
goods. As a result, incorporation into an empire is likely to push peripheral areas toward
a more subordinate economic status as they lose the ability to access cores outside of the
empire due to military hostilities or protectionist economic policies. In the particular
context of the political economy of empires, this agrees fully with Eisenstat’s (1993
[1963]:Chapter 7) assertion that the primary goal of empires is the economic exploitation
of their hinterlands. The differences in the economic effects of rulership strategies can be
considered primarily in terms of how effective the imperial core was at extracting
47
resources from the province and what, if any, reciprocal benefits the imperial presence
provided.
More indirect forms of imperial rule will generally have fewer economic effects
on provincial areas. They produce few changes to existing economic structures, for either
production (no imperial workshops) or distribution (no reorganization of market
networks). Provinces under indirect rule also retain a greater ability to negotiate their
position vis-à-vis multiple cores. Indirect rule will generally have somewhat negative
overall economic effects on provincial areas, since the empire extracts an additional level
of taxes from provincial areas while providing few additional services that might
otherwise promote economic development. Indirect rule may not even result in an
increase in trade with other provincial areas due to increased interregional stability, as
indirect rulers often allow interregional conflicts among provinces to continue as a means
of channeling aggression away from the state. As a result, the economic (non)effects of
highly indirect rule should be visible as a continuation of pre-imperial economic trends,
including little change in the quality or diversity or long-distance trade, or the
organization of craft production. Overall wealth levels may show a slight tempering of
any prior trends toward economic growth.
Direct rule has the potential to produce much larger economic effects in
provincial areas, but the degree to which it will produce economic changes is a product of
how different the provincial and state economic systems are, as well as how collective a
state is. It may result in changes to the organization of production, either directly (such as
state-sponsored workshops) or indirectly (such as the intensification of production due to
increased market opportunities). It is also likely to cause changes in the regional
48
circulation and distribution patterns of goods, due to a greater imperial ability to enforce
preferential trade with some areas at the expense of others. Due to the general trend for
core areas to favor themselves, this is likely to produce a pattern of increased trade with
the imperial core, at the expense of prior trade partners. The shift from local provincial to
imperial rule will likely result in changes to the basic spatial patterning of economic
institutions, either due to the introduction of new institutions or if the empire places its
center of control in a different geographic location than the previous independent
provincial center of power. As a result, direct rule can be expected to produce visible
changes in household-level economic activities. However, outside of a general shift
toward increased trade with the imperial core, the specifics of such changes are
dependent on variables other than the presence of direct rule alone.
Non-collective or network oriented rulership will also cause relatively few broad
economic changes in provincial areas. Due to this form of rulership’s reliance on external
finance, leaders have little incentive to develop local economies, as these are not their
primary tax bases. In addition, elite control over most of the resulting long-distance trade
goods is unlikely to promote widespread demand for such items. While elites pursuing
network strategies are likely to develop geographically extensive trade ties in order to
access a broad range of exotic goods for marking their statuses, the products exchanged
along these ties do not enter general circulation. In fact, elites may seek to restrict
commoner access to broader exchange networks in order to preserve the exclusivity of
their own access to particular goods. In addition, rulers in non-collective states generally
have less penetrative social power, and thus less ability to change existing or establish
new economic institutions in general. As a result, non-collective rule should generally
49
appear archaeologically as a limited and uneven distribution on non-local/luxury goods
among commoner households. It should also not promote an increased reliance on market
provisioning, thus limiting the development of higher levels of independent (rather than
elite-attached) craft specialization.
Relatively collective rulership is more likely to promote internal economic
development. This is due in large part to its greater reliance on internal finance; a
growing internal economy offers a growing tax base for a collective state. In addition, the
lower degree of restrictions on purchasing offers greater potential for economic growth.
This is likely to result in an increase in market exchange and an increase in craft
production for trade above the household level in a feedback loop of economic growth
(Millett 2001). As a result, collective rulership is likely to result in increased trade
(though it may be biased toward the imperial core), increased productive specialization,
and have neutral to positive effects on local levels of wealth.
In summary, direct and indirect rule can primarily be differentiated by the degree
of interruption they produce in existing systems, including in commoner households,
regardless of the directionality of that change. In contrast, the collectiveness of rule is
marked by both the direction and evenness of changes. Relatively non-collective
rulership is likely to produce uneven and generally negative effects in commoner
households, while relatively collective rule promotes general economic growth among
much of the population.
50
Cultural Effects of Rulership Strategies
Ancient states2 were generally multiethnic, as a fundamental result of their
processes of expansion, and generally did not see internal ethnic homogeneity as an
inherent goal (Hall 1998). In many cases, differences among subject populations of
empires were fostered, or at least tolerated (Sinopoli 2001). The association of a state
with a uniform cultural identity, and the association between ethnicity and demands for
self-determination are products of modern nation-states and should not necessarily be
projected into the past (Kedourie 1960). However, identity differences also provided
potential fracture-lines leading to conflict and collapse (Emerson and Hedman 2016). The
combination of these two factors means that the provinces of ancient empires are fertile
grounds for the study of the formation and maintenance of cultural identities. While the
initial imperial decisions about the rule of particular provinces were based on the
relationship between the core and province, the structure of ongoing interactions
depended on the relationships among three groups – the imperial core, the local
provincial population, and the individuals from the core who take up residence in the
province (Stein 2002c). This is particularly true of cultural (rather than economic)
interaction, where provincial groups are likely to have a greater range of choices whether
they participate in imperial systems or not. Official state/provincial interaction is likely to
be concerned with official imperial ideology. In contrast, colonist/provincial interaction is
In this case, I use the term “state” in a sense closer to my definition of “empire”, rather than a broader
definition including city-states, which may well be largely monoethnic.
2
51
likely to promote the transmission of a wider range of quotidian practices from the
imperial core, as well as more diverse forms of hybridization between foreign and local
practices (Wells 2005). As a result, I consider both forms of interaction as possible
vectors for the introduction of non-local cultural practices and objects into provincial
settings when evaluating the effects of different rulership strategies. In addition to
discussing the general cultural effects of different rulership strategies, I also consider how
these strategies would be expressed in two particular domains of commoner household
activity: ritual practices and foodways.
The results of imperial economic policies can be separated from the effects of
imperial policies targeted toward cultural integration based on the visibility of the artifact
classes involved. Economic shifts are likely to cause changes in the frequencies and
sources of a wide range of goods, including both differentiated (“branded” or
producer/regionally identifiable items, such as decorated pottery) and non-differentiated
items (“commodities” or goods that generally cannot be distinguished among producers,
such as obsidian blades or plainware pottery). In contrast, imperial policies specifically
oriented toward producing cultural changes, such as the integration of diverse cultural
groups, are likely to concentrate on differentiated types of goods, especially those used in
high-visibility contexts.
As with economic change, the directness of imperial rule is associated with the
generally expected degree of cultural change. In cases of relatively indirect rule, there is
less contact overall between local people and people from the core. Because the majority
of the service provisioning and tax collection remains in local hands, there are fewer
officials assigned to provincial areas and they are less likely to personally interact with
52
the lower levels of the bureaucracy, let alone the bulk of the population. In addition, it
may be less likely to see large numbers of colonists under conditions of indirect rule.
There is no strong state mechanism for promoting the official movement of groups into
provincial areas and provisioning them with goods once they arrive. In addition, the
indirect state generally lacks the authority to directly enforce privileges toward colonies
in provincial areas, such as preferential tax statuses, military protection from local
aggression, and/or differential access to non-local goods. As a result, those colonies that
do occur under indirect rule are likely to be largely self-organized (e.g. religious exiles,
trade outposts) and depend heavily on their maintaining good relationships with their host
communities. Indirect rule states are unlikely to seek to impose cultural changes on
provincial areas. First, they generally lack the power to do so, and second, because they
rule through local proxies, the local cultural justifications for rule by these proxies cannot
be overtly challenged.
In contrast, under more direct forms of rule, there will generally be a higher
degree of overall cultural interaction between the imperial core and provincial areas.
Under this form of rule, the empire has both the infrastructural penetration to interact
with a broad portion of the local population and sufficient control of force to mandate and
support colonies. Bureaucratic functions, including public service provisioning, are
largely taken over by the empire, which means that they can be run according to imperial
norms or adjusted to incorporate imperial elements. As a result, the local population is
exposed to imperial practices in relatively formal settings. In addition, direct rule states
have a greater ability to support state-sponsored colonies, either of people from the core
or of people moved from one provincial area to another, likely leading to greater
53
interaction in informal settings as well. Whether or not direct rule states seek to impose
cultural practices on conquered areas is largely subject to ideological factors, and as a
result, the overall degree of integration is subject to substantial variation.
From an imperial perspective, the relationship between the directness of rule and
the maintenance of local identities is a double-edged sword:
“At the same time that direct rule stimulates cultural identities in some contexts, it
also affords the center with a greater capacity to suppress collective action on the
basis of these same identities. Indirect rule does the reverse – it supplies more
autonomy to peripheral groups, endowing them with a greater capacity to
challenge the state, but by the same token it also removes much of their incentive
to seek greater autonomy.” Siroky et al 2013, p.3.
As a result, the intensity of local resistance cannot be taken as a direct reaction to a
particular form of imperial rule, but rather should be considered part of an ongoing
feedback process between imperial and local strategies.
The degree of collective action in a state and the associated rulership strategies
impacts the vectors for cultural interaction. Less collectively oriented states will generally
have a high degree of elite integration into pan-imperial elite cultural practices, but will
not have strong reasons to promote imperial cultural practices among non-elites. As a
result, local elites are likely to serve as the primary brokers of imperial culture in local
settings. As with economic benefits in relatively non-collective states, this is likely to
produce an overall low level of integration into imperial cultural systems and an uneven
distribution of those traits that are present, as exposure to foreign practices and access to
54
the necessary knowledge will be strongly mediated through patron-client relationships
between local elites and commoners.
More collectively oriented states will generally feature a higher degree of cultural
integration due to multiple aspects of their organization. They will promote the
development of a shared group identity between leaders and subjects in order to promote
civic participation and improve voluntary compliance with social norms. In addition,
public goods provided by the empire will serve to develop a shared set of practices
among their users. However, public goods are by definition provided above the
household level and their use is not likely to be visible in household contexts.
In addition to interaction with imperial practices in official settings, local people
may also interact with colonists. Relatively collective states have sufficient social
penetrative power to enforce the movement of people from one area and then support
them once they arrive in the new area. They generally also allow greater individual rights
relative to the state, so households which choose to move for economic or personal
reasons are generally permitted to do so. Both of these means of commoner movement
allow for relatively high levels of interaction between local and immigrant populations,
with a higher subsequent potential for the transfer of cultural practices and goods.
The widespread presence of imperial material culture in commoner households
will be the most common in relatively collective states, as this type of rulership will
encourage commoner participation in state activities (such as festivals and feasting), and
will not place barriers on the circulation of foreign goods. In contrast, less
collective/network orientations among elites will tend to limit the circulation of imperial-
55
linked items as these are an important means of demonstrating exclusive ties among
provincial and imperial elites.
Ritual Practices. Religion – and its material expression in ritual – extends across a
spectrum from inclusive to exclusive practices (Insoll 2004:Chapter 2; Renfrew 1994:50),
though archaeologists often focus on one side or the other (Fowles 2013). This distinction
between inclusive and exclusive religious practices maps well onto the distinction
between more and less collective forms of governance (Carballo 2015:Table 3.3). Crossculturally, Peregrine (2012), finds that that the secular vs. divine basis of rule is strongly
correlated with the degree of social collectivity of a society, with less collective societies
being more likely to use religiously based justifications for rulership. The degree of social
collectivity in a society is also likely to influence which contexts ritual is used as a
legitimating force. Smith (2002) divides Aztec ritual practices along public/private and
state/domestic dimensions. In network-oriented societies, elite energy will primarily be
invested in private state rituals, or public rituals that reinforce social differences. In
contrast, in corporate-oriented societies, state rituals will generally be public affairs
drawing on broad participation and may grade into public household practices. This
distinction has broader implications for the degree of homogeneity in domestic ritual
practices, as more collective/corporate social organizations will promote broadly
integrative practices, resulting in increased household homogeneity, while the
individualistic (and often ancestor-linked) orientation of network strategies will produce
more variation as households or other social grouping seek to distinguish themselves
from others.
56
In Central Mexico, there is a contrast between religious concepts and symbol sets
linked to concerns of general (corporate) interest, especially agriculture, and those linked
to rulership (Carballo 2015). The former are much more stable over broad temporal and
geographic ranges of Mesoamerica and are found in both commoner and elite contexts. In
contrast, items and symbols linked to rulership cults and found primarily in elite contexts,
tend to have much shorter cycles of use. In the Aztec case, this distinction between
corporate and network strategies in religion can be seen in the dual-pyramid form of the
Templo Mayor itself. The Tlaloc side of the temple is the expression of a very
longstanding Mesoamerican symbolic complex, which dates back to at least the
Formative period. With agricultural associations to water and fertility, the Tlaloc side of
the temple drew on concerns of collective societal interest. In contrast, the Huitzilopochtli
half of the temple, while partially drawing on longstanding symbolic complexes
surrounding the Old Fire God, gave them a particular Mexica ethnic- and elite classbased focus by conflating older traditions with the Mexica patron god.
Foodways. Foodways also provide a useful means of evaluating forms of
rulership, as they have both economic and cultural dimensions. Direct and indirect rule
are likely to be related to the overall degree of change in food-related practices, as an
aspect of overall cultural interaction under the two forms of rulership. In addition, forms
of rulership which produce economic hardship and/or reward economies of scale are
likely to result in changing strategies of food consumption. In the Basin of Mexico, Aztec
rule resulted in changes to both the intensity of work devoted to food preparation and in
the portability of the resulting foods (Biskowski 2000; Brumfiel 1991).
57
The distinction between corporate and network leadership strategies maps onto
the difference between inclusive and exclusive feasting events (van der Veen 2003). The
former demonstrate power through the volume of food provided and the number of
people served. They generally feature cuisine similar to that served in other contexts,
distinguished primarily by its quantity during the feasting event. Such events are socially
integrative, involve individuals from multiple social classes, and are often provided as
rewards for cooperative labor. In traditional communities in modern Mesoamerica,
collective community labor must be reciprocated with alcoholic beverages and/or food
provided by the town leadership, or people will not participate (Carballo 2015). This
makes such events an integral part of corporate leadership strategies. Archaeologically,
inclusive feasting events will be visible as unusually large accumulations of serving
vessels and food refuse that are otherwise similar to what is found in domestic contexts.
Inclusive feasting events are also likely to have effects on more quotidian household
foodways, as they can introduce news foods to a broad portion of the population at once.
In contrast, exclusive feasting events are characterized by restricted invitations
and distinctive cuisines. They are likely to be hosted by elites for other elites. Hosts serve
exotic foods or food involving complicated preparation methods, allowing them to
demonstrate their connections to restricted goods and/or knowledge. Visitors will be
expected to reciprocate, at which time they are expected to demonstrate their own ability
to acquire rare foodstuffs and their knowledge of the status-appropriate methods for
preparing and serving them. As such, exclusive feasting plays an important role in
network-oriented leadership strategies. Archaeologically, exclusive feasting should be
visible as a higher diversity of vessel forms and rare food remains in both immediate
58
feasting contexts and elite households in general. On an everyday household level,
exclusive feasting’s broader effects will result in uneven knowledge of or access to new
food related practices.
The Aztec Empire
The Aztec Empire was founded in AD 1428, when the city-states of Tenochtitlan,
Texcoco, and Tlacopan formed an alliance and tribute-sharing agreement. This
arrangement developed out of a background of competing central Mexican city-states and
confederacies embedded in a larger Postclassic Mesoamerican world-system (Smith and
Berdan 2003a) characterized by a high degree of economic and information exchange.
The alliance promptly proceeded to conquer the rest of the Basin of Mexico and by the
1440s had begun a series of conquests farther afield. By the Spanish arrival in AD 1519,
the alliance was dominated by Tenochtitlan and received regular tax payments from citystates across Mesoamerica (Barlow 1949; Smith 2015b). The imperial territory can be
broadly divided into tributary provinces, which were city-states grouped into provincial
administrative units that owed regular tax payments to the imperial capitals, and strategic
provinces, which were allied client-states that served as military buffers against hostile
states and were negotiated with on an individual basis (Berdan, et al. 1996) (Figure 2.1).
Traditional conceptions of the Aztec Empire hold that it exercised a relatively low
degree of direct control over its conquests, even characterizing it as the quintessential
hegemonic state (Hassig 1988). This directly contrasts with the results of collective
action based evaluations of the Aztec Empire, which found that it was highly collective
59
(Fargher and Blanton 2007; Fargher, et al. 2011). This apparent contradiction is likely a
result of the latter perspective conflating services provided by the empire proper, with
those provided by the local city-state, especially outside of the Basin of Mexico. The
results, however, still suggest that commoner choices could and did impact Aztec state
policy, especially at intermediate levels of the control hierarchy.
Figure 2.1 The maximum extent of the Aztec Empire, showing the imperial core,
territorial, and strategic provinces.
Imperial and Local Strategies
The negotiation between state and local goals can be seen in the interplay of the
strategies used by each. The Aztec Empire’s strategies are divided into four major
60
categories, referred to as political, economic, elite, and frontier strategies (Berdan, et al.
1996; Berdan and Smith 2003). Activities in these four categories overlap, but the
divisions make a useful framing device. Ethnohistorically, applications of these strategies
included the promotion of imperial ideology through public ceremonies, elite
intermarriage, interference in questions of succession, reassignment of market and/or
head-town locations, and the immigration and resettlement of populations. All of these
strategies were part of longstanding Mesoamerican traditions of political control, though
the Aztec Empire expanded them to their largest extent (Smith and Berdan 1996b). The
goal of imperial strategies, Aztec or otherwise, is to provide a steady income stream to
the imperial core in a cost-effective manner.
In contrast, provincial strategies include bolstering (elite collaboration with
imperial political interests), emulation (elite and non-elite use of an imperial style),
resistance (the attempted blocking of imperial interests up to and including active
rebellion), exodus and internal population movement (migration beyond or within
imperial boundaries), information control (concealing information for provincial benefit),
appropriation (the selective adoption of imperial styles for local ends), complicity
(collaboration, usually by elites, for economic gain), and assimilation (integration into the
dominant society) (Stark and Chance 2012). As can be seen in the diversity of options in
this list of provincial strategies, provincial populations have a wide range of potential
responses to imperial actions.
Both imperial and provincial strategies include options that reinforce, options that
undercut, and options that are effectively neutral relative to the goals of the opposing
group. None of these strategies are limited to a particular form of rulership, either.
61
Instead, they provide a general perspective on the more concrete, specific methods that
rulers and ruled can employ in their interactions with either other.
Previous Studies of Aztec Imperialism
I now provide a brief sketch of the relationship between the commoners and the
empire and the provinces and empire more generally, based on existing research. Because
world systems approaches and middle ground/hybridization models both predict that
distance plays an important role in structuring empire/province interaction, I discuss the
evidence from the Basin of Mexico and the provinces separately.
The Basin of Mexico
The Basin of Mexico formed both the political core of the Aztec Empire and one
of the primary cores of the greater Postclassic Mesoamerican world system. The region
has a long history of archaeological investigation, though it has not always been oriented
toward the types of contexts that can be used to answer questions about the organization
of imperial rule. Projects that do have the potential to contribute to such questions include
the Basin of Mexico survey project (Parsons 1971, 2008; Parsons and Whalen 1982;
Sanders 1965), which systematically recorded sites across most of the Basin, more
intensive surveys at Huexotla (Brumfiel 1976), salvage work in Tlateloco (González Rul
1988a, 1988b), reconstructions of historical excavations at Chiconautla (Elson 1995;
Nichols, et al. 2009), and excavations at Chalco (Hodge 2008), Xaltocan (Brumfiel
2005b; De Lucia 2011; Overholtzer 2012), Cihuatecpan (Evans 1988) and other rural
62
sites in the Teotihuacan region (Parsons 1966). In addition, there has been extensive work
at the Templo Mayor site, covering the primary temple of the Mexica capital of
Tenochtitlan and some surrounding buildings (Boone 1987; López Austin and López
Luján 2009; López Luján 2005), which provides a view of official Aztec state ideology
(See Figure 2.2 for site locations). In the Basin of Mexico, the beginning of the Triple
Alliance in the early 1400s traditionally marks the division between the Middle and Late
Postclassic periods (sometimes also called the Early and Late Aztec periods).
Archaeological studies of Aztec imperialism have documented substantial changes within
the Basin of Mexico during these periods, as well as substantial intraregional variation.
Figure 2.2 Comparative Postclassic Sites in Central Mexico used for analysis in this
dissertation
63
Economically, there are three major trends in the Basin of Mexico during these
periods. The first is an increasing amount of trade within the Basin. A combination of
stylistic analysis and INAA sourcing of ceramics from a variety of sites across the Basin
of Mexico demonstrates a shift from Middle Postclassic multi-producer, multi-directional
trade primarily limited by the political boundaries of the mini-empires of this period, to
decreased interregional trade combined with increased trade with the capital cities of the
Triple Alliance following their political consolidation of the Basin in the Late Postclassic
(Garraty 2007; Garraty and Stark 2002; Hodge, et al. 1992; Minc 2009; Minc, et al. 1994;
Nichols, et al. 2002; Nichols, et al. 2009). Obsidian provisioning shows a similar shift to
heavy dependence on the Aztec-controlled Pachuca source and a decrease of local blade
production in most of the Basin (Millhauser 2005; Pastrana 1998).
Second, there is an increase in site and/or regional craft specialization within the
Basin. Cotton spinning increased across the Basin, either in response to increased tribute
demands or increased market opportunities (Brumfiel 1991). Evidence for the production
of other types of goods outside of the imperial capitals is highly variable, with both high
(Charlton, et al. 1991; Evans 1992; Otis Charlton, et al. 1993) and low (Brumfiel 1980,
2005b) frequency cases. The overall pattern is of food production intensification where
possible, primarily in the southern portion of the Basin, accompanied by the
intensification of craft activities in both urban and less agriculturally productive areas
(Blanton 1996; Nichols, et al. 2002).
Third, trade from outside the Basin of Mexico decreases. The diversity and
quantity of ceramics from outside the Basin decrease over time (Huster 2015). In
addition, the very high dominance of Pachuca obsidian (and near-exclusive secondary use
64
of the Otumba source) at Late Postclassic sites in the Basin (Golitko and Feinman 2015;
Pastrana Cruz 2007) can be considered another aspect of this reduction in external trade.
This point has not traditionally been considered, but is an important corollary to the
increase in goods exported out of the Basin of Mexico during this time.
Culturally, the Basin of Mexico becomes increasingly culturally homogenous
over the course of the Postclassic and the high degree of pre-imperial cultural interaction
within the Basin of Mexico makes it difficult to distinguish cultural changes resulting
from imperial actions. In addition, the Triple Alliance drew on and manipulated
preexisting symbols as a means of establishing legitimacy (Brumfiel 2007). Despite these
difficulties, the presence of the empire appears to have had little effect on the symbolic
repertoire of the average commoner household (Brumfiel 1996, 1998; Klein and Victoria
Lona 2009), with household ritual assemblages showing few connections to official state
ideology.
In one of the few case studies to address identity at the local level, genetic
research at Xaltocan has demonstrated that ethnohistorical sources likely overemphasized
the degree of site abandonment and population replacement which occurred as a result of
Aztec conquest (Mata-Míguez, et al. 2012). However, substantial population movement
did occur earlier in the site’s history providing a good benchmark for identifying
archaeological population movements within Central Mexico (Overholtzer 2014), and the
period under Aztec rule did see the active manipulation of ethnic symbols (Overholtzer
2015).
In summary, the archaeological evidence from the Basin of Mexico under Aztec
rule supports what could be considered an intermediate position between Hassig’s
65
argument for indirect rule (Hassig 1985, 1988), and Blanton and Fargher’s argument for
relatively collective rule (Blanton and Fargher 2008; Fargher and Blanton 2007). The
period under Aztec rule did produce significant changes at the level of the commoner
household. These included greater local economic integration and increased regional
specialization, both of which can be considered markers of economic growth (Millett
2001). However, this was accompanied by a reduction in the standard of living in some
areas, which would indicate that the local degree of collective action was only semieffective at protecting commoner interests. Culturally, Aztec rule did produce greater
cultural homogeneity within the Basin, but this appears to have been a side effect of
increased intraregional interaction, rather than a deliberate effect of imperial policy.
The Provinces
The provincial evidence for the effects of incorporation into the Aztec Empire is
more diverse than that seen in the Basin of Mexico, a not unexpected result of the
variability in Aztec strategies and local responses (See Figure 2.3 for site locations).
Three general trends are apparent, however. First, distance from the Basin of Mexico
played a large role in the degree of interaction, both economic and cultural. Second, when
distance is taken into account, the amount of imperial effort invested in controlling a
province was generally inversely related to the cooperativeness of the province. Third,
the local elite adoption or non-adoption of aspects of Aztec culture had a significant
influence on what commoners in the surrounding region also chose to use. Previous
studies in four regions, Morelos, Veracruz, Oaxaca, and Guerrero, can be used to
illustrate these points. Data for these regions are drawn from case studies representing a
66
wide range of pre-and post-Aztec conquest settings and balances of power, including
rural Cuexcomate and Capilco and urban Yautepec for Morelos (Smith 2006a, 2006c;
Smith, et al. 1989), urban Totogal and its surroundings, lowland secondary center
Callejon del Horno and rural areas in the Mixtequilla and urban Cuetlaxlan for Veracruz
(Garraty and Stark 2002; Ohnersorgen 2001, 2006; Skoglund, et al. 2006; Venter 2008),
Coixtlahuaca and the central Oaxaca Valley for Oaxaca (Blanton 1983; Kowalewski, et
al. 2010), and several settlements including an Aztec fortress in the Oztuma area for
Guerrero (Silverstein 2000) (Figure 2.2, Figure 2.3).
Figure 2.3 Comparative Postclassic Sites in Western Mesoamerica with prior
archaeological work discussed in this dissertation, with the territories of the Aztec
Empire shown
67
To begin, provincial regions closer to the Basin of Mexico tended to have both a
longer history of pre-imperial interaction and a higher degree of economic integration
once under Aztec rule. Sites in Morelos show that local marketing regions expanded, but
regional networks contracted in favor of increased trade with the Basin of Mexico once
the region became part of the Aztec Empire. The standard of living decreased for most
people, with the exception of nobles in the regional capital of Yautepec (Smith 2010).
This closely parallels the pattern seen in the Basin of Mexico and while technically a
province of the empire, Morelos may more accurately be considered part of the imperial
core. While sites in Morelos contain the full range of Aztec decorated vessel types as
tradewares, more distant regions, such as Veracruz, tend be limited to a few types. When
plotted against each other, the distance from the Basin of Mexico and the frequency of
Aztec ceramics form a clear drop-off curve (Sergheraert 2009; Smith 1990).
Several sites and regions present clear exceptions to this drop-off curve, however.
These can generally be described as “problem areas” for the Empire. For example, the
Oztuma region was both rebellious and on the Aztec/Tarascan frontier, resulting in the
establishment of a Nahua-Aztec colony and a border fortress. Frequencies of Aztec
ceramics are significantly higher in the fortress than in the surrounding area (Silverstein
2001). Likewise, Cuetlaxlan was a former Tlaxcallan ally, with an unstable relationship
with the Aztec Empire. Within the region, the frequency of Aztec ceramics (including
local imitations) is highest in a particular area within the capital of Cuetlaxlan, followed
by large sites generally and finally by the rural hinterland. The entire region, however,
has higher than geographically predicted frequencies (Garraty and Stark 2002;
Ohnersorgen 2006). As in Morelos, the differences in wealth become far more
68
pronounced during the Late Postclassic, with (mostly urban) elites becoming wealthier at
the expense of lower-class households. The Middle to Late Postclassic rearrangement of
settlement in the Mixtequilla may also be a result of imperial actions, though this
hypothesis remains tentative until surface remains can be better dated (Garraty and
Ohnersorgen 2009).
Third, the types of Aztec items adopted by elites were a large factor in what was
subsequently used by commoners in the same area. At Cuetlaxlan, imports are dominated
by Aztec III Black-on-Orange and Texcoco-Molded frying-pan censers (sahumadors).
These are the same types that are found in the surrounding hinterland, often as locally
produced imitations. In a negative case, most Mixtec and Zapotec elites in Oaxaca did not
adopt Aztec-style goods (Boone 1996), and neither did the surrounding commoners,
resulting in the recovery of only a few dozen Aztec sherds in the entire Oaxaca Valley
survey (Flannery and Marcus 2003; Kowalewski, et al. 2010), and only occasional Aztec
artifacts in more extensively investigated contexts (Whittington and Workinger 2015).
As a result of this variation, Aztec/provincial interactions can be grouped into
three general categories. First, there are regions, such as Morelos, which closely follow
the general Basin of Mexico pattern of increased integration into the Basin economic
system at the expense of other regions, increased local specialization, and slightly
decreasing wealth. Like the Basin of Mexico, these sites can be considered intermediate
on both the directness and collectiveness of rule which they experienced. Second, there
are provincial sites with higher levels of Aztec material culture than their surrounding
areas, such as Cuetlaxlan and Castillo de Teayo on the Gulf Coast and Oztuma on the
Tarascan frontier. These can generally be matched to historical accounts of Aztec
69
administrative centers, garrisons and/or colonies. (Interestingly, the reverse is not true –
some known garrison/colony locations cannot be located on the ground.) Economically
and culturally these sites are strongly tied to the Basin of Mexico, with a wide range of
Basin-style goods used in both high and low visibility contexts. In and of themselves,
these sites were likely directly ruled outposts of the Aztec Empire, but their degree of
leverage relative to the empire is unknown. While such sites could potentially offer a
venue for more intensive interaction with local people, this does not usually appear to
have been the case, and they cannot be considered sufficient evidence for direct rule of
broader provincial areas in the absence of other evidence. Third, there are sites, scattered
across a wide portion of the empire, which show more limited Aztec influence. These
sites include cases such as the Mixtequilla region on the Gulf Coast and Coixtlahuaca and
its surrounding area in Oaxaca. In these cases, only a narrow range of Aztec goods occur
at the provincial site and these are often accompanied by local imitations. The frequency
of Aztec goods is generally higher in urban areas, and they may be absent in surrounding
rural communities. Unfortunately, to date, none of these sites has been investigated in
sufficient detail to differentiate the pre- and post-Aztec portions of the Late Postclassic,
making specific conclusions about economic and cultural change difficult. Due to the
limited evidence for Late Postclassic Central Mexican influence in general, these sites
can be considered examples of relatively indirect rule, most likely with little collective
power relative to the empire.
70
Case Study Regional and Site Background
Calixtlahuaca and the surrounding Toluca Valley can be placed against these
three variables for empire/province interactions described above: distance, loyalty, and
elite preferences. The Toluca Valley is located immediately west of the Basin of Mexico,
making it one of the most centrally located provinces. Ethnohistoric sources document at
least one episode of rebellion and reconquest, suggesting a “problem province” status for
the region. Finally, the evidence for elite interaction appears relatively high, with the
region appearing a fair number of times in documentary histories, and Aztec style goods
found in burials near monumental architecture. Calixtlahuaca is not a clear fit within any
of the three types of Aztec/provincial interaction described above. The site lacks the
history of interaction and co-development with the Basin of Mexico characteristic of
Morelos, but also shows a higher degree of evidence for changes under Aztec rule than
most “limited influence” cases.
Ethnohistorical work on the area has drawn on both the traditional Mexica-centric
sources and a steadily increasing number of local colonial documents. The local colonial
documents from the region include both the records of extensive court cases between the
Marquesado del Valle (Hernán Cortez’s personal estate), the Spanish Crown, and local
communities (García Castro 2006; Hernández Rodríguez 2011), records of other court
cases (Ruiz Medrano and Noguez 2004), and Nahuatl-language wills spanning several
centuries (García Castro 2000; Pizzigoni 2007). As is common in documentary sources,
those for the Toluca Valley contain a number of contradictions and have been subject to
very little archaeological verification to date. In particular, the place-name Matlatzinco is
71
variably used to refer to the greater Toluca Valley as a region, the city-state of
Calixtlahuaca, and the city-state of Tollocan (Umberger 2008). Similarly, the site/citystate of Calixtlahuaca is referred to as Matlatzinco and Calixtlahuaca, both of which are
Nahuatl terms, and once, given the pre-Aztec name of Pintambati (Hernández Rodríguez
2011:189).
The Toluca Valley was a multi-lingual and probably multi-ethnic region, with
speakers of Matlatzinca, Mazahua, Otomi, and Nahuatl all recorded in the early colonial
Relaciones Geographicas for the region (García Castro 1999). Basalenque (1975
[1642]:prologue) lists three group names for the Matlatzinca who lived in the Toluca
Valle: Nentambati (“those from the center of the valley”), Nepyntatuhui (“those from the
land of corn”), and Matlatzingos (“those who make nets”, but this is a Mexica name). He
also lists Pirindas and Charenses as names for those Matlatzinca living within the
boundaries of the Tarascan Empire, outside of the Toluca Valley proper, demonstrating
the fuzzy relationship between location of residence, language, ethnicity, and terms for
group identity in Postclassic Central Mexico. The Florentine Codex describes multiple
named groups in the Toluca Valley, but describes their lifeways collectively, suggesting
that cultural differences among such groups were subtle (Sahagún 1950-82: Book 10:
Chapter 29).
During the Early Postclassic, the Toluca Valley follows the same general
trajectory as much of central Mexico, with mention of Toltec, Chichimec, and
unspecified previous occupants variously intermarrying and fighting with each other and
outsiders. According to (Alva Ixtlilxochitl 1975-77 [1600-1640]:53), the rulers of Tula
72
sent children to marry into families in the Toluca Valley, though this may be part of the
later ideological justification for bringing the area under Aztec rule.
By the Middle Postclassic, Calixtlahuaca was the dominant city-state in the
Toluca Valley, based on textual sources (Tomaszewski and Smith 2011). If this is the
case, the site rose to prominence quickly as modern archaeological work places its
foundation during this same time period (Huster and Smith 2015). This contrasts with
García Payón (1956/57), who had argued for a site chronology beginning during the
Classic period. While there were some Classic and Epiclassic vessels recovered during
his excavations, they are from offering contexts and, given the lack of other evidence for
an earlier occupation of the site, are best interpreted as heirloom objects. This period may
also see the origin of the Chimalli (shield) dynasty, centered at Calixtlahuaca. Multiple
members of the Late Postclassic and Early Colonial native nobility in the Toluca Valley
had surnames incorporating the term chimalli, and there is a more generalized reference
to the “Chimallis of Calixtlahuaca” in one court case (Hernández Rodríguez 2011).
Umberger has argued that this term is associated with the bird image often depicted on
shields in Matlatzinca style sculptures and reliefs at Calixtlahuaca, and less frequently on
portable stone items at other sites in the region. At this time, there were three “lords of
the Matlatzinca” (Zorita 1963 [1566-1585]:194-200), though scholars disagree as which
site or sites they may have ruled (García Castro 1999:53-56). Calixtlahuaca/Tenango/
Malinalco, Calixtlahuaca/Toluca/Tenango, and Calixtlahuaca/Toluca/Tenancingo have
all been proposed as possible sets of candidates.
During the 14th century, the northeastern portion of the Toluca Valley was briefly
incorporated in the Tepanec state before regaining independence in the turmoil
73
surrounding the formation of Triple Alliance (Hernández Rodríguez 1988). While sites in
the northwestern Toluca Valley are more clearly subordinate to the Tepanec state, the
central valley is only mentioned as providing tribute twice – referred to once as Tolocan
in the Anales Tepanecas, and once as Matlatzinco in the Carta de Azcapotzalco de 1561
(Santamarina Novillo 2006:509-511). Given that both references list one toponym but not
the other, they most likely refer to the same place, and Calixtlahuaca is a strong
contender for the location in question. If this is the case, there was likely a short period of
time during which Calixtlahuaca was at least nominally subordinate to the Tepanec state.
In the 1470s the Toluca Valley was conquered by the Triple Alliance in an effort
to form a buffer zone between the expanding Aztec and Tarascan Empires and to acquire
access to a productive maize-growing area (García Castro 1999:58). The conquest
occurred under the Mexica rulers Tizoc and Axayácatl (Berdan and Anawalt 1992
[1541]; Umberger 2008), with further campaigns to deal with rebellion under Ahuitzotl
and Moctezuma II (Chimalpahin 1965 [1606-1631]; Cuauhtitlán 1985 [1606-1631]) .
During the conquest, the Aztecs were aided by the ruler of Tollocan (modern Toluca),
against the ruler of Calixtlahuaca. Of the three “Lords of the Matlatzinca” at this time,
only Chimaltzín of Toluca survived.
Following the Aztec victory under Axayácatl, many Matlatzinca (traditional
residents of the Toluca Valley region) fled west into Tarascan territory, and lands in the
Toluca Valley were divided between Tollocan, Basin city-states that had assisted in the
conquest, and high-ranking Aztec individuals. Calixtlahuaca fell under the direct control
of Tenochtitlán in this subdivision. Lands left empty by fleeing Matlatzinca were
resettled by large numbers of agriculturalists from the Basin of Mexico (Carrasco
74
1950:277-279; Cuauhtitlán 1985 [1606-1631]:57; Zorita 1963 [1566-1585]:22,263ff,
266). Settlers came from the particular Basin cities that had been granted control over a
particular Toluca Valley town, resulting in a mosaic of communities occupied by
Matlatzinca and groups from different Basin of Mexico source communities.
Officials of the Triple Alliance were established in both Calixtlahuaca and
Tollocan (modern Toluca), but the latter became the primary tribute collection point,
creating a reversal of the pre-Aztec local balance of power (Alva Ixtlilxochitl 1975-77
[1600-1640]:2:145; Chimalpahin 1965 [1606-1631]:105-107), a situation that continues
to the present day. Hernández Rodríguez (2011) argues that this power shift was largely
political and that Calixtlahuaca maintained much of its religious importance. The primary
deity image from Matlatzinco (in this case probably Calixtlahuaca) was removed to
Tenochtitlán, where it was kept in its own temple (Durán 1951 [1581]:2:272; Sahagún
1950-82:2:171-172).
Based on recent survey work, Calixtlahuaca covered 234 hectares at its maximum
extent, making it relatively large for a Postclassic urban center (Smith 2008). Recent
radiocarbon dates demonstrate that the site was founded around AD 1130 and abandoned
by 1530, shortly after the Spanish conquest. The ceramic chronology for the site allows
this span to be divided into three phases, described in more detail in Chapter 3 (Huster
and Smith 2015). The site provides an interesting contrast between local and Aztec
cultural traditions. On one hand, it features a highly distinctive local ceramic tradition,
has limited Aztec imports in domestic contexts, and has a distinctive of a single
monumental architectural core. This is juxtaposed against distinctively Aztec-style
architectural elements in the last phase of monumental construction (Sergheraert 2011)
75
and, regionally, the largest collection of Aztec-style sculpture in the provinces (Umberger
1996).
Even based on this brief summary, it is clear that the ethnohistory of the Toluca
Valley provides examples of multiple imperial and provincial strategies. Imperial actions
include intermarriage and alliances with local nobility, the rearrangement of economic
systems, propaganda in the form of capturing and replacing the central deity statue, and
the introduction of colonists. Local responses include bolstering, migration, and
resistance. Archaeology offers a means of identifying additional strategies and measuring
the effects of those already described.
76
CHAPTER 3
CHAPTER 3. SITE BACKGROUND AND PROJECT METHODS
This project is based on data from the site of Calixtlahuaca in the Toluca Valley
of central highland Mexico. Calixtlahuaca was a Middle-Late Postclassic (AD 11301530) city-state capital. My research for this project occurred in conjunction with the
broader Calixtlahuaca Archaeological Project (CAP), a multi-year investigation of urban
processes and political change at the site directed by Dr. Michael E. Smith. This chapter
describes the physical setting of the site, field and lab work by the CAP, specialized
artifact samples and analyses used in later chapters, and the development of the site
chronology.
Geology and Environment
The Toluca Valley (sometimes called the Upper Lerma) is located immediately
west of the Basin of Mexico, in the Central Mexican highlands. It is slightly higher than
the Basin of Mexico, resulting in both cooler temperatures and higher rainfall. It is part of
the Neovolcanic axis, and the landscape is dominated by the Nevado de Toluca, Mexico’s
fourth highest mountain. The Toluca Valley forms the headwaters of the Rio Lerma,
which eventually drains much of west Mexico. The geology is primarily igneous, with
both tertiary and quaternary formations. Temperatures are lowest during January, and
nighttime freezes are not uncommon. The area receives an average of 29.4 inches of rain
annually, with the majority falling between June and September. During portions of
prehistory with higher than modern rainfall levels, parts of the Toluca Valley were
77
covered with shallow lakes and/or marshes, but these features never dominated settlement
and transport patterns to the degree that the lakes in the center of the Basin of Mexico did
for that region. Human occupation of the Toluca Valley began as early as the Archaic,
but settlement density remained light until the collapse of Teotihuacan (Sugiura
Yamamoto 1998). Stylistic similarities and chemical sourcing of artifacts show
fluctuating ties to central and west Mexico (Kabata 2010; Sugiura Yamamoto 2005).
The archaeological site of Calixtlahuaca is mostly located within the boundaries
of the lands of the modern village of San Francisco Calixtlahuaca, in the Municipio of
Toluca, State of Mexico, Mexico (Figure 3.1). The site area is bounded to the south by
the Parque Sierra Morelos, to the east by the village of San Marcos , to the north by the
village of San Francisco Calixtlahuaca, and to the west by the ejido lands belonging to
the community of Tecaxic. The site covers a primary hill, Cerro Tenismo (Tenhizo on
some maps), and extends across a smaller secondary hill to the east, Cerro San Marcos.
The central portion of the site is an official Instituto Nacional de Arqueologia e Historía
(INAH) archaeological zone. There is a small on-site museum, managed by the Municipo
of Toluca. Much of the site (including most of the INAH zone) is held as “Tierras
Comunales” by the residents of San Francisco Calixtlahuaca. For practical purposes, this
is similar to an ejido, with land title controlled communally, but with particular families
or individuals having usufruct rights to particular plots. However, in a unique situation,
the land is owned by a communal village organization, rather than by the federal
government. Based on pedestrian survey by the Calixtlahuaca Archaeological Project,
evidence for ancient occupation extends well beyond the official site zone boundaries,
covering approximately 264 hectares (Smith, et al. 2009).
78
Figure 3.1 Map of the site of Calixtlahuaca showing site boundaries, monumental
architectural groups, and excavation unit locations
79
The site’s environment is characteristic of highland Central Mexico. Cerro
Tenismo ranges from approximately 2650 to 2920 masl. The site receives around 800mm
of rainfall annually, with the majority falling during the July-September rainy season.
The lower two thirds of the site are currently used for subsistence maize milpa and
maguey cultivation. The upper portions are used primarily for grazing (mostly sheep,
with some cattle and turkeys), though evidence of past cultivation and discussion with
local residents suggest that this is due to decreased interest in subsistence agriculture
among younger villagers, rather than any inherent unsuitability of the land for agriculture.
There are tejocote (crabapple) and capulin (wild cherry) fruit trees on the hill, which are
harvested opportunistically. The village of San Francisco Calixtlahuaca is steadily
encroaching on the site, even within the officially designated archaeological zone.
Site Description
As noted above, the archaeological site of Calixtlahuaca covered a maximum
extent of about 264 hectares. The site contains a number of monumental structures,
originally located, mapped and excavated by José García Payón. In contrast to most
Postclassic cities, which feature a single monumental core, Calixtlahuaca’s monumental
architecture is scattered from the base to the summit of Cerro Tenismo (Figure 3.1). This
arrangement of two-to-four structure clusters is likely a result of the hillside nature of the
site, where creating sufficiently large, flat, terraced areas would have required more labor
than the construction of the monumental architecture itself. The monumental architecture
includes a palace (Figure 3.1 A, known locally as the Calmecac), a round pyramid
80
(Figure 3.1 C, Figure 3.2), a two pyramid and altar complex (Figure 3.1 D, Figure 3.3), a
probable elite residence (Figure 3.1 F, known locally as the Panteon due to the number of
burials encountered there), and several other unrestored mounds or mound groups. The
results of García Payón’s excavations of the monumental architecture were partially
published in a number of works (García Payón 1936, 1938, 1941a, 1941b, 1956/57, 1979,
1981), but the specific details of what artifacts came from which monumental structure
are decidedly fuzzy (Smith 2003e). What can be ascertained is that many of the
monumental structures had multiple (2-4) construction phases, the last of which is likely
associated with the Aztec occupation of the site, due to differences in stone selection and
construction techniques (Sergheraert 2011). Sculptures recovered at the site include both
pieces in the imperial Aztec style and pieces in a distinctly local style (Umberger and
Hernández Fahan submitted 2014). The majority of the surviving artifacts from García
Payón’s excavations are stored at the Centro Cultural Mexiquense, on the outskirts of the
city of Toluca.
Outside of the complexes of monumental architecture, the majority of the site was
covered with terraces, featuring a mixture of residential architecture and agricultural
fields. The ancient terraces covering the majority of the site were narrow and connected
to a system of large paved drains to prevent erosion during the rainy season (Borejsza, et
al. n.d.). Soil on the terraces was heavily manipulated by humans in order to produce
consistent textural characteristics beneficial for agriculture. The Postclassic terrace
system collapsed after the site was abandoned following the Spanish conquest, and the
modern terraces in use on the hill offer only the most general outline of the ancient
system. The majority of the Postclassic houses excavated at the site were one or two
81
room structures, built of wattle and daub on stone wall foundations (Figure 3.4) Some
houses may have had adobe walls. Exterior pavements were common and some small
interior rooms were also paved. Unlike in many Postclassic Central Mexican urban
settlements, houses occur singly, rather than being grouped into compounds or clusters
around a central courtyard. Like the lack of a single monumental core, this pattern may
be result of the effort needed to produce large level areas on a steep hillslope.
Figure 3.2 The round pyramid (Group C) at Calixtlahuaca
82
Figure 3.3 Monumental Group D at Calixtlahuaca
Figure 3.4 Typical house at Calixtlahuaca (Excavation Unit 309)
83
Archaeological Work
Archaeological work in the Toluca Valley has been relatively limited to date,
especially for the Postclassic period. Sugiura (2000) has surveyed the majority of region,
but she is just beginning to publish the Postclassic portion of the resulting dataset (e.g.
Sugiura Yamamoto 2011). All of the Late Postclassic towns listed in the Toluca Valley
provinces of the Matricula de Tributos can be confidently matched to geographic
locations (Smith and Berdan 1996a). However, other than Calixtlahuaca itself, only three
other Postclassic sites in the Toluca Valley, Huamango, Teotenango, and Cerro Toloche
have been subject to systematic excavation. In the cases of the first two of these sites,
excavation focused primarily on the Early Postclassic monumental cores of the sites (Piña
Chán 1975, 1981). The third site, Cerro Toloche, is only located a few miles from
Calixtlahuaca but would have been part of the prehispanic city of Tollocan (Toluca).
Work at Cerro Toloche began after the completion of the Calixtlahuaca project and
artifact analyses are still underway at this time (Jaramillo Lunque and De la Peña Virches
2012, 2014). Based on initial results, the site has a very similar artifact assemblage to
what has been found at Calixtlahuaca, and future comparisons will be informative. A few
other sites in the broader Toluca Valley and surrounding areas have also been excavated
during salvage work, including the Cerro de los Magueyes at Metepec (Carbajal Correa
and González Miranda 2003), and various excavations in Valle de Bravo (Murillo
Rodríguez 2002; Reinhold 1981). Unfortunately, these projects have not been
comprehensively published, making comparisons difficult.
84
Calixtlahuaca itself has long been recognized as an archaeological site, and was a
regular stop on the 19th century Central Mexican antiquities collecting circuit (Huster
2013). More formal archaeological work at the site has consisted of the excavations of
José García Payón on the monumental architecture from 1930-1938 (García Payón 1932,
1936, 1941a, 1941b, 1956/57, 1979), various consolidation projects by INAH
(Villanueva Villalpando 1999; Zúñiga Bárcenas 1992), and the current Calixtlahuaca
Archaeological Project (CAP) (Smith 2006b, 2011). The García Payón project mapped
the site, and excavated and restored approximately two thirds of the monumental
structures at the site. These include the royal palace, several temple complexes, and
several architectural groups of unknown function.
Beginning in 2006, the modern Calixtlahuaca Archaeological Project collected
data from household and terrace contexts, as a complement to the previous work on
monumental architecture at the site. The project consisted of one summer survey season,
one six-month excavation season and five subsequent summer lab seasons. The survey
established the site boundaries and produced a set of systematic surface collections. The
survey material has been analyzed by Novic (2015) for information on neighborhoods
and social clustering within the site. The 2007 season featured excavations in twentyseven areas scattered across the core of the site (Figure 3.1). The targeted goals of the
excavations were approximately evenly divided between domestic contexts and broader
terracing and land modification processes. Thirteen of the excavations produced middens
and/or domestic architecture with associated refuse scatters. Once the various
stratigraphic levels within the excavations were dated, several proved to have
components dating to multiple phases of the site’s occupation, providing a total of
85
eighteen chronologically distinct household components for analysis (Table 3.1). During
the subsequent lab seasons, all artifacts recovered from domestic contexts were cataloged
and subjected to basic visual classification. I did four additional months of attribute
analysis in the lab during fall 2011 specifically for this dissertation.
Unit
307
315
316
320
323
324
303
307
308
311
316
322
307
309
316
317
324
327
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Yata
Yata
Yata
Yata
Yata
Yata
DS-1 Contexts
N. Lots M³ Excav. N. Sherds
6
0.87
4,770
24
5.29
13,890
11
2.34
3,050
9
2.02
3,840
13
3.91
8,915
3
0.59
914
7
1.75
8,951
5
1.98
20,280
5
0.76
3,729
13
3.08
5,030
27
9.46
15,800
4
0.91
1,668
11
3.47
10,200
9
4.08
3,094
7
3.89
7,451
14
5.42
9,638
6
1.76
3,438
4
1.26
948
DS-2 Contexts
N. Lots M³ Excav. N. Sherds
9
2.49
5,810
41
10.50
16,775
17
4.77
4,710
37
8.03
12,189
44
14.05
26,947
3
0.59
914
8
1.75
9,043
10
3.83
22,330
9
1.35
4,359
22
6.22
7,838
47
13.76
22,563
7
1.52
1,855
13
4.41
10,257
17
5.75
4,217
17
6.08
10,091
25
6.64
10,860
6
1.76
3,438
8
2.96
1,266
Table 3.1 Excavated household components in the DS-1 (core household) and DS-2
(extended household) samples with lot counts, volume excavated, and sherd counts.
86
Field Methods
During the 2006 survey season, the project worked out of the site museum. The
immediate goals of the survey were to establish the boundaries of the site, locate potential
areas to excavate during the subsequent season, and identify broad spatial patterning in
artifacts. In addition to providing information on the spatial layout of the site, this season
also allowed the project to adjust the previously developed whole-vessel ceramic
classification system for use with sherds (Smith 2003e). Details of the survey methods
and findings can be found in the survey project report (Smith 2006b) and associated
articles (Smith, et al. 2009; Tomaszewski 2006).
Beginning in 2007, the immediate excavation goal was to generate a sample of
house and terrace contexts from across the site, in order to better understand the
occupational history of the site as an urban center. This resulted in the placement of 27
excavation units3 across the site, labelled as Units 303 through 329. Unit placement was
based on a combination of the archaeological potential of a given section of the site, cross
referenced against landholder attitudes toward archaeological investigation.
3
The CAP used the terms Unit-Locus-Lot to identify both survey and excavation contexts, rather than the
more common Operation-Unit-Lot. This work uses the project terminology, so “Unit” is a major excavated
area, such as a house excavation or several test pits within a single field. Unit codes are also used to refer to
several other non-excavated contexts, including “300” for all survey collections, “301” for a profile drawn
during the survey, and “302” for donations by local residents. “Locus” refers to the grid square within the
unit. Excavated loci were of variable dimensions, but usually covered between one and four square meters.
Architectural features were also assigned locus numbers if they were excavated apart from the grid system.
In the survey, locus is used to refer to a single surface collection. “Lot” refers to the excavated level within
the locus, beginning with 1 at the surface. Lot depth is variable, using a combination of arbitrary and
stratigraphic breaks, so a given lot number may refer to variable depths in different loci. For further
descriptions of excavation procedures, see Smith 2011.
87
Approximately half the units were targeted toward locating domestic structures and half
were focused on aspects of the terraces and drain system. About one third of each
category did not locate its targeted objective, and units intended to investigate one topic
sometimes located features of interest to the other and were expanded accordingly. Each
unit was excavated by a team of one or two archaeologists and 4-10 local workers, and
checked at least once a day by Dr. Smith. I personally supervised the excavation of units
304, 306, 309, 313, 317, 324, and 325, and was able to observe the remainder of the
house excavations on a regular basis.
Within each house-focused unit, excavation began with a trench through the
highest probability area. If this trench encountered architecture or midden deposits,
additional loci were added until an area one to two meters around all architectural
remains was completely exposed. Once the architecture was exposed, a limited number of
units would be further excavated to expose any deeper layers of cultural material. When
possible, a two-meter zone around the architecture was cleared down to just below the
Postclassic ground level in an effort to locate midden deposits, or, if no clear midden
could be delimited, to produce an adequate sample of medium-to-high density secondary
refuse. Middens were excavated until they reached sterile. Within each locus, lots were
removed in 10 or 20 cm arbitrary levels unless a change in stratigraphy was observed.
Excavated soil was screened through ¼ inch mesh at the discretion of the lead
archaeologist for the unit, and the screened/not screened status of each lot was recorded.
For most house excavations, plowzone lots were not screened but most lots associated
with architecture and all lots from middens were screened. For unscreened lots, workers
removed and bagged all artifacts observed during excavation. Details of the excavation
88
method and findings can be found in the project report (Smith 2011) and associated
articles (Smith, et al. 2013).
All excavated artifacts were initially processed at the field lab in San Francisco
Calixtlahuaca. All artifacts were washed and the very high priority ceramic lots and all
lithics were given a basic analysis. Following the end of the excavation season, all of the
project artifacts were moved to a lab facility at the Colegio Mexiquense, in the nearby
town of Zinacantepec. All further artifact analyses, which took place between 2008 and
2012, took place at this lab, and are discussed as appropriate below.
Chronology
One of the primary data analyses for the CAP was the establishment of a site
chronology. Because of the relatively limited previous work in the Toluca Valley, most
local ceramic types did not have known date ranges other than “Postclassic,” limiting the
possibility of using marker types to date particular components. The full process of
developing the site chronology is presented in Huster and Smith (2015), and I present a
brief summary of the results here. We first created three clusters of ceramic lots, based on
discriminate analysis of ceramic lots, with the initial seed lots for each group selected
based on ceramic cross-ties to other regions. These groups were confirmed to be
chronological due to their stratigraphic consistency across the site. We assigned absolute
dates to the three groups based on a series of 54 radiocarbon dates. These dates were used
to generate phase date ranges using Bayesian analysis, and further confirmed the
temporal distinctiveness of the three ceramic groupings.
89
Based on ceramic seriation and the Bayesian analysis of associated radiocarbon
dates, Calixtlahuaca was occupied from AD 1130 to 1530. The occupation of the site can
be subdivided into three phases, Dongu (AD 1130-1380), Ninupi (AD 1380-1450) and
Yata (AD 1450-1530) (Huster and Smith 2015). These were referred to as Phases 2, 4
and 6 during analysis and the numbered labels are retained for identifying household
components by phase. On a regional scale, these can be considered equivalent to Middle
Postclassic, Late Postclassic-A, and Late Postclassic-B phases at other Central Mexican
sites.
Samples
The Calixtlahuaca Archaeological Project excavations included over one and a
half thousand individual lots. These recovered approximately half a million sherds as
well as smaller numbers of artifacts of other classes. These artifacts come from a range of
contexts representing both different original contexts of use and deposition, and different
formation processes in the time between deposition and excavation. As a result, the
project defined five potential samples, referred to as DS-1 through DS-5, of which three
are used in this dissertation (Figure 3.5).
90
Figure 3.5 Schematic of the DS-1 (core household sample) to DS-5 (all excavated
contexts) samples of contexts excavated by the Calixtlahuaca Archaeological Project. N
values are the number of excavated lots in the sample.
The samples begin with the DS-5 sample. This sample includes all of the artifacts
from all of the project’s excavations. This encompasses a wide range of contexts,
featuring everything from middens to slopewash. It also includes lots that received a
range of post-excavation processing, ranging from a brief looking over before discard to
full artifact classification. I use this sample when I am discussing the range of variation
present in a particular artifact class at the site, because good examples of particular
artifact types may not have come from the most secure contexts.
The DS-4 and DS-3 samples were generally not used in this dissertation. The DS4 sample consists of all of the lots at the site with completely classified ceramics. The
DS-3 sample consists of all of the lots at the site which could be confidently assigned to
91
one of the three primary phases of occupation. Many of these lots do not have completely
classified ceramics. The DS-4 and DS-3 samples partially overlap each other; there are
phased lots without completely classified ceramics and lots where the classified ceramics
could not be assigned to a phase.
The next sample, the DS-2, is an extended household sample. It includes all
securely dated lots peripherally or well associated with household contexts. As a result,
associations are somewhat more tentative than in the more restricted DS-1 sample. I use
this sample for analyses of rare artifact classes, such as figurines or jewelry, where using
a broader sample is the only way to establish a reasonable sample size.
The most commonly used sample of contexts is the DS-1, or core domestic
context sample. A subset of the Calixtlahuaca Archaeological Project excavations was
designated as the domestic context sample and is generally used as the core sample by all
project researchers focusing on household information (as opposed to terraces and
broader landscape use). The domestic context sample consists of lots with a good
chronological assignment to a single phase and that are clearly associated with house
architecture, from clear midden deposits, and/or from unassociated high-density deposits
that did not show signs of significant redeposition. It does not include plow zone or other
surface disturbed contexts, terrace fill or gully fill. The sample includes lots from thirteen
excavation units. Because many of the units have more than one temporal component,
there are eighteen discreet components within the thirteen excavation units. Two units are
continuously occupied through all three phases, one included portions of unrelated Dongu
and Yata phase occupations, and the remainder are single-phase occupations. There are
six components dating to each of the three phases. The unit locations cover a broad swath
92
of the site core, from the valley floor to the upper hillslopes, including materials from
multiple neighborhoods in each phase. None of the components with architecture are
highly distinctive on the basis of architectural style or house size, suggesting that the
sample does not include any elite households. (For a further discussion of household
status and wealth, see chapter 6.) For the purposes of my analysis, I treat all of the DS-1
lots dating to a single phase in a given excavated unit as a single analytical unit, referred
to as a household component. This means each household component usually includes
materials from more than one stratigraphic context. This lumping is necessary to provide
sufficient sample sizes for many of the smaller components.
Each component should be interpreted as a household series (sensu Smith 1992b),
rather than a household in the ethnographic sense. While the ethnographically observable
household is a single family grouping, which may grow, shrink, and reform over time, the
archaeological household is a composite view of a series of households over a time span
much longer than the life of any one individual member of the household. As a result, the
archaeological remains in each household component are an average of activities over
multiple generations.
Analyses
The work in this dissertation draws on four major ceramic datasets and references
a number of additional analyses of specific ceramic artifact types (i.e., figurines, spindle
whorls) or non-ceramic artifacts (i.e., chipped, ground stone). The following sections
93
provide details on the nuts-and-bolts of classification, sampling and recording
procedures.
General Ceramic Classification
All of the ceramics for the DS-1, DS-2, and DS-4 samples were given a basic
classification during the 2007 excavation season or the 2008-2012 lab seasons. The
classification system used by the project is based on a combination of vessel form and
decoration. Paste is generally only factored into the classification system as a secondary
means of identifying certain imported types, such as Aztec Black-on-Orange wares.
Local ceramics lack clear differentiation in pastes, with most being made from a mediumcoarse buff paste.
The ceramic classification system divides ceramic sherds into six major form
categories, cross-cut by decorative families. As a result, “type” refers to a specific
decorative pattern for a specific vessel form. The form based categories are bowls
(including simple, tripod, and molcajete bowls), jars (necked vessels without spouts),
basins (large open vessels), comals (tortilla griddles), other vessel forms (a miscellaneous
category including freestanding censers, sahumadors, spinning bowls, pitchers, copas,
and other rare forms), and non-vessel ceramics (figurines, spindle whorls, beads, pipes,
worked sherds, etc). Major decorative groupings are referred to as families. Decorative
groups identified from the regional review of museum collections are named using the
letters A through H and J to K, though the latter two families did not occur in the
excavated project material. In addition, comparable groupings of material from the Basin
of Mexico, such as Aztec Black-on-Orange wares and Black-on-Red wares are also
94
referred to as decorative families. Decorative families are usually based on color
combinations; thus all Family C types are polychrome on white, while all E types are redon-natural. A full listing of families and illustrations of representative sherds can be
found in Chapter 8. Within each family, specific decorative types are numbered, so
sherds are recorded as A-1, A-2, B-1, B-2 and so on. While the “0” number for a family
is always “Family known, specific decorative type unknown”, all other numbers are
unique to the family and an A-1 has no particular relationship to a B-1. About one third
of the vessel types in the classification system fall outside the decorative family system.
These are either non-local types (e.g., Chalco-Cholula Polychrome), or
undecorated/simply decorated pieces with distinctive forms (e.g., Biconical censers).
Within each type, sherds were identified as rims, molcajetes (grater bowls), appendages,
or bodies. The general ceramic classification system used at the site was developed prior
to the beginning of the project by Smith (2003e) for recording whole vessels in museums
around the Toluca Valley. It was subsequently amended in 2006 for use on sherds and
continued to be amended as necessary throughout the lab seasons. When type definitions
were changed, all potentially affected previously classified sherds were reclassified to
insure consistency throughout the dataset. Any lots from which I drew the random sample
for attribute coding in 2011 were briefly reclassified at the time of sampling to insure
standardization. The final dataset includes between 900 and 20,000 classified sherds per
DS-1 domestic context component. Copies of the type list and lot coding sheet may be
found in Appendix 1 and a more detailed ceramic coding guide will be uploaded to the
Digital Archaeological Record (tDAR, www.tdar.org) as part of the project’s data
management plan.
95
The majority of the general ceramic classification was completed by lab workers
from the village of San Francisco Calixtlahuaca. Once they completed a provisional
classification, the ceramic lot was checked by one of the season’s senior project members
(Dr. Michael Smith, Cynthia Heath-Smith, Juliana Novic, or myself, depending on the
season). In cases where someone other than one of the lab workers did the initial
classification, the lot was still checked by a second person to insure consistency. The
resulting data were entered into an Access database in the field.
I use the general ceramic classification data in analyses in four chapters. When I
am using all classified sherds of all types, even for limited samples of contexts, this is my
largest dataset and allows for strong conclusions. All DS-1 household components have
at least 900 sherds. In my discussion of trade in Chapter 4, I analyze all DS-1 sherds by
likely source region. Similarly, in my discussion of wealth in Chapter 5, I use either all
sherds or all non-eroded sherds as my sample for analysis. In other cases, I use subsets of
the classified ceramics. These include ritual vessels in Chapter 7 and separate analyses of
cooking and serving vessels in Chapter 8. In order to maintain adequate sample size, I use
total sherd counts for all intra-site comparisons and some inter-site comparisons. When
performing comparisons with other sites where ceramics were reported using rim-sherd
only classifications, and where this would result in sample of less than ten sherds for a
given household at Calixtlahuaca, I either remove the household from comparison, or
compare the Calixtlahuaca households as a single phase-based total.
Attribute Analysis
96
The second major dataset collected and used in this dissertation was a more
detailed attribute analysis of a sample of sherds from each domestic component. The
sample consisted of a randomly selected sample of up to 200 rim, 50 appendage
(handle/support), and 30 interior-incised sherds per component. Attribute data consisted
of up to nine items for each sherd, including detailed vessel form, rim and appendage
form, rim thickness and diameter, decorative motifs, paste type, and incision patterns in
grinding bowls (see Appendix 2 for code descriptions and sample coding sheets). These
attributes encompass a range of functionally equivalent variation in technological style,
likely resulting from learned differences in production techniques (Hegmon, et al. 2000;
Lemmonier 1986). However, because later analyses showed that many of the attributes
recorded in the field did not correspond to the groups produced by INAA and ceramic
petrography, much of the collected attribute data were not included in this dissertation.
Sherds were selected for attribute analysis as follows. For each component, lots
were randomly selected until reaching either 5000 total sherds, or three lots, whichever
involved a larger number of lots. Lots that had previously had samples taken for thin
section petrography were selected only after all other lots in the component to avoid the
difficulty of reuniting sherds that were boxed for export. For each component, the three
selected lots with the highest sherd counts were weighed by type. If the three largest lots
were all from the same unit, the smallest was replaced with the largest lot from a different
unit (see Appendix 2 for lots selected for each component). This weighing provided data
on fragmentation rates by type.
Within the selected lots for a component, the rim sherds, appendages, and incised
pieces (molcajetes and ollas) were each laid out as a group. Using a list of random
97
numbers generated for the range necessary by http://www.graphpad.com/quickcalcs/
randomn1.cfm, with numbers put in order and duplicates removed, I then counted from
the first sherd, taking for attribute analysis those that corresponded to the numbers on the
randomly produced list. Counting continued directly from one lot to the next. Pieces
belonging to more than one category were placed with rims if including a rim and
anything else, and as supports if molcajetes and support were both present. If selected,
however, all applicable attributes were recorded. If there were fewer sherds of a given
group in the DS-1 component than the total to be sampled, attributes were recorded for all
sherds present, but additional sherds were not added from other non-DS-1 lots. This
occurred most often when sampling incised sherds.
All of the original attribute analysis was completed between August and
December, 2011. With the exception of three household components where Rosario
Endañu, a Mexican archaeology student, assisted with recording less subjective variables
(paste, sherd thickness, vessel diameter, and rim-arc percentage), I personally did all of
the analysis. I revised the attribute coding for the first five classified components in
summer 2012 to insure consistency. The resulting data were entered into an Access
database by student volunteers at Arizona State University. A revision of 10% of the
entries found errors in less than .5% immediately after data entry, and this was further
reduced by ongoing checking and correction of clearly anomalous cases during analysis.
The attribute data is a reasonably large sample per household, and has the
advantage of being a randomly selected and consistently sized sample for each household
component. In chapter 5 I use attribute data on rim diameter, thickness, and lip form from
three common vessel forms to address the standardization of ceramic production by
98
phase. In chapter 6 I also use attribute information on average rim arc percentage by
household component to compensate for variation in formation processes when
calculating vessel equivalents as a component of household wealth.
Instrumental Neutron Activation Analysis
Neutron Activation Analysis is considered the most accurate method for
identifying the bulk chemical composition of artifacts. It functions by bombarding a
sample of the artifact (for ceramics, a powdered sample) with neutrons and then
measuring the resulting decay of the radioactive isotopes formed. At the Missouri
University Research Reactor, samples are subjected to both a short and long irradiation
cycle, allowing for the identification of 27-33 elements. The method has been widely
used in Mesoamerica. Within the Basin of Mexico, where there is long history of using
the technique and a strong comparative database of sherds and clay samples. These data
have been used to distinguish at least six major source groups, with eighteen possible
subgroups (Crider 2011:80). The areas to the north (Tula), east (Basin of Mexico), and
southeast (Morelos) of the Toluca Valley are represented in MURR’s comparative
database, allowing for the identification of sherds from these areas. However, previous
work in the Toluca Valley itself is limited to a single study of 64 sherds from the site of
Tlacotepec (McVicker, et al. n.d.) and a small, currently unpublished sample of
Epiclassic sherds from the southern Toluca Valley. As a result, geographic source group
assignments are stronger for sherds imported from outside the Toluca Valley.
A stratified random subsample of thirty sherds from seventeen of the eighteen
domestic components was selected for INAA analysis at the Missouri University
99
Research Reactor (see Appendix 3 for INAA samples). (The eighteenth household
component, 327-Ph6, was identified and defined after INAA samples had been
submitted.) The rim sherds from the attribute analysis of each domestic component were
divided into bowl, jar and other vessel categories, and sherds too small for INAA (2 cm2)
were removed. Because there is little apparent variation in breakage rates due to
differences in paste or vessel form at the site, the removal of small sherds was not
expected to introduce significant bias into the sample. A sample of 15 bowl, 10 jar, and 5
other vessel rims was then selected from the remaining adequately-sized sherds using the
same procedure as the attribute random sampling. Sherds were assigned to stratification
groups based on their primary vessel form code from the attribute analysis, with tripod
bowls, molcajetes, and simple bowls stratified as bowls, jars sorted as jars, and all other
vessel forms placed in the other vessel stratification category. This produced a very close
correspondence with the broad vessel form categories used in the general ceramic
classification, with two specific exceptions. Thick-rim vessels, listed as an “other vessel
form” in general classification, were sampled with the jars. Decorated basins, which were
included with bowls of the same decorative pattern in the general classification, were
sampled with other vessels for INAA. Due to the reassessment of the site chronology
between the time of sample selection and the writing of this dissertation two household
components, 303-Ph2 and 311-Ph4, ended up with double samples of 60 sherds. In
addition, the broader CAP also submitted 150 sherds for INAA. These sherds were
selected by type from types suspected to be non-local, and greatly improved MURR’s
ability to identify rare, imported, source groups.
100
I use the INAA source group assignments provided by MURR. These include
nine groups, referred to as Groups 1-9, and a remaining unknown fraction, referred to as
Group Unknown. One group, Group 9, consists exclusively of sherds from a single
component, including types that should not otherwise cluster together. As a result, it is
considered a result of contamination, either prior to excavation, or during processing and
is excluded from analysis. Of the remaining groups, three (Groups 1-3) are generally
associated with the Toluca Valley, three are associated with the Basin of Mexico (Groups
4-6), and two are associated with areas to the south or southwest of the Toluca Valley
(Groups 7 and 8). These groups are described in more detail in Chapter 4.
The INAA sample had two primary functions. The first was, as a random sample, to
provide an independent measure of ceramic variation within and among households,
based on the distribution of source groups. The second was to associate decorative types
from the general classification or particular attributes from the attribute analysis with
geochemical source groups. For this second function, I also included the INAA results of
the sample of 150 sherds from the general project. This general sample consisted of
samples of suspected imported types. As most of these types are relatively rare overall
and thus appear infrequently in the random sample, the addition of the general project
data allows for much firmer source attributions for these types.
The INAA sample is used to support my discussion of regional trade patterns over
time, as well as my attributions of particular types as local or non-local for purposes of
signaling foreign ties. The number of INAA samples per household (30, usually minus 49 sherds which could not be assigned to a group), provides large, representative samples
per phase, which I use to discuss trade over time. Due to the increased probability of
101
random variation in individual households, I use INAA results to discuss inter-household
variation only in conjunction with the larger samples provided by the general
classification. Both the random and type-based INAA samples are used to establish
regional affiliations for specific ceramic types. Types of particular interest were
overrepresented in the type-based sample, but there were still a fairly large number of
types which were only represented by a few INAA samples. In these cases, geographic
attributions from INAA are supported by a broad literature-based discussion of sites and
time periods where the types are present or absent elsewhere in Central Mexico.
Petrography
Thin-section petrography is a technical analysis method that complements INAA,
and the CAP used it as a supplemental sourcing method. While INAA identifies the bulk
composition of ceramics, petrography focuses on the inclusions (non-clay fraction).
These may include natural inclusions in the clay which were not removed during
processing or intentional additions during processing, such as shell, grog, or sand. This
provides information both about manufacturing decisions in the production of the pottery,
and provenience information, based on geologically or culturally distinctive combinations
of inclusions.
The CAP submitted three sets of samples for petrography to Dr. Jennifer
Meanwell. The first group consisted exclusively of material from the surface survey.
Sherds were selected by macroscopic paste group, in an effort to check the validity of
these groups. The results of this set of sherds generally showed that the field-designated
paste groups were not analytically valid categories. The second group was a stratified
102
random sample of sherds by household component, for the two excavation units with
continuous occupation through all three phases. These samples consisted of 22 sherds
from each phase at each household, plus a small number of additional sherds from
suspected non-local types. The sample from each household component was stratified by
vessel form, with bowls, jars, and other vessels selected separately. The third group was
selected for correlation between the INAA and petrographic datasets. It consisted of 87
sherds from the INAA sample, selected to provide 7-10 examples of each INAA source
group.
Based on the combination of these three groups of samples, Dr. Meanwell defined
three major local petrographic groups (I-III), one minor local group which occurred only
in the survey (IV), one general unspecified local category (V), and eight much less
common non-local source groups (VI-XIV). The non-local groups can be divided
between those associated with types from the Basin of Mexico (VI-IX), and those
associated with types from areas to the south or southwest of the Toluca Valley (X-XIV).
These groups are described in more detail in Chapter 4. All further analyses use Dr.
Meanwell’s petrographic group assignments.
My use of the petrographic analyses is similar to those from the INAA samples.
The petrographic results are used in conjunction with INAA and the general ceramic
classification to identify changes in interregional trade patterns over time, with results
considered at the phase level. Because most households were not sampled for
petrography, this analysis is not used to discuss inter-household variation. The
petrography samples were also used to establish source regions for ceramic types, in
conjunction with the INAA and literature-based evaluation of sources.
103
Lithic Analyses
Lithics were analyzed by form and function, and samples were submitted for XRF
analysis. All excavated lithics were given a basic classification by form (blade, flake,
tool, eccentric, etc.) and by material (grey obsidian, green obsidian, chert, basalt) by the
lab staff. In addition, Dr. Bradford Andrews performed a more detailed analysis of most
of the DS-1 and DS-2 lithics, recording source material, technological artifact type, and
wear visible with a hand lens. Each artifact type code is also associated with a production
technology (core-blade, bifacial, bipolar, unknown), and a stage of production (general
production, core rejuvenation, finished tool, etc). (See Appendix XX for a list of lithic
codes, and Andrews (2013) for the initial results). The lithic classification is primarily
used to discuss craft production at Calixtlahuaca in chapter 5. In addition, the total lithic
counts and green obsidian counts are used as one line of evidence for evaluating
household wealth in chapter 6. Most household components have reasonable lithic
samples; only 316-Ph2 and 327-Ph6 have less than 50 lithic artifacts.
A sample of the obsidian artifacts was also analyzed via XRF by Dr. Adrian
Burke to determine provenience. While the obsidian samples were originally selected as a
quasi-random sample from about half of the household components, the sourcing
revealed that there was substantial sampling bias introduced due to the correlation
between obsidian sources and particular artifact types. As a result, the obsidian sourcing
data are valid at the overall phase level, but not at the individual household level. The
XRF samples were initially processed by Dr. Adrian Burke to test experimental analytical
methods. Any samples that he could not process due to technological limitations on
sample object size were then sent to MURR for analysis. In addition, two or three
104
samples from each of Dr. Burke’s source groups were sent to MURR to confirm their
geographical assignments. (See Burke and Gauthier (2013) for initial sourcing results.)
The double tested samples showed near-perfect agreement between labs. The lithic
sourcing data are primarily used to discuss trade at Calixtlahuaca, and they provide a
valuable comparison to the patterns identified in the ceramic analyses.
Other Analyses
In addition to the primary ceramic and lithic datasets, this project also drew on a
number of additional analyses of CAP data, performed both by me and others associated
with the project. These include an analysis of ceramic figurines, an analysis of textileproduction artifacts, and an analysis of ground stone forms.
The excavations recovered 369 figurines or figurine fragments, of which 136
come from DS-2 household contexts. The figurine assemblage was initially analyzed by
Rosario Endañu, a Mexican archaeology student at the Universidad Autónoma de Estado
de México, for use in her licenciatura thesis. It was later reanalyzed by Dr. Michael Smith
to create a higher degree of coding comparability with his previous projects. Both coding
systems recorded combinations of paste, stylistic affiliation, and subject matter for each
piece. Efforts to have a small number of figurines exported for INAA were unsuccessful.
Figurines are used to discuss cultural changes in ritual practices in chapter 6. In order to
maintain adequate sample sizes, figurines are formally analyzed at the phase level, with
some more subjective discussion of the patterns seen in individual households.
I analyzed the textile production artifacts, including spindle whorls, spinning
bowls, and tabular basalt desfibrador scrapers, from the site for one of my MA papers
105
(Huster 2009). While refinements to the site chronology and the DS-1 sample list have
rendered many of my original conclusions obsolete, I do draw heavily on the basic
catalogs and artifact classifications generated by my previous work. Textile production
artifacts are used to discuss changes in craft production.
The ground stone artifacts from the excavations (n=560) were originally classified
into broad functional categories by Charles and Maria Stapleton, with input from Dr.
Michael Smith. I recorded more detailed attributes for the ground stone by functional
class, including details of form and wear. Maize grinding tools were analyzed on a
comparative regional basis by Justin Mortensen for his Honors thesis (Mortensen 2014).
Ground stone artifacts are used to discuss changes in both household wealth, and changes
in food preparation practices.
Collections and Data Management
The artifacts excavated by the CAP are currently stored at the project’s lab at the
Colegio Mexiquense in Zinacantepec, Mexico. Upon submission of the final project
report to INAH, the majority of the ceramics not included in the DS-1 or DS-2 samples
will be discarded. With the possible exception of burned daub, all other artifact classes
will be kept, regardless of context of recovery. One copy of the project’s paper records
(excavation forms and lab forms) is stored with the artifacts in Mexico and a second copy
is located at Arizona State University. A copy of project records, including scans of field
and lab forms, databases, artifact catalogs, and photographs will be uploaded to tDAR
beginning in summer 2016 and will be publically available there (tDAR id: 401248).
106
CHAPTER 4
CHAPTER 4. “EVERY KIND OF MERCHANDISE SUCH AS MAY BE MET
WITH IN THE LAND4”: REGIONAL EXCHANGE SYSTEMS
The Postclassic period in Mesoamerica is characterized by an increase in
interregional interaction. One of the major ways this occurred was an increase in trade,
including increased medium and long distance trade, trade in a broader range of goods,
and greater access to trade goods across a wide spectrum of the population (Berdan 2003;
Blanton, et al. 2005; Golitko and Feinman 2015). However, the degree to which this
increase in trade was affected by the political expansion of the Aztec Empire remains
subject to substantial uncertainty (Nichols, et al. 2009), especially in those portions of the
Aztec Empire outside of the Basin of Mexico. The Aztec Empire is known to have
relocated markets to remove them from local elite control, which may or may not have
affected commoner access to goods. In addition, the need for regular tax (tribute)
payments could have either promoted long distance exchange or reduced the surplus
available for purchasing luxury goods. Thus, the economic impacts of Aztec rule on
provincial households have important implications for evaluating Aztec imperial
activities in provincial areas.
In this chapter, I address the question of interregional exchange at Calixtlahuaca
over time, in order to distinguish between broader Postclassic economic trends and the
4
Cortés, Hernando
1962
Five Letters of Cortés to the Emperor. Translated by J. B. Morris. Norton, New York. Pp
87-89
107
effects of Aztec rule. I focus on three aspects of exchange: the volume of material traded,
the diversity of regions from which it came, and the evenness of household access to nonlocal goods. The first two of these aspects address the influence of the directness of Aztec
rule; to what extent was the Triple Alliance able to alter preexisting trade patterns in
directions that benefitted themselves? I expect more direct rule to appear as a noticeable
deviation from prior trends in trade, particularly as a greater proportion of imported
material from the Basin of Mexico, at the expense of trade connections with other areas.
The second two of these three aspects of exchange are used to address the balance
between more and less collective strategies of rulership. Under more collective forms of
rulership, provincial areas have greater leverage relative to the core and are thus more
likely to maintain preexisting trade connections. In addition, under relatively collective
rule, changes in household wealth are more likely to be evenly distributed, meaning that
interhousehold variation in access to foreign goods should remain relatively stable.
I analyze exchange patterns for ceramics using a combination of INAA,
petrography, and type based classification. I analyze exchange patterns for obsidian using
XRF sourcing. The results of these analyses show an increase in the diversity and volume
of long distance trade between the Dongu (MPC) and Ninupi (LPC-A) phases. This
demonstrates that prior to the Aztec conquest of Calixtlahuaca, the site was becoming
progressively better integrated into broader Postclassic market systems. The preexisting
local organization of social power was also sufficiently collective to insure that access to
non-local goods was relatively evenly distributed among households. Under Aztec rule,
during the Yata (LPC-B) phase, the source diversity and evenness of source distribution
among households decrease for both ceramics and obsidian. However, the total quantity
108
of imported ceramics increases, while the total quantity of imported obsidian decreases.
The percentage of non-local goods from the Basin of Mexico increases from each phase
to the next for both ceramics and obsidian. This corresponds to a sufficient degree of
imperial economic control over economic processes to reorient them in directions
favorable to producers and traders from the imperial core. However, this control was
most likely achieved primarily via network strategies of rulership, as the distribution of
non-local goods becomes more uneven under Aztec rule and access to prior trade
connections was diminished.
The Study of Commercial Economies in Mesoamerica
The presence of longstanding, well-integrated market systems in Postclassic
Central Mexico has been well documented on the basis of early historic documents
(Cortés 1962:pp 87-89; Díaz del Castillo 1963:232-234; Sahagún 1950-82:Book 8, p. 6769; Torquemada 1975-83:v.4, 348-352) and archaeological analyses of various forms of
material culture (Garraty 2007, 2009; Nichols, et al. 2002; Smith 1999, 2010; Stark and
Ossa 2010). Together, these two forms of evidence provide a baseline for the Aztec
economy. They describe a market system in which diverse types of goods were available
to all consumers and with negotiable prices. This description provides a good broadbrush picture. Stylistic similarities (Boone 2003; Smith 2003d) and shared types of
artifacts (Smith 1990) during the Postclassic demonstrate, at a minimum, the widespread
exchange of ideas. Sourcing techniques (e.g. Braswell 2003 for obsidian; Skoglund, et al.
2006 for ceramics) and more quotidian ethnohistoric information on trade in the
109
Relaciones Geograficas (Hirth 2013) support this perspective, demonstrating the
movement of goods as well as ideas across much of greater Mesoamerica.
However, the widespread acceptance of a commercial economy during the
Postclassic has not come easily to archaeology. Historically, there has been a sense that
premodern economies were relatively static, with most production oriented primarily
toward household subsistence, and that any longer-distance economic interaction was
driven by elite actions (Polanyi, et al. 1957; Wittfogel 1957). This perspective also
appeared in Mesoamerican archaeology, in both culture-historical and ecologicaladaptation paradigms (Chapman 1957; Sanders and Santley 1983; Sanders and Webster
1988). More recently, widespread work has demonstrated that Mesoamerican economies
featured substantial long distance exchange (Hirth 2013), that the organization of trade
changed over time (Braswell 2010; Feinman 1996; Golitko and Feinman 2015), and that
at least during the Epiclassic and Postclassic periods, commercial exchange played a
significant role in provisioning commoner households (Garraty 2009; Hirth 1998; Smith
2003c).
At the same time, this research has produced an emphasis on the commercial
nature of the Postclassic Mesoamerican economy, which masks a large degree of
variation in the volume of goods traded and the distances over which they travelled.
Studies of exchange patterns on a smaller geographic scale have demonstrated the
presence of reasonable diversity in the quantity and types of items, distance of exchange,
number of trading partners, and the relationship between political units and trade patterns
(Hodge and Minc 1990; Kowalewski, et al. 2010; Minc 2009; Skoglund, et al. 2006;
Venter 2008). More recent syntheses have emphasized that market economies do not
110
exist on a presence/absence basis and that looking at why and how particular aspects of
commercialization develop is more useful than simply identifying the presence of a
market based economy (Feinman and Garraty 2010; Garraty and Stark 2010). Similarly,
the relationship between increasing general commercialization and the political economy
of the state also bears investigation.
Patterns of Exchange in Postclassic Central Mexico
Traded goods have both a source and a destination. As a result, studies of trade
can examine what is traded within a region, what is traded out of, and what is traded into
a region. Each of these dimensions of trade can also be examined on multiple geographic
scales, such as the site, the sub-region (such as the Basin of Mexico), a region (the
Central Highlands), or a macroregion (all of Mesoamerica). Studies of trade in Central
Mexico have focused largely on exchange within or out of the Basin of Mexico, with few
attempts to examine the multidimensional trade networks in which the Basin was
enmeshed.
The Basin of Mexico
The Basin of Mexico provides one of the most in-depth case studies of
Mesoamerican trade, with the majority of information coming from Instrumental Neutron
Activation Analysis (INAA) sourcing studies of ceramics, which have produced the
largest set of INAA data on ceramics in Mesoamerica (Nichols, et al. 2014), over the
course of multiple studies (Brumfiel 2005b; Crider 2011; Garraty 2007, 2013; Hodge
111
2008; Hodge and Minc 1990; Hodge, et al. 1992; Minc 2009; Stoner, et al. 2013). While
pottery cannot be taken to characterize the movement of all goods in an exchange system,
it provides a broad-brush proxy for the general characteristics of exchange in a region.
During the Middle Postclassic, intra-Basin trade occurred primarily, but not
exclusively, within political subdivisions of the Basin. After the establishment of the
Triple Alliance, the former economic divisions of the Basin relaxed, but remained
present, and the Mexica capital of Tenochtitlan supplied increasing amounts of pottery to
other portions of the Basin at the expense of ceramic producers in other portions of the
region (Garraty 2007). After the Spanish conquest, the city of Texcoco came to dominate
the production of many of the types of decorated ceramics used in the Basin (Garraty
2013).
Research on goods exported out of the Basin of Mexico has focused on ceramics
and obsidian. The frequency of exported ceramics generally follows a drop-off curve as a
function of distance (Sergheraert 2009; Smith 1990). There are three types of commonly
exported ceramics: Aztec Black-on-Orange, Aztec Black-on-Red (Guinda Redwares),
and Texcoco molded/filleted sahumadors. Measuring the export frequency of these types
is complicated by their local production in some provincial areas (e.g.Ohnersorgen 2006).
A fourth type, Texcoco Fabric Marked (sometime also called Texcoco Saltware), is found
in limited areas outside of the Basin, but has a much smaller distribution than the
previous four types (Smith 1990), and likely travelled as packaging for salt rather than for
its own sake. Green obsidian, from the Pachuca, Hildalgo source, the majority of which
was routed through the Basin of Mexico, drops off with distance from the Basin of
Mexico (Smith 1990). Cities located on trade routes, such as Otumba, became specialized
112
producers of trade goods, destined for markets both within and beyond the Basin of
Mexico (Nichols 2013).
The discussion of goods imported into the Basin of Mexico has focused primarily
on goods brought into the Basin as tax payments to the Triple Alliance, rather than on
trade-based mechanisms. While calculations of goods brought into the Basin as taxes
demonstrate that such items would have met a small but consistent portion of the region’s
needs (Drennan 1984; Sanders, et al. 1979:184, 378), this movement of goods is next to
invisible archaeologically. This is partially due to the perishable nature of many tax
goods, such as foodstuffs or textiles, and partly due to the lack of systematic excavations
in the imperial capitals where the majority of taxes in kind likely ended up. Excavations
at the Templo Mayor have recovered a great diversity of items imported from many
different regions, but it is a unique context and should not be considered characteristic
Tenochtitlán more generally, let alone the Basin of Mexico as a whole. Wide-scale test
pits at Tenochititlán’s sister city of Tlatelolco recovered extremely few non-local
ceramics (González Rul 1988a), despite the site being location of the largest market in
the Late PostclassicBasin of Mexico
Excavations at other locations in the Basin have generally focused on intra-Basin
political and economic relationships, rather than on ties to regions outside the Basin of
Mexico, and many project typologies do not even report categories for imported wares
from outside the Basin. This is likely a combination of a historically nationalistic focus in
Mexican archaeology emphasizing the primacy of the Basin of Mexico (Robles García
2012), a low level of imported ceramics, and a lack of familiarity with imported types.
Michael Smith noted that he sometimes finds Tlahuicua (Morelos-style) polychromes
113
misidentified by researchers in the Basin of Mexico, and my personal experience
suggests that many Middle and Late Postclassic Toluca Valley types would be
misidentified as “odd” examples of Mazapan wavy line by researchers more accustomed
to the Basin of Mexico ceramic sequence. Overall, however, most excavations of
domestic contexts in the Basin of Mexico show very low levels of ceramics from outside
the region.
Central Mexico Beyond the Basin
Outside of the Basin of Mexico, both intra- and interregional trade appear to have
been more multidirectional. In an analysis of both modern and historic excavation
collections, Smith (2003b) showed that ceramics stylistically characteristic of four
different subregions in Central Mexico (Toluca Valley, Malinalco, Eastern Morelos, and
Western Morelos) were traded among all four areas. In addition, ceramics from the Basin
of Mexico were recovered in all of these areas, and most areas also had ceramics from
other areas in or beyond Central Highland Mexico (ranging from Tonatico to the
Huasteca). While the focus on elite contexts in many of the older excavations included in
this study may elevate the quantity of imported ceramics recovered, the general pattern is
still one of far more diverse exchange than is seen in the Basin of Mexico proper.
In adjacent regions, imported ceramics from the Basin of Mexico usually
increase in frequency over the course of the Postclassic, while those from other regions
decrease (Huster 2015). This generally indicates that outside of the Basin of Mexico,
Aztec rule produced a shift from a more laterally integrated market network to a dendritic
market pattern centered on the Basin of Mexico. However, where there is sufficient
114
chronological control, Aztec ceramics often began increasing prior to an area’s political
incorporation into the empire, suggesting that the increase in Aztec ceramics was not
exclusively a result of political conquest.
Trade and Empire At Calixtlahuaca
Both geographic and cultural distance can serve as potential barriers to trade,
though the latter likely played a larger role at Calixtlahuaca. The site is approximately
60km from Tenochtitlan, and within 100km of most of the Basin of Mexico. This places
Calixtlahuaca outside of the 4-8 km radius which people will generally travel for their
regular market needs (Blanton 1996). However, assuming 36km/day as a reasonable
travel distance (Drennan 1984), Calixtlahuaca is within two days travel from
Tenochtitlan. It is also well within the 550km calorically efficient radius for the transport
of maize by overland human transport (Drennan 1984, 1985). This theoretically predicted
radius of trade is supported by documentary evidence. The Matricula de Tributos page of
taxes demanded from the Toluca Valley includes maize (Matrícula de Tributos
1980:lamina 13), and maize from the Toluca Valley is among the varieties sold by the
maize seller in Florentine Codex, which is presumably describing vendors in the Basin of
Mexico (Sahagún 1950-82:Book 10:65-66). Ethnographically, peasant craft producers
from the Toluca Valley continued transporting mats and other reed products to Mexico
City on foot, in 70kg loads using tumplines, well into the 20th century (García Sánchez
2008:198-202). This demonstrates that the regions were close enough to allow for the
transport and exchange of both foodstuffs and relatively bulky items of other types. As a
115
result, geographic distance should not be considered a barrier to trade, and a lack of
exchange between the two regions is more likely due to either a lack of economic
integration or political barriers to trade.
After the Aztec conquest of the Toluca Valley, the Aztec Empire shifted the
primary regional center of administration from Calixtlahuaca to Tollocan, where they
established a military garrison and governor (Smith and Berdan 1996a). Calixtlahuaca
remained under the nominal control of local rulers. Calipixque (tax collectors) were
present in both Tollocan and Calixtlahuaca. It was a common Aztec strategy to remove
markets and their associated revenue from the control of local elites. This likely occurred
at Calixtlahuaca, though whether this involved the movement of political control,
economic control or both to Tollocan is not clear.
Hypotheses About Trade At Calixtlahuaca
The Toluca Valley sits between the areas traditionally considered the cores of the
Central and West Mexican culture areas. During the Late Postclassic, these two regions
came under the political control of the Aztec and Tarascan Empires, respectively. Due to
limited prior work in the Toluca Valley, it is relatively unknown to what degree the
region was integrated into the market networks extending from either of these core areas
during the Postclassic. I evaluate regional trade at Calixtlahuaca for both ceramics and
obsidian.
The two dimensions of variation in rulership, directness and collectiveness,
produce partially overlapping expectations, due to the degree of correlation between
116
indirect rule and network (non-collective) imperial rulership. However, the directness of
rule can generally be evaluated based on the degree of resulting change in the provincial
area, regardless of the directionality or form of the change. In contrast, the collectiveness
of rule is broadly associated with the directionality of change, with more collective rule
generally expected to produce outcomes more favorable to commoners. Based on the
expectations of either relatively indirect (Hassig 1985) or relatively collective rule
(Blanton and Fargher 2008), I present two scenarios for the effect of Aztec rule on long
distance trade:
1. If Aztec rule was primarily indirect, as argued by Hassig (1985, 1988), it
should have had relatively few effects on long distance trade at Calixtlahuaca,
but those effects which do occur should be primarily negative for the local
population. Under indirect rule, more goods are extracted out of provincial
areas, but this extraction is done using preexisting mechanisms. As a result,
there will generally be an overall reduction of wealth, resulting in less ability
to purchase foreign goods. However, at the same time, there should be
relatively few changes in the diversity sources of goods available. Non-local
goods may also become more unevenly distributed if access to distant source
locations becomes interrupted due to imperial attempts to control local elites.
2. Alternatively, if Aztec rule was relatively collective, as argued by Blanton and
Fargher (2008), it should be expected to be accompanied by increased
economic growth and generally positive economic effects for provincial areas.
117
Under such conditions, the expansion of the Aztec Empire should reinforce
the prior development of well integrated market systems. This should be
visible archaeologically as an overall increase in the degree of trade, both due
to increased purchasing power and increased regional specialization in the
production of particular goods. However, due to imperial influence over the
development and organization of the market system, a disproportionate
portion of the increase in goods moving within the system may be from the
imperial core. Under relatively collective rule, lateral trade links, visible as the
diversity of sources from which goods come, should remain comparable (or
increase) to the period prior to imperial rule, as commoners have relatively
high market access to such goods and the state has no reason to interrupt
access. Finally, the interhousehold variation in access to non-local goods
should remain comparable to pre-imperial levels, due to most commoners
acquiring such items on the market rather than via elite patronage.
Ceramic Importation
As a result of the general lack of knowledge about regional trade patterns and
potential proveniences for ceramic types found at Calixtlahuaca, I used multiple methods
to investigate ceramic exchange at the site. I used INAA and thin section petrography to
establish source regions and provide a general overview of exchange over time. The
results of the specialized analyses were then used to support literature-based assignments
of probable source regions by type, allowing for a broader discussion of trade based on
118
the general CAP classification of household ceramics. Source regions were assigned
based on three lines of evidence: INAA analysis, petrographic analysis, and type
frequencies at other sites within and outside the Toluca Valley.
Figure 4.1 Geographic source macroregions for ceramics excavated at Calixtlahuaca
The two technical analysis methods both identify a number of individual groups.
While the different methods produce different specific groupings, they can be crosscorrelated to each other and to geographic locations at macroregional levels. As a result, I
assign ceramics to one of three regional macrogroups: local, Basin of Mexico, and the
South-Southwest portion of the modern state of Mexico (Figure 4.1). Source groups
assigned to the local macrogroup are assumed to be from the Toluca Valley, though not
119
necessarily exclusively from Calixtlahuaca itself. They make up the majority of the
sampled ceramics, have geology consistent with the Toluca Valley, and consist primarily
of vessel forms and decorative types found at sites in the Toluca Valley. The Basin of
Mexico macrogroup consists of those groups made up of types stylistically characteristic
of production in the Basin of Mexico (Figure 4.2), and, in the case of INAA, which match
the chemistry of the Basin. The S/SW State of Mexico macrogroup consists of groups
primarily of decorative types found in Malinalco, Ixtapan de la Sal, or Valle de Bravo
museum collections (Figure 4.3). These groups are both chemically and petrographically
distinct from the Toluca Valley, but their geographic assignment is based strictly on type
comparisons, as MURR has no reference samples from this macroregion. A small number
of very rare decorative types, none of which were included in the petrographic or INAA
samples, were identified as being from other geographic regions on a stylistic basis.
At the macroregional level, ceramics at Calixtlahuca show an increase in imports
from the Basin of Mexico over time, with the largest increase between the Ninupi and
Yata phases. Imports from the S/SW and other minor regions decrease over time, though
the differences are not statistically significant for all methods.
120
Figure 4.2 Selected ceramic types associated with the Basin of Mexico source
macrogroup, based on both INAA and petrographic analyses
121
Figure 4.3 Selected ceramic types associated with the S/SW State of Mexico source
macrogroup, based on both INAA and petrographic analyses
INAA
I submitted 569 ceramic sherds for INAA at the Missouri University Research
Reactor (MURR). As described in the previous chapter, these sherds represent a stratified
random sample of seventeen out of the eighteen excavated household components at
Calixtlahuaca (“Household sample”). The random sample of sherds was supplemented by
122
an additional 150 sherds selected by type (“Type sample”). These additional sherds were
selected to represent 5-6 sherds per type for a variety of suspected non-local types, as
well as some common local decorated types. As a result of this dual sampling strategy,
the frequency of imported ceramics can be considered both on a household-by-household
basis and on a type frequency basis. The catalog of INAA samples and the results from
MURR can be found in Appendix C.
Based on the INAA results, MURR assigned ceramic samples from Calixtlahuaca
into nine groups (Stoner and Glascock 2013). These are identified as Groups A-H and
Group Guinda in the original report, but I have relabeled them as Groups 1-9 to avoid
confusion with the letter-named ceramic decorative families used at Calixtlahuaca.
Approximately 18% of the sherds could not be assigned to one of these groups. These
groups include three local groups (Groups 1-3), three Aztec-associated groups (Groups 46), two groups associated with areas of the State of Mexico to the south and southwest of
the Toluca Valley (Groups 7-8), and one group resulting from contaminated samples
(Group 9). I made geographic assignments on the basis of MURR’s chemical reference
groups for the Basin of Mexico and Morelos, and stylistic similarities for all other
regions.
The INAA samples generally show an increasing frequency of imports from the
Basin of Mexico over time (Groups 4, 5, and 6), accompanied by a decreasing level of
imports from the regions to the west and southwest of the Toluca Valley (Groups 7 and 8)
(Table 4.1). However, due to sample size, the a chi-square test comparing phases does not
show significant differences for this analysis (²=4.63, df=4, p=.33). The average
diversity of groups per household component increased from each phase to the next, from
123
four groups per household during the Dongu phase, to four and a half during the Ninupi
phase, and just over five and a half during the Yata phase.
Unit Phase
307 Dongu
315 Dongu
316 Dongu
320 Dongu
323 Dongu
324 Dongu
Dongu Total
Local
1 2 3
19
18
11
25
2
22 1
11 5
3 6
98 12 13
303 Ninupi 18 1
307 Ninupi 18
308 Ninupi 19
311 Ninupi 48 1
316 Ninupi 24
322 Ninupi 21
Ninupi Total 148 2
307 Yata
309 Yata
316 Yata
317 Yata
324 Yata
Yata Total
15
21
20
21
22
99
2
1
1
3
7
Basin of Mexico S/SW St. Mex. Unknown
Total
4
5
6
7
8
9 Unkn. Total Known
1
10
30
20
1
30
29
3
30
27
1
6
30
24
1
1
3
9
30
21
2
2
17
30
13
3
1
4
3
46
180 134
2
2
2
1
1
2
1
6
1
6
1
2
1
4
1
1
34
1
1
1
2
2
1
1
1
1
1
5
2
1
1
3
4
1
36
1
1
1
1
2
1
2
4
8
6
7
6
4
35
60
30
30
60
30
30
240
22
21
24
52
24
26
169
7
4
5
6
5
27
30
30
29
30
30
149
23
26
24
24
25
122
Table 4.1 INAA ceramic sourcing results by household component, with groups arranged
by macroregion (Local, Basin of Mexico, South-Southwest State of Mexico, and
Unknown). Household sample only.
Local INAA Groups. Group 1 (Group A in the report) is the most common local
group (n=422). It accounts for 70% of the total sherds and 75% of the sherds in the
household sample that could be assigned to a group. It is the dominant group during all
124
three phases of the site’s occupation. It includes examples of all major vessel forms, and
a wide range of plain and local-style decorated types. It also includes some, but not all, of
the pieces originally identified as locally produced versions of Aztec III or III/IV Blackon-Orange. (See Table 4.2 for all INAA group to ceramic type correlations.)
Group 2 (Group C in the report) is a secondary local group of 24 samples,
accounting for 4-4.5% of the sourceable sherds in both the combined and random
samples. It is somewhat associated with excavation units 323 (5 sherds), and 324 (10
sherds). Like Group 1, it contains a range of plainware and local decorated types, in all
major vessel forms. The only one of the two analyzed negative design sherds (G-types)
assigned to a group is in this group.
Group 3 (Group H in the report) is a minor local group, consisting of 26 samples.
It is somewhat associated with material from 315-Ph2, with this component accounting
for 58% of the group. This suggests that either this unit had an unusual depositional
environment, or unique trade ties to a particular group of nearby potters. In contrast to
Group 9, this group is unlikely to be the result of processing contamination, as the 315Ph2 sherds include cases from both the household and type-based samples, which would
have been processed on different days, both in the field and at MURR. The group
includes a wide range of local vessel forms and decorative types, though jars are
noticeably underrepresented.
Aztec INAA Groups. Group 4 (Group B in the report) is an imported group of
unknown origin consisting of 33 samples. The types in the group consist exclusively of
plainwares and Aztec Black-on-Orange bowls. The Black-on-Orange bowls in this group
125
crosscut the field classification of pastes, including examples of all three paste groups
associated with the decorative style. Chemically, the group matches neither the local
groups in this sample, nor the general or sub-regional Basin of Mexico or Morelos
clusters in the MURR database. It may represent the northeastern Toluca Valley, an area
which saw more interaction with the Tepanec state, or a previously unknown production
location in the Basin of Mexico. The group as a whole is temporally sensitive, increasing
in frequency from each phase to the next, but the types assigned to the group also shift
over time, from mostly plainwares to mostly Aztec Black-on-Orange. Based on the
decorated types in the group, it is considered an Aztec-associated source group for the
purposes of further analyses.
Group 5 (Group D in the original report) consists of imports from the Basin of
Mexico. It includes 24 samples, only five of which are from the random sample. This is a
direct result of the emphasis on selecting presumably non-local types for the type based
sample. The majority of sherds assigned to this group are Aztec Black-on-Orange, with
two plain or eroded bowls as the only exceptions. With one exception, this group consists
of pieces field-typed as being “imported” (the unmodified Aztec III and III/IV codes) or
“unsure” (Local B types) pastes, rather than locally produced imitations of Aztec wares
(Local A types). The group also shows a motif-based bias, including only two examples
of the later Aztec III/IV variant, though it occurs in equal quantities in the Ninupi and
Yata random samples. While the chemistry of this group is consistent with the Basin of
Mexico as a general region, none of the samples are a clear match to any of the known
sub-regional source groups within the Basin of Mexico.
126
Group 6 (Group Guinda in the report) is closely correlated with Basin of Mexico
style guinda Black-on-Red wares; the group contains only type 192 and 208 sherds (the
two type codes for guinda bowls), and ten of the eleven samples from these types were
assigned to this group. While two of the sherds match the Tenochtitlan reference group
specifically, the majority are not a clear match to a single subregion within the Basin of
Mexico. They may be sherds drawn into a single cluster based on the underlying
similarities of the type of clay used in different subregions, rather than due to a common
origin. Guinda Redwares have a distinctive very fine buff or light grey paste. This
suggests specialized processing, as clays from the Basin of Mexico generally fire to an
orange or at least orange-tinted color. Guinda wares sourced to the Basin of Mexico and
Morelos by other projects show no distinction in macropaste or decoration between the
sherds from the two regions (Smith 2006c). The samples analyzed at Calixtlahuaca are
definitely not locally produced, and this group is treated as an Aztec source group for
further analyses.
S/SW State of Mexico INAA Groups. Group 7 (Group E in the report) includes 18
sherds and consists of some plainwares and a variety of relatively uncommon decorated
types. The decorated types include C-1 and C-2 polychromes, incised redwares, and some
Red-on-White types (decorative family D). Based on the types included in this source
group, it is probably from an area peripheral to the Toluca Valley, likely to the south,
such as Malinalco or Ixtlan de la Sal. Eight of the sherds in this group were matched to
Morelos references groups by MURR. Given that they are stylistically clearly not from
Morelos, it is likely that the sourcing is picking up on similar geological formations to the
127
south of the Toluca Valley, rather than south of the Basin of Mexico. It is treated as an
imported group in further analyses.
Group 8 (Group G in the report) includes 10 sherds, consisting exclusively of
incised redwares and a few plainwares. Based on the frequency of incised redwares from
archaeological contexts around the Toluca Valley and surrounding regions, this source
group likely comes from the greater Valle de Bravo (Malacatepec) region. Based on the
amount of chemical variation among the samples, it is likely that this group includes
pieces from a relatively wide geographic area, which could be subdivided if there was a
larger set of reference samples for areas peripheral to the Toluca Valley. The samples in
this group are treated as imports in further analyses.
Unassigned INAA. Group 9 (Group F in the report) (n=39) has an almost one to
one association with the first set of samples taken from Unit 303 and is considered
contaminated and treated as unassigned in later analyses, as there is no reason that the
range of types present in this component should form a unique group. Based on the
elevated concentrations of Mn and Co that cause these samples to form a group, it is
unclear whether the contamination resulted from the depositional environment or from
some aspect of sample preparation.
There is also a remainder of sherds not assigned to any of the above groups,
referred to as “unknown”, which accounts for 18% of the total sample. The majority of
these sherds are probably moderately anomalous paste variations from the regions
described above. A sample of 14 INAA-unassigned sherds was included in the
INAA/petrography cross-correlation sample and could all be assigned to existing
128
petrographic groups, supporting the premise that they were the result of occasional
variations in production.
Type
Code
Local
1 2 3
Basin of Mex.
4 5 6
S/SW
7 8
Unasign.
9 Unk.
N.
1
Aztec Imports
Aztec III
21
Aztec III/IV
114
Aztec III, Local B
181
Aztec III/IV, Local B 182
Aztec III, Local A
271
Aztec III/IV, Local A 274
Aztec Or, unkn.
1
Aztec Or Spn. Bowl 67
Guinda Bl/R
208
Guinda, fine line
192
Copa, Guinda
113
Sahumador,
Mold/Fillet
64
Thin-Walled Basin
42
Other Imports
B-5 Bowl
207
B-11 Bowl
137
C-2 Bowl
214
D-0 Bowl
215
D3 or D4 Bowl
188
G-0 Bowl
230
Coyotlatelco Bowl 166
Other dec Bowl
122
Other dec Jar
38
Plain and Eroded
Bowl, Eroded
10
Bowl, Plain
11
Jar, Eroded
30
Jar, Plain
31
Thick-Rim Jar
167
Comal
45/46
3
4
2
4
5
2
2
3
1
10
1
7
1
1
1
2
1
1
1
7
3
1
1
1
1
3
1
1
2
4
1
2
2
1
1
2
1
1
2
1
3
2
1
1
1
1
2
3
2
2
1
7
7
1
7
3
2
1
1
1
8
19
8
15
5
8
47
84
59
88
28
16
2
1
1
1
1
32
42
40
56
21
4
1
1
1
5
1
1
1
9
1
3
3
2
1
3
1
1
2
2
1
1
2
3
6
9
3
Table 4.2 INAA ceramic sourcing group to project ceramic type correlations
129
14
7
16
6
14
4
1
2
7
4
1
Type
Code 1
Decorated Bowls
B/E frag
264 10
Red rim
200 14
B-0
202 21
B-1
203
2
B-2
204
8
C-1
213 14
E-0
221
5
E-1
222 13
E-5
225 13
E-6
226
3
E-9
277 18
E-10
132 17
E-11
133
4
E-12
173
E-14
185
E-16
184
1
Decorated Jars
B/E frag
266
1
B-0
253
2
E-0
255
6
E-2
256 19
Other Forms
Biconical Censer
267 14
Other Censer
170
2
Scored Censer
101/18 2
Sahumador, Toluca 262
8
Crude Unfinished
134
9
Miniature Vessel
71
1
Copa
70
1
Asymmetrical
168
1
Pitcher
72
1
Spoon
100
2
Total
Local
2 3
Basin of Mex.
4 5 6
S/SW
7 8
Unasign.
9 Unk.
1
1
2
1
1
1
2
1
2
1
1
3
1
2
1
1
1
1
1
1
4
3
4
1
1
1
1
4
4
4
2
2
4
2
1
1
1
1
1
1
1
421 24 26
1
1
3
5
1
2
8
1
1
1
33 24 10
18 10
39 134
N.
14
21
26
2
9
19
7
19
21
4
27
22
5
1
2
1
1
2
10
22
20
5
7
11
21
1
1
2
2
3
741
1
Bowl unless other form given
Table 4.2 (continued) INAA ceramic sourcing group to project ceramic type correlations.
130
Petrography
Petrographic thin section analysis was used as an independent check on the results
produced by INAA. As described in the methods chapter, the petrography samples were
selected using a triple strategy. First, Julie Novic selected samples based on macropaste
from the surface survey material (Meanwell 2013b). Second, a stratified random sample
of 22 sherds was selected for each of the three phases from the two households with
continuous occupation, Units 307 and 316, to provide a general picture of the relative
frequency of petrographic groups at the site over time (Meanwell 2013a). In addition,
following the completion of INAA, a sample of 8-15 sherds from each INAA source
group (n=85) were also selected for petrography for two reasons. First, this sample allows
for the correlation of the ceramic groupings produced by the two methods. Second, it
created for a much wider petrographic sample of types, allowing for more relatively rare
types to be assigned to petrographic groups (Meanwell 2014). All petrographic analyses
were performed by Dr. Jennifer Meanwell and her reports are attached as Appendix E.
The petrographic analysis resulted in the identification of 13 groups, labelled
groups I-III and VI-XV. (IV and V were excluded from the final numbering scheme to
avoid confusion with earlier versions). Based on the mineral inclusions and the
decorative types in each group, the groups correspond to the same three macroregional
divisions as INAA groups. Each macroegion includes a general group for sherds that
could only be identified at the macroregional level. There are three major local paste
groups, I-III and group VI is a used for local sherds that were not a match to a specific
group. Groups VII-IX are from the south or southwest of the State of Mexico, and Group
X is the general macroregional group for this area. Groups XI-XIII are from the Basin of
131
Mexico and Group XIV is the general group for this macroregion. Group XV is used for
all sherds that did not match one of the previous groups, and are thus probably imports
from another region. Approximately 5% of the samples could not be assigned specific or
macroregional groups.
Unit Phase N.
307 Dongu 22
316 Dongu 22
307
316
307
316
Total
Ninupi 22
Ninupi 22
Yata
Yata
22
22
132
I-A
4
3
7
6
2
8
I-B
4
6
10
Local
I-C I-D
6
3
1
9
1
II
2
2
4
III
6
4
10
2
2
1
1
2
4
5
9
4
2
6
25
6
7
7
6
6
2
8
6
6
4
4
8
0
2
1
3
23
23
23
3
9
Basin of Mexico
XI XII XIII XIV
0
0
S/SW
X
Other
XV
2
2
1
1
3
0
0
0
1
1
1
3
4
3
1
1
2
1
1
1
1
2
2
4
2
1
3
1
1
2
1
1
2
1
2
3
8
6
6
0
Table 4.3 Ceramic petrography results by household, with fabrics grouped by geographic
macroregion (Local, Basin of Mexico, South-Southwest State of Mexico, Other).
Household random petrographic samples only.
The average household pattern, based on the stratified random household samples
from Units 307 and 316, shows a dramatic increase in imported ceramics between the
Dongu and Ninupi phases (Table 4.3). The total average frequency of imported ceramics
remains approximately even between the Ninupi and Yata phases. The pattern also shows
a steadily increasing frequency of Aztec imports from each phase to the next, at the
expense of other imports. Based on a chi-square test with Yates correction, the phase
132
totals do deviate significantly from random, when unassigned sherds are excluded (Yates
²=8.03, df=4, p=.09). There are some differences among households in the sources of
imports, as well as the total frequency of imports, but these do not override the general
pattern, and are generally consistent with the INAA results for these individual household
components. There is also a general agreement between the two methods about which
specific sherds were imported from a general macroregion, though specific groups do not
map onto each other perfectly (Table 4.4).
Local
4
Group 2
4
Group 3
1
2
2
III
2
2
2
1
2
VI
XI
XII
XIII
VII
VIII
IX
X
XV
1
2
2
2
2
Basin of
Mexico
Group 1
II
Group 4
2
7
1
Group 5
4
2
2
Un- S-SW
known Mex.
Local
I-A I-B I-C I-D
Basin of Mexico S/SW State of Mexico Unk.
Group 7
Group 6
1
1
Group 8
1
Group 9
1
Unknown
5
1
6
1
1
3
1
1
2
2
1
1
2
5
1
1
Table 4.4 Ceramic INAA group (rows) to petrographic group (columns) correlations,
arranged by macroregion (Local, Basin of Mexico, South-Southwest State of Mexico,
Unknown) for all sherds analyzed using both methods.
Local Petrographic Groups. Group I is the most common petrographic group at
the site, accounting for approximately 50% of the household sample and 121 sherds
overall. It contains intermediate to mafic inclusions, primarily plagioclase feldspar, with
133
little evidence for deliberate tempering. It consists primarily of local types, and includes
the full range of vessel forms found at the site. Meanwell divided it into four subgroups,
referred to as I-A to I-D. The subgroups are distinguished by combinations of dark and
light amphiboles in the temper and moderately and very optically active paste fabrics. Of
the subgroups, I-B and I-D occur in a more limited range of types and INAA groups,
while I-A and I-C generally occur in a wider range. (See Table 4.5 for all petrography to
ceramic type correlations.)
Group II accounts for approximately 7% of the household sample (n=19 overall).
It is generally similar to group I, but contains smaller mineral fragments and fewer voids.
The fabric is also generally lighter in color than Group I. Due to the mineralogical
similarities with the previous group, this is also considered a local variant. This group
includes common plainware forms, and a somewhat unusual mix of decorated types. It
also includes all three of the Aztec style Black-on-Orange sherds which were assigned to
a local petrographic or INAA group. With the exception of one Polychrome-on-White,
the other decorated sherds are small fragments which could not be assigned to a specific
decorative type.
Group III accounts for approximately 16% of the household sample (n=46
overall). It is generally similar to both Groups I and II, but has a higher iron content in
both the fabric and temper, which causes it to fire to a more reddish color when not
reduced. Some, but not all, of the sherds in this group were grog tempered, but the sherds
used as temper are not distinctive from the new vessels, indicating the reuse of ceramics
from the same production source. This group includes both plainwares and a range of
local decorated types. The local decorated types contain more B-types (local redwares)
134
and less E-types (local Red-on-Natural) than would be expected from the sample as a
whole. This petrographic group also includes all of the comal sherds that were not
matched to a clearly imported group. I tentatively consider this group to have been
produced within the Toluca Valley, but most likely not at Calixtlahuaca itself due to the
somewhat unusual combination of otherwise local types in the group.
Group VI contains sherds which generally fall within the range of variation
encompassed by the previous local groups but which were not a specific match to a
particular group (n=3). These sherds show both similar mineral inclusions and production
techniques to the local groups in general.
Aztec Petrographic Groups. In addition to the local petrographic groups described
above, there are also nine petrographic groups associated primarily with non-local
decorative types or INAA groups. Of these, four are associated with ceramics from the
Basin of Mexico, four are associated with types more commonly found to the south or
southwest of the Toluca Valley, and one consists of sherds not matching any of the
known groups. All of the groups include relatively small numbers of samples, many of
which come from the non-random samples selected for INAA/petrography correlations.
Group XI (n=11) consists almost exclusively of Aztec III or III/IV Black-onOrange ceramics. The single exception is one plain jar sherd. This group is characterized
by highly processed, likely levigated, clay. The total volume of inclusions is much lower
than in the local types described previously, though the types of minerals are similar, as
expected based on the similar volcanic geologies of the Basin of Mexico and the Toluca
Valley. This group shows evidence of firing at a high temperature. All of the Aztec
135
orangewares in the group were originally typed as either probably imported (project types
21 and 114) or imported status unknown (Local B Aztec variants), though not all sherds
in these types were assigned to this petrographic group. This petrographic group is
strongly, though not perfectly, associated with INAA Group 5. This group is considered
Aztec-affiliated for purposes of further analysis.
Group XII (n=9) consists of Aztec III or III/IV Black-on-Orange and a mix of
plainware vessel forms. This group is characterized by high firing temperatures.
Tempering is variable; most sherds show a coarse temper similar to that seen in the local
groups, but a few examples show some evidence for the deliberate removal of inclusions,
followed by the addition of sand temper. The Aztec Black-on-Orange assigned to this
group consists mostly of sherds originally assigned as locally produced imitations
(“Local A” types). The plainwares assigned to the group include both plain bowls and
comals. This group is strongly associated with INAA Group 5.This group is considered
Aztec-affiliated for purposes of further analysis.
Group XIII (n=7) consists of four pieces of Basin of Mexico style Black-on-Red
ceramics (“Guinda”), one piece of Aztec Black-on-Orange, one eroded bowl and one
local sahumador fragment. Guinda ceramics are typically not useful for petrographic
purposes because they are made of very fine clay with organic temper, providing almost
no mineral fraction for identification. The clay matrix is slightly optically active, and
most examples have a dark core and voids consistent with the use of organic temper.
Only one of these sherds was also sampled for INAA, and it was assigned to the Group 6
(the guinda INAA group). This group is considered Aztec-affiliated for purposes of
further analysis.
136
Local
Type I-A I-B I-C I-D II III VI
Aztec 1
Aztec III
21
Aztec III/IV
114
Aztec III, local B
181
Aztec III/IV, loc. B 182
Aztec III, local A
271
Aztec III/IV, loc. A 274
Guinda
208
Guinda, fine line
192
Miniature vessel
71
Texcoco Salt Ves.
68
Other Imported 1
B-5
207
B-11
137
B-7
209
C-2
214
D-0
215
D-3 or 4
188
Coyotlatelco
166
Other Dec. Jar
38
Unkn. Ves. Frag.
60
Plain and Eroded
Bowl, Eroded
10
Bowl, Plain
11
Jar, Eroded
30
Jar, Plain
31
Jar, Thick Rim
167
Comal
45/46
Spoon
100
Decorated Bowls
B/E frag.
264
Red Rim
200
B-0
202
B-1
203
C-1
213
E-0
221/269
E-1
222
E-5
225
E-6
226
E-9
277
E-10
132
E-11
133
E-12
173
E-14
185
G-0
230
Basin of Mexico
XI XII XIII XIV
4
3
1
1
1
3
S/SW St. Mex. Unk.
VII VIII IX X
XV
1
1
2
1
2
1
1
2
2
1
1
3
1
2
1
3
1
1
1
1
1
1
1
1
1
1
4 9
2
16 4
6
3
1
1 3
2 1 5 5 1
2
4
10 3 2 9
5 2 1 1
4
1 1
1
1
1
2 1 1
1
1
1
2
1 3 1 1
1
1
1
4
1
1
2
1
1
1
1
1 1
1 1 1 1
1
1
1
2
2
1
1
2
3
1
1
1
1
1
1
1
1
2
1
1
2
N.
5
3
3
4
5
2
3
2
1
3
5
4
1
1
1
1
1
1
1
8
35
12
50
17
12
1
4
2
3
1
6
2
4
8
2
4
1
1
1
2
Table 4.5 Ceramic petrographic sourcing group to project ceramic type correlations
137
Local
Type I-A I-B I-C I-D II III VI
Decorated Jars
B/E frag.
266
B-0
253 1
E-2
256
2
Ritual
267 6 2
Biconical Censer
Scored Censer
101/180 1 2
Sahum., Local
262
1
Censer, Other
170
Crude Unfinished 134 1 2
Handle/Support
169
1
Total
Basin of Mexico
XI XII XIII XIV
S/SW St. Mex. Unk.
XV
VII VIII IX X
N.
1
1
2
2
1
1
5
2
1 2
1
1
1
2
1
1 3 1
1 1
43 31 35 12 19 46 3
10 9
7
12
6
3
1
1
5 13
14
13
6
2
2
12
3
268
1
Bowls unless otherwise noted
Table 4.5 (continued) Ceramic petrographic sourcing group to project ceramic type
correlations
Group XIV (n=12) includes sherds generally falling within the range of variation
seen in the Basin of Mexico associated groups, but which could not be matched to a
specific group.
S/SW State of Mexico Petrographic Groups. Group VII consists of six decorated
sherds from low-frequency types: three incised redware pieces, one sherd each of C-1 and
C-2 polychromes, and one piece coded as Coyolatelco, which should probably more
accurately be identified simply as sloppy red-on-white. This group is characterized by a
relatively low level of inclusions, and inhomogenous (poorly mixed), optically active,
fabric. The sherds in this group show evidence of the addition of both deliberately sorted
sand temper, and grog temper from both local and non-local vessels. The petrographic
group has a strong correlation with INAA Group 7. All sherds in this petrographic group
138
are assigned to INAA Group 7, though there are also three Group 7 sherds not in this
petrographic group. This group is considered to be associated with regions to the south or
southwest of the Toluca Valley for purposes of further analysis.
Group VIII Group consists of three pieces of incised redware. This group is
generally similar to the preceding Mixed Imports group, but shows a larger fraction of
grog, somewhat less optical activity, and a mineral mix characteristic of a more felsic
source region. The inclusions in this group are bimodal, indicating deliberate tempering.
All three sherds in this group were assigned to INAA Group 8. This group is considered
to be associated with regions to the south or southwest of the Toluca Valley for purposes
of further analysis.
Group IX consists of one piece of incised redware, one E-0 Red-on-Buff bowl,
one crude unfinished bowl and two plain jar sherds. The sherds in this group are
generally similar to those in the Mixed Imports and Grog-Tempered groups, but show a
much higher degree of optical activity. Due to differences among the sherds in this group,
Dr. Meanwell believes it is unlikely that they were produced in the same location. The
samples show evidence for deliberate tempering, and an unusual (more felsic) mineralogy
when compared to the main local groups. Both sherds in this group were assigned to
INAA Group 8. This group is considered to be associated with regions to the south or
southwest of the Toluca Valley for purposes of further analysis.
Group X (n=13) consists of sherds which fell into the general range of paste,
temper, and production techniques characteristic of the S/SW State of Mexico groups, but
which could not be matched to a specific group. The group consists primarily of plain or
eroded types. Dr. Meanwell considers the S/SW State of Mexico groups provisional at
139
the individual group level and feels that some of them might combine into a single group
if a larger set of samples were examined.
Type Based Analysis
Because the Calixtlahuaca Archaeological Project coded all analyzed ceramics
into type categories, it is these types that ultimately need to be associated with probable
source regions for reasons of sample size. For both INAA and petrography, the random
ceramic samples by household can be used to directly evaluate the frequency of imported
ceramics in each household component. However, for both methods, these samples
include only a fraction of the total ceramics from the component. In addition, neither
method covers all of the households included in the final DS-1 sample. One household
component was defined after samples were selected for INAA (327-Ph6), and the random
household sample for petrography only covered the two households that were
continuously occupied through all three phases of the site’s history, 307 and 316. In both
cases, the relatively small number of samples per household (n=30 for INAA and N=22
for petrography) also creates a higher degree of random error. As a result, the data
produced by these specialized methods of analysis become more useful when they can be
extrapolated to the ceramic assemblage as a whole, bolstering regional identifications
based on type distributions from different areas.
The Calixtlahuaca Archaeological Project’s ceramic type classification system is
described in more detail in the preceding chapter on methods. I assigned types to one of
six regional proveniences (or as unknown) based on a combination of factors. For types
for which INAA or petrographic data are available, the type is assigned to the region
140
which dominated the sourcing samples (Table 4.2, Table 4.5), supported by regional type
affiliations in the published literature. For types for which no sourcing data are available,
or where only two samples were run and they divided between source regions, types are
assigned to a source region based on published descriptions of ceramics in the relevant
literature. This numbers of INAA and petrography samples run, relative to the
frequencies of types in the assemblage, can be seen in Table 4.6. Due to the strong
association between type frequencies in the overall assemblage and the dual sampling
strategy used, most types occurring in more than trace frequencies are well represented
among the samples selected for technical analysis. Excluding types with less than four
successfully sourced (INAA and/or petrography) samples only results in a 4-8%
reduction in total and/or rim sherd count per phase and the patterning of source regions
by phase remains consistent. This demonstrates that the minor types assigned to source
regions primarily (or exclusively) on the basis of the regional literature are likely
generally accurate and that the overall quantities of such types are not likely to have a
large effect on the type based analysis as a whole. Specific regional references are listed
in the subsequent sections on each region. While projects in adjacent regions have not
always published complete artifact catalogs, ceramic typologies are commonly included
in reports and can be used for comparative purposes. A complete list of ceramic types and
their regional assignments can be found in Appendix A. I present the general list of
source regions below, followed by a discussion of several special cases, where regional
assignment was more difficult. Due to the much higher sample size, the results of chisquare tests of phase and source region are highly significant for both rims and total
sherds (Rims - ²=470.64, df=6, p=<.000; All Sherds - ²=1662.57, df=6, p=<.000).
141
General Ceramic Source Regions. Source Unknown/Not Applicable. The type is
either not a vessel, or not from the Postclassic. For non-vessel items, function overrides
potential source in the classification system used by the project, so types may contain
both local and imported items (e.g., all figurines, including those likely of Aztec
manufacture, are classified under a single type code in the general ceramic database).
Vessels were generally assigned to “non-Postclassic” types on the basis of pan-regional
traits, and thus also contain an unknown mix of local and imported pieces. NonPostclassic types are quite rare.
Local. This regional group includes types manufactured in the general area the
Central Toluca Valley surrounding Calixtlahuaca, though not necessarily at the site itself.
This is the “default” geographic source region for all types, unless there is a good reason
to place them elsewhere. These types fall primarily (and often exclusively) into INAA
groups 1, 2 or 3, or petrographic groups I-III and VI. Types in this group are comparable
to those reported from the historic García Payón excavations at Calixtlahuaca (García
Payón 1936, 1979, 1981), and more recent projects at the Toluca Valley sites of
Teotenango (Tommasi de Magrelli 1978; Vargas Pacheco 1975), Metepec (Carbajal
Correa and González Miranda 2003), and Cerro Toloche (Jaramillo Lunque and De la
Peña Virches 2012, 2014).
Basin of Mexico. This regional group includes types associated with the Basin of
Mexico. Types in this group were assigned primarily to INAA groups 4, 5, or 6 and the
petrographic groups XI-XIV. It also includes types not sampled for technical analyses
which are stylistically characteristic of the Basin of Mexico. Types assigned to this group
are comparable to those described for the Middle and Late Postclassic Basin of Mexico
142
by Parsons (1966) and illustrated by Séjourné (1970, 1983). Common decorated wares
characteristic of the Basin of Mexico have been subject to extensive INAA sourcing
across multiple projects, both in and outside of the Basin (Garraty 2013; Hodge and Minc
1991; Minc 2009; Nichols, et al. 2009). While provincial copies of Aztec Black-onOrange and Texcoco molded sahumadors are fairly common, they can also be reliably
visually distinguished from pieces produced in the Basin of Mexico. Additionally, some
Aztec Black-on-Red was produced in Morelos, but it does not appear to have circulated
widely – outside of Morelos, it has only been identified on a low frequency basis in the
immediately adjacent southern Basin of Mexico. None of the sourced Aztec Black-onRed samples at Calixtlahuaca were from Morelos. Additionally, Chalco-Cholula
polychromes cannot always be distinguished between the two titular source cities, one
within the Basin and one in Puebla. However, as none of this type was recovered at
Calixtlahuaca, this does not create any potential for regional misattribution.
Split Local/Basin of Mexico type. This group includes types that split
approximately evenly between local and Basin of Mexico source groups in INAA or
petrographic analyses, or where there was a discrepancy between the expected regional
attribution and the results of a small number of sourcing samples. (Ladles are the
exception – of three INAA samples two are local and one is unknown. They are placed in
this category based on paste colors in the type collection, which are approximately evenly
divided between the buff-brown shades characteristic of local plainwares and the finer
orange-tinted shades characteristic of the Basin of Mexico.) For purposes of estimating
the frequency of imports, types in this group are assigned as 50% local, 50% Basin of
Mexico.
143
Type
Code Type Description
Local
168 Asymmetrical bowl
70 Copa, fragment
60 Unknown Vessel Frag.
72 Pitcher, other dec
184 E-16 Bowl
206 B-4 Bowl
228 E-8 Bowl
185 E-14 Bowl
173 E-12 Bowl
223 E-3 Bowl
169 Hollow support/handle
224 E-4 Bowl
0
Eroded uncertain
101/180 Scored Censer
71 Miniature vessel
253 B-0 Jar
170 Other censer
266 B / E fragment, Jar
204 B-2 Bowl
203 B-1 Bowl
256 E-2 Jar
222 E-1 Bowl
226 E-6 Bowl
133 E-11 Bowl
277 E-9 Bowl
225 E-5 Bowl
262 Sahum, Tol Valley form
255 E-0 fragment, Jar
213 C-1 Bowl
134 Crude unfinished
264 B/E fragment, Jar
221 E-0 Bowl
267 Hourglass censer
132 E-10 Bowl
200 Red rim frag, Bowl
167 Thick-rim Jar
202 B-0, Jar
30 Eroded Jar
10 Eroded Bowl
31 Plain Jar
11 Plain Bowl
N.
Sherds
N. INAA INAA N. Petrog Petrog N.
Rims N. Run Known N. Run
Known
5
6
38
46
11
15
21
20
23
26
439
44
3,280
228
89
503
127
552
89
83
1,906
98
128
318
267
248
256
961
245
498
959
1,384
1,828
1,360
487
679
1,975
40,923
3,097
53,183
5,090
3
4
8
10
10
11
11
11
12
16
18
21
33
38
39
41
43
45
47
71
75
76
78
102
104
106
123
149
157
227
332
357
441
483
487
660
862
1,515
1,676
1,923
2,050
2
1
2
1
2
1
1
1
2
1
0
1
7
1
2
5
1
9
2
22
19
4
5
27
21
11
10
19
21
14
7
20
22
21
28
26
59
47
88
84
4
1
2
3
1
8
2
20
15
4
5
19
17
9
6
18
10
10
5
15
20
17
23
25
42
36
70
59
1
1
1
1
1
1
3
3
6
1
1
2
1
6
1
1
2
1
5
4
1
2
8
2
5
4
0
1
2
8
2
12
4
11
4
13
4
2
17
3
12
8
50
35
13
4
2
16
3
11
7
50
33
Table 4.6 Number of samples sourced by type, with DS-1 total and rim sherd counts, and
INAA and petrography samples submitted and successfully sourced. Excludes types with
less than ten DS-1 rim sherds unless examples were sourced.
144
Type
Code Type Description
Basin of Mexico
68 Texcoco fabric marked
1
Aztec orange, untyped
42 Thin-walled Basin
67 Spinning bowl, Az. Or.
192 Guinda, variant B
114 Aztec III/IV
64 Sahumador, Texcoco
182 Aztec III/IV, Local B Bowl
113 Guinda goblet
21 Aztec III Bowl
181 Aztec III, Local B Bowl
208 B-6 Bowl
Morelos
(3 Total Sherds, None Sourced)
South/Southwest State of Mexico
209 B-7 Bowl
166 Coyotlatelco? Bowl
38 Other decorated, Jar
188 D-3 or 4 Bowl
214 C-2 Bowl
207 B-5 Bowl
137 B-11 Bowl
215 D-0 Bowl
230-232 G Bowls
Mixed Local/Basin of Mexico
274 Aztec III/IV, Local A Bowl
100 Ladle
271 Aztec III, Local A Bowl
45/46 Comal
Other Non-Local Regions
122 Otros tipos decorados
N.
Sherds
N. INAA INAA N. Petrog Petrog N.
Rims N. Run Known N. Run
Known
258
1
38
20
29
21
51
34
41
106
147
212
1
11
13
19
21
24
26
29
35
60
63
1
3
2
4
7
2
6
1
14
16
7
1
2
1
4
5
1
5
1
13
12
7
1
1
58
5
16
25
26
31
54
1
1
3
4
7
11
13
18
30
1
1
3
1
7
7
7
2
18
25
177
593
12
23
35
173
44
16
3
0
2
3
2
3
4
4
5
3
3
5
2
3
1
0
3
1
4
4
5
1
1
1
1
1
1
5
4
1
2
1
1
1
1
1
4
4
0
2
4
3
14
16
2
2
10
8
2
1
5
12
2
1
5
10
1
1
Table 4.6 (continued) Number of samples sourced by type, with DS-1 total and rim sherd
counts, and INAA and petrography samples submitted and successfully sourced. Excludes
types with less than ten DS-1 rim sherds unless examples were sourced.
South or Southwest State of Mexico. This group includes types assigned primarily
to INAA groups 7 and 8, or petrographic groups VII-X. These are generally minor
decorated types at Calixtlahuaca, particularly Red-on-White, complex Polychrome-on145
White, and incised redwares. In addition, types without INAA data that occur in higher
frequencies in museum collections from Malinalco and Ixtapan de la Sal (Smith 2003b),
historic excavations in the Valle de Bravo (Reinhold 1981), and/or regional survey near
the Mexico/Morelos/ Guerrero border (Arana 1990) are placed here.
Morelos. All of the types assigned to this group are stylistically characteristic of
Morelos. Stylistic descriptions of decorative types characteristic of Morelos are based on
Smith (Smith 2006a, 2006c), and have been supported by prior INAA sourcing work of
pieces recovered within Morelos. (While Guinda types were produced in both the Basin
and Morelos (Smith 2006c), the examples at Calixtlahuaca form a single INAA group
more likely matching the Basin of Mexico and are placed there.) INAA assigned seven
sherds from Calixtlahuaca to Eastern Morelos and one to Western Morelos (all from
group 7), but of the six of these that are decorated, none are stylistically consistent with
material from Morelos. Because Morelos is one of the only regions immediately south of
the Central Mexican Plateau with INAA reference samples, the samples in this study are
probably matching here by default rather than true association, and the associated types
are not placed in this general geographic region.
Other, more distant source regions. This group includes all other source regions,
including the Huasteca, the Balsas, and the Tarascan Empire. This group also includes
sherds coded as “other decorated type” during analysis, as these were pieces that did not
match any known type in the classification system and were therefore presumably not
from a local or immediately adjacent region. Assignments are based on decorative styles.
All of the potential types in this category are either extremely rare, or did not occur at
Calixtlahuaca.
146
Special Cases – Aztec Black-on-Orange. The correlation between the ceramic
classification type divisions for Aztec Black-on-Orange and the associated INAA and
petrographic results for these same types are a special case. At Calixtlahuaca, Aztec
Black-on-Orange occurs almost exclusively in the form of tripod bowls, most of which
are molcajetes. These vessels are distinguished from local-style molcajetes by their vessel
body shapes, support forms, rim forms, and pattern of incising on the interior. As a result,
they can be identified even in eroded and fragmentary states. However, “imitation” Aztec
Black-on-Orange was sometimes produced for local use in provincial parts of the Empire.
The Toluca Valley may have been one of these regions, as previous INAA samples from
the site of Tlacotepec from the Field Museum’s Starr collection showed two distinct
chemical groups of Aztec Black-on-Orange ware (McVicker, et al. n.d.). Unfortunately,
something about the sample preparation for these pieces left their results incompatible
with the Calixtlahuaca samples (See MURR report in Appendix D), so it is unknown
whether the two groups at Tlacotepec correspond to the Aztec INAA groups 4 and 5
identified at Calixtlahuaca.
It was recognized in the field that the Aztec Black-on-Orange ceramics recovered
at Calixtlahuaca had a continuum of pastes. The finer of the pastes were a good match for
traditional examples of the type from the Basin of Mexico. In addition, we also recovered
a range of coarser pastes, some of which fired to the traditional orangeish hue, and others
which were cream colored. These were sorted into three paste based categories: Basin of
Mexico (types 21 and 114) which visually matched pieces from the Basin of Mexico on
the basis of paste color and texture, Provisionally Local A (types 271 and 274) which was
made from a coarse, usually cream colored paste, and Provisionally Local B (types 181
147
and 182) which was orange colored, but coarser than would be expected from pieces
produced in the Basin.
These three type based classifications only broadly parallel the results of the
INAA and petrography groups. Aztec Black-on-Orange sherds were also assigned to
three INAA source groups: the primary local Group 1, the Basin of Mexico Group 5, and
the geographically unknown Aztec Group 4. Sherds assigned to the Basin of Mexico
source group 5 are almost exclusively from the “Basin of Mexico” paste types. Aztec
B/O sherds assigned to the primary local group are exclusively from the “Local Aztec A”
paste types, and are all from the cream-firing portion of this group. However, the third
INAA group, Unknown Aztec Group 4 includes Aztec Black-on-Orange sherds assigned
to all three paste groups. Despite the visual variation in these INAA Group 4 Aztec
sherds, petrography did place most of them into a distinctive grouping, Group XII.
The ware also exhibits temporal variation, including both Aztec III and the later
Aztec III/IV decorative variants. Both decorative variants occur in all three macropaste
categories. However, within the INAA source groups, there is a definite temporal bias,
with later decorative variants being more likely to be assigned to Group 4. Aztec III/IV
variants were identified conservatively, with ambiguous or eroded sherds coded as Aztec
III. As a result, the apparent temporal bias in the INAA groups may be overestimating the
degree of the shift because some of the Aztec III sherds are decoratively ambiguous. The
temporal bias in INAA groups is less pronounced, but still present when Aztec B/O is
considered by the phase of the context where each piece was recovered. Approximately
the same number of sherds from each paste group were selected for analysis from each
phase.
148
As a result of this ambiguity, the finer two categories of Aztec Black-on-Orange,
those originally coded as imported and uncertain (Local B), are considered to be imported
types in this analysis, while the types originally coded as local (Local A) are considered a
split type and divided between the local and Basin of Mexico source regions.
Special Cases – Comals Due to a combination of cultural and compositional
factors, comals are a special case at Calixtlahuaca. As a type, their distribution generally
parallels other stylistically Aztec types; they are almost completely absent during the
Dongu phase and increase steadily after that. They also never reach the frequencies seen
in the Basin of Mexico and Morelos, where comal sherds account for 15-30% of rim
sherds in household assemblages. (See Chapter 8 for a more extensive discussion of
comals in a regional context.) This would suggest that comals are certainly a non-local
idea, whether or not they are locally produced.
The compositional data for comals are ambiguous. Of the sixteen samples
submitted for INAA, five are local (in groups 1 and 2), and three are Aztec (in group 4),
and the remaining eight are unassigned. There are no other types represented by more
than four sherds that have this high an unknown assignment rate. Petrography was more
successful at assigning the samples to known groups. All petrographic samples were
assigned; six were local and two were imported. However, four of the six local
assignments were to group III, an otherwise relatively rare local group. Ethnohistorically,
the Florentine Codex separates comal production from that of other ceramics; for both
types of artisans and types of clay, comals are one of the only types of ceramics given a
separate entry (Sahagún 1950-82:Book 10: p. 83; Book 11: p. 256-257). If this means that
149
comals were produced by specialized potters, then it is not unexpected that they have
paste recipes outside of the norm. Due to the general regional evidence that comals are a
non-local type, and the mixed but local-trending compositional evidence, comals are
assigned to the Split Basin of Mexico/Local group.
Ceramic Exchange: Results
All three methods – INAA, petrography, and type-based source attributions –
show a pattern of increasing ceramic imports over time. This is accompanied by a
regional shift in the source of imports, with ceramics from the Basin of Mexico
accounting for a larger proportion of total imports over time. Imports from all regions
other than the Basin of Mexico and the South-Southwest State of Mexico are very rare
during all phases (Figure 4.4). The average number of non-local source regions per
household increases between the Dongu and Ninupi phases before decreasing to its
lowest level during the Yata phase. This is accompanied by a more uneven distribution of
access to ceramics from all sources during the Yata phase.
Because the INAA and petrography samples were stratified random samples
divided into bowls, jars, and other vessels, each of these categories must be weighted by
the frequency of the vessel form category within a component’s original ceramic
assemblage in order to estimate the actual frequency of different source groups. This
allows differences in vessel form frequency among components and time periods to be
accounted for in the analyses. The results of this weighting procedure for the INAA and
petrography samples can be seen in Tables 4.6 and 4.7.
150
Figure 4.4 Summary of imported ceramics by macroregion by phase, based on INAA (A),
petrography (B), types, rim sherds only (C), and types, all sherds (D).
151
Vessel
Unit Phase Form
307 Dongu Bowl
Jar
Other
Total
% Form
54.99
33.33
11.68
Basin of S/SW St.
Mex.
Local Mexico
54.99
0.00
0.00
33.33
0.00
0.00
11.68
0.00
0.00
100.00
0.00
0.00
Other
0.00
0.00
0.00
0.00
316 Dongu Bowl
Jar
Other
Total
53.63
34.95
11.42
53.63
26.21
11.42
91.26
0.00
0.00
0.00
0.00
0.00
8.74
0.00
8.74
0.00
0.00
0.00
0.00
307 Ninupi Bowl
Jar
Other
Total
62.39
29.72
7.88
31.20
29.72
6.57
67.49
7.80
0.00
0.00
7.80
15.60
0.00
1.31
16.91
7.80
0.00
0.00
7.80
316 Ninupi Bowl
Jar
Other
Total
52.96
37.26
9.78
39.72
27.95
9.78
77.44
13.24
9.32
0.00
22.56
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
307 Yata
Bowl
Jar
Other
Total
60.46
27.52
12.02
45.35
24.08
6.01
75.43
15.12
0.00
4.01
19.12
0.00
3.44
0.00
3.44
0.00
0.00
2.00
2.00
316 Yata
Bowl
Jar
Other
Total
54.95
34.39
10.66
27.48
30.09
7.11
64.67
27.48
0.00
1.78
29.25
0.00
4.30
0.00
4.30
0.00
0.00
1.78
1.78
95.63
72.47
70.05
0.00
15.18
24.19
4.37
8.46
3.87
0.00
3.90
1.89
Dongu Mean
Ninupi Mean
Yata Mean
Table 4.7 Ceramic petrography results by macroregion for each of six sampled
household components, with weighting corrections for vessel form frequencies for the
original component
152
Unit
Phase
1
307
315
316
320
323
324
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
303
307
308
311
316
322
Ninupi 83.67
Ninupi 85.55
Ninupi 80.15
Ninupi 90.51
Ninupi 100.00
Ninupi 80.77
307
309
316
317
324
Yata
Yata
Yata
Yata
Yata
2
3
4
5
95.00 0.00 0.00 0.00
63.83 0.00 36.17 0.00
91.06 0.00 8.94 0.00
91.76 4.12 0.00 0.00
53.14 22.41 0.00 4.65
17.14 45.71 0.00 11.43
5.00
0.00
0.00
0.00
0.00
0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 4.12 0.00
0.00 4.65 15.15
0.00 11.43 0.00
95.00 5.00 0.00
100.00 0.00 0.00
100.00 0.00 0.00
95.88 0.00 4.12
75.54 4.65 19.80
62.86 11.43 11.43
1.38
1.97
8.40
2.94
0.00
4.62
4.75
0.00
3.06
0.00
0.00
0.00
0.00 4.75
0.00 0.00
0.00 0.00
0.00 0.00
0.00 0.00
4.51 10.10
0.00
0.00
0.00
0.00
0.00
0.00
89.11 6.13 4.75
98.03 1.97 0.00
88.54 11.46 0.00
97.06 2.94 0.00
100.00 0.00 0.00
80.77 9.13 10.10
79.04 12.49
86.68 13.32
90.84 9.16
90.72 4.64
100.00 0.00
5.45 0.00
0.00 12.48
0.00 8.40
1.81 4.75
0.00 0.00
0.00 0.00
6
7
8
Local
BoM S-SW
63.92 10.08
77.81 0.00
82.71 3.55
86.40 4.33
89.10 10.90
5.04
8.88
4.58
0.00
0.00
7.45
4.44
4.58
4.64
0.00
5.04
4.44
0.00
0.00
0.00
0.00
4.44
4.58
0.00
0.00
5.04
0.00
0.00
0.00
0.00
3.44
0.00
0.00
4.64
0.00
Dongu Mean
Ninupi Mean
Yata Mean
68.66 12.04
86.77 1.21
79.99 5.77
7.52
4.27
3.70
2.68
3.22
4.22
0.83
1.30
1.90
0.00
0.75
1.80
3.37
2.48
1.01
2.52
0.00
1.62
88.21
92.25
89.46
3.51
5.27
7.92
5.89
2.48
2.62
Dongu StD
Ninupi StD
Yata StD
30.49 18.65 14.48
7.48 2.20 5.28
9.93 4.61 3.77
4.67
2.97
2.67
2.04
2.09
2.60
0.00
1.84
2.47
4.50
4.19
2.25
6.18
0.00
2.25
15.40
7.37
7.60
4.54
4.43
5.59
8.14
4.19
3.84
1.74
0.92
0.63
2.45
1.60
1.37
1.34
1.69
2.24
2.45
2.45
1.37
0.17
0.08
0.08
1.29
0.84
0.71
1.38
1.69
1.46
Dongu CoV
Ninupi CoV
Yata CoV
0.44
0.09
0.12
1.55
1.82
0.80
1.93
1.24
1.02
1.39
8.48
0.00
0.00
4.64
0.00
Table 4.8 INAA ceramic group frequencies by household component, after weighting by
vessel form frequencies in the total original assemblages to account for stratification by
form in the original sample. Excludes unassigned samples.
153
Both specialized samples include a percentage of sherds that did not match one of
the groups described above. These are treated differently for the INAA and petrography
samples. The unidentified INAA samples likely contain a mixture of imported sherds
from areas with too small a sample to form an identifiable group, and local ceramics with
somewhat anomalous paste recipes. This is supported by MURR’s description of the
unidentified sherds, the varied petrographic placements of the INAA-unknown sherds
submitted for both techniques, and the diversity of types included in the category.
Attempts to use discriminant analysis to assign these sherds to existing groups were not
successful; unassigned sherds from types that generally otherwise showed a strong
association with groups from one particular region were not consistently assigned to that
region. As a result, the unknown INAA samples are excluded from further calculations.
This matches the usual treatment of unassigned samples in INAA studies of Aztec
ceramics (Garraty 2006:Tables 6.1, 7.1).
In contrast, the much smaller fraction of petrographically unknown sherds are
generally either plainwares or types that would reasonably have been expected to have
been imported. Because the petrography included macrogroup level categories for sherds
that could be matched only at the regional level, rather than the group level, any sherds
remaining outside of these categories are highly likely to be from a region outside of
those represented in the numbered groups. The presence of plainwares in this unknown
category does not negate their foreign status; both INAA and petrography placed
substantial numbers of plainware sherds in the Basin of Mexico and S/SW State of
Mexico categories, demonstrating long distance trade in plainwares as well as in
154
decorated types. As a result, the petrographically unassigned sherds are treated as imports
of uncertain provenience.
Import Quantities and Sources
All three ceramic analysis methods, INAA, petrography, and type classification,
show similar patterning in the quantity and sources of imported ceramics at Calixtlahuaca
over time. However, there is some variation in the specific quantities and timing of
changes between methods.
The INAA data show only minor fluctuations in the frequency of local ceramic
production, which is addressed further in the next chapter. This method shows a steadily
increasing frequency of sherds in Aztec-associated groups, more than doubling over the
course of the site’s occupation. This is accompanied by an initial decrease in ceramic
imports from areas in the south or southwest of the State of Mexico between the Dongu
and Ninupi phases, followed by a relatively stable level of imports between the Ninupi
and Yata phases. In the INAA sample, the total percentage of imported ceramics is
highest during the final Yata phase, after the Aztec conquest of the site. However, as
noted previously, the chi-square tests of the original sample count results do not show
that the trends in the INAA data are strongly significant, and the calibration by vessel
form does not change the percentages of the overall assemblage attributed to each source
enough to change this.
The similarly weighted petrographic data show a sharp increase in the overall
ceramic imports between the Dongu phase and the two following time periods, from
about 5% non-local ceramics to 28% in Ninupi and Yata phases. This is much higher than
155
the total imported group frequency in the INAA samples, suggesting that either the local
petrographic groups represent a smaller geographic area around Calixtlahuaca than the
corresponding INAA groups, or that a majority of the INAA-unassigned samples are nonlocal. Aztec imports increase from each phase to the next, starting at completely absent
during the Dongu phase and reaching a high of 22% by the Yata phase. Imports from the
S/SW of the Toluca Valley rise from 4 to 8% from the Dongu to Ninupi phases before
dropping to less than 2% during the Yata phase. This latter pattern of imports contrasts
with those produced by other methods, but is consistent with the INAA results for
households 307 and 316 only, suggesting that it is due to the particular households
selected for petrography. Imported sherds of unknown origin show the same pattern as
those from the S/SW of the Toluca Valley, increasing from the Dongu to Ninupi phases,
before decreasing to an intermediate level during the Yata phase. In this case, the
calibration by vessel form makes the changes in import frequencies over time more
pronounced than they were in the original data set, suggesting that the already significant
chi-square results for the raw counts of petrographic samples underestimate the
significance of the calibrated results.
Import patterns based on the general ceramic classification were calculated for
both rim sherds and all sherds (Table 4.9). The two show similar patterns, though the
absolute percentage of imports is about 60% lower for the all sherd count calculations.
Because jars, which are mostly local, add more body sherds per vessel than bowls, which
account for most of the imports, this is not an unexpected pattern. Types associated with
the Basin of Mexico increase from each period to the next, with a particularly large
increase between the Ninupi and Yata phases. In both samples, imports from the S/SW of
156
State of Mexico drop between the Dongu and Ninupi phases. In the rim-based sample,
imports from this region continue dropping during the Yata phase, while in the all sherd
calculations, the frequency of imports from the S/SW is about the same between the
Ninupi and Yata phases. Imports from any other region are very rare during all phases,
but reach their highest level during the Ninupi phase. Based on two-tailed t-tests of
means of the percentages of each source group by household component, the increases in
imports from the Basin of Mexico are statistically significant between all pairs of phases
(Rims - p>.90, All Sherds – p>.95). In contrast, none of the between-phase changes for
imports from the S/SW region were statistically significant at the .90 level, mainly due to
the high degree of variability among households within each phase.
157
Rim Sherds
All Sherds
Unit
307
315
316
320
323
324
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
Local
98.54
98.80
98.62
98.10
95.92
87.01
Other
BoM S/SW Import
0.97 0.24 0.24
0.33 0.58 0.29
0.00 1.38 0.00
0.89 1.01 0.00
0.48 3.48 0.12
1.30 11.69 0.00
Local
99.54
99.77
99.45
99.58
99.22
98.70
Other
BoM S/SW Import
0.30 0.15 0.02
0.06 0.11 0.06
0.32 0.23 0.00
0.29 0.13 0.00
0.08 0.61 0.09
0.17 1.13 0.00
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
97.92
98.75
97.87
96.23
96.97
91.55
0.49
0.73
1.83
2.89
2.60
7.49
1.47
0.47
0.30
0.70
0.16
0.48
0.12
0.05
0.00
0.18
0.27
0.48
99.51
99.52
99.07
99.00
98.93
98.39
0.20
0.33
0.84
0.80
0.96
1.43
0.25
0.13
0.08
0.14
0.08
0.06
0.04
0.02
0.00
0.06
0.04
0.12
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
91.62 7.93
84.34 15.66
91.47 8.53
91.07 8.57
96.76 2.68
100.00 0.00
0.44
0.00
0.00
0.27
0.56
0.00
0.00
0.00
0.00
0.09
0.00
0.00
96.11
94.19
97.11
96.20
99.31
99.68
3.80
5.75
2.85
3.13
0.60
0.32
0.09
0.07
0.04
0.53
0.09
0.00
0.00
0.00
0.00
0.14
0.00
0.00
Dongu Mean
Ninupi Mean
Yata Mean
96.17
96.55
92.54
0.66
2.67
7.23
3.06
0.60
0.21
0.11
0.18
0.01
99.38
99.07
97.10
0.20
0.76
2.74
0.39
0.12
0.14
0.03
0.05
0.02
Dongu StD
Ninupi StD
Yata StD
4.61
2.60
5.38
0.48
2.55
5.44
4.38
0.46
0.25
0.13
0.18
0.04
0.38
0.42
2.09
0.12
0.45
2.04
0.41
0.07
0.20
0.04
0.04
0.06
Dongu CoV
Ninupi CoV
Yata CoV
0.05
0.03
0.06
0.72
0.95
0.75
1.43
0.77
1.18
1.21
0.96
2.45
0.00
0.00
0.02
0.57
0.59
0.74
1.04
0.55
1.46
1.34
0.88
2.45
Table 4.9 Ceramic type classification based import frequencies by household component
and source macroregion (Local,Basin of Mexico, South-Southwest State of Mexico, Other
Region), with rim sherd and total sherd based values.
158
The total quantity of imports also increases over time in all four samples, though
there are variations in the timing of the change. The INAA and rim sherd type analyses
show similar total imports during the Dongu and Ninupi phases, followed by a sharp
increase during the Yata phase. In contrast, the petrography shows the sharp increase
occurring between the Dongu and Ninupi phases, and the total sherd-type analysis shows
a steady progression in increasing imports from each phase to the next. I consider the first
pattern to be the most accurate, given the broader sample compared to the petrography
and the lower influence of differential fragmentation rates based on vessel size when
compared to the total sherd sample.
Inter-Household Variation in Ceramic Exchange
In addition to the site-level patterns of ceramic exchange over time, the household
components within each phase also vary in their quantities and sources of non-local
ceramics (Figure 4.5, Figure 4.6). Component level data are available for six components
based on petrography (Units 307 and 316 for each phase), all components except 327-P6
for INAA, and all components for type-based analysis. This discussion draws primarily
on the INAA and type-based datasets due to their more extensive coverage, with
reference to the petrographic sample where appropriate. (Table 4.7, Table 4.8, Table 4.9)
159
Figure 4.5 INAA ceramic imports by household component. Values from Table 4.6.
Figure 4.6 Rim sherd type-based ceramic imports by household component. Values from
Table 4.9.
160
Not unexpectedly, the INAA samples have a higher proportion of household
components in which imports from one or both non-local source regions are completely
absent when compared to the type-based data. Given the average frequency of imported
ceramics, where each source region accounts for an average of 2-6% of the ceramics for a
given household component, and a sample size of thirty sherds, each source region would
only be expected to be represented by one or two sherds in the INAA sample, making it
reasonable that sampling would occasionally miss a region completely. Despite this, there
is a high level of agreement at the household component level between the INAA sample
and the type based ceramic analysis. There is a pattern of increasing within-phase
differentiation over time. Both datasets have steadily increasing coefficients of variation
for the total quantity of imported ceramics over time. This suggests that access to
imported ceramics was becoming increasingly differentiated over time. The two datasets
also tend to identify many of the same households as having relatively high or relatively
low frequencies of imported ceramics, as well as ceramics from specific source regions.
At the same, the diversity of sources per household increases between the Dongu
and Ninupi phases, before decreasing to the lowest overall level during the Yata phase.
Based on the typological analysis of all rim sherds, during the Dongu phase, all but one
household had ceramics from more than one of the three primary import sources (Basin
of Mexico, S/SW of the Toluca Valley, and Other Regions) and three had imported
ceramics from all three groups. During the Ninupi phase, five out of six household
components had imported ceramics from all three import sources, and the sixth had
imports from two. However, despite the higher overall levels of imported ceramics during
the final Yata phase, one household component has no imported ceramics, two have
161
imports exclusively from the Basin of Mexico, and only one has ceramics from a region
other than the Basin of Mexico or S/SW of the Toluca Valley. When this is calculated as
the average number of source regions per phase, Dongu phase components have an
average of 2.3 sources per component, Ninupi components have 2.8, and Yata phase
components have only 1.5. The typological analysis of all sherds, which is more likely to
identify rare categories of imports due to the larger sample size, shows a similar (though
higher overall) pattern, with the average number of import source regions per household
going from 2.5 to 2.8 to 2.0 over the three phases. Collectively, these data show that the
occupants of Calixtlahuaca saw a widespread increase in the diversity of ceramic
exchange between the Dongu and Ninupi phases. This was followed a reduction in the
diversity of sources present during the Yata phase, as well as a much more uneven
distribution of imported ceramics. This is consistent with the site’s increasing integration
into broader regional exchange networks prior to Aztec rule, followed by an interruption
of such trade under Aztec rule with a potential shift toward more network-oriented
provisioning strategies as households relied on personal, rather than market connections
for non-local goods.
Obsidian Exchange
A sample of obsidian was submitted to Dr. Adrian Burke at the University of
Montreal for experimental high-resolution sourcing using XRF. As a result of the
technical limitations of the size of samples that the facility’s equipment could use and the
smaller database of reference samples, a significant portion of the samples either could
162
not be processed or could not be matched to a source. The remaining samples, as well as
a few successfully sourced samples from each source were then analyzed via XRF at
MURR. The overall sources and source frequencies are very similar to those from the
more systemically random sample survey obsidian samples processed at MURR
(Glascock 2012; Novic 2015), when phase length is taken into consideration (Table 4.9).
(Due to the much longer length of the Dongu phase, it presumably contributed a larger
proportion of the surface material at the site than the shorter Ninupi and Yata phases.)
Context
Dongu
Ninupi
Yata
All excavated,
weighted by phase
Survey
Pachuca Otumba
15.54
34.11
34.88
21.66
47.49
22.97
25.32
20.05
Ucareo
45.57
35.25
28.08
Other
4.77
8.21
1.45
40.27
45.80
4.71
2.90
29.70
31.30
Table 4.10 INAA-based obsidian source percentages for excavated phases and survey at
Calixtlahuaca
Sampling Calibration
Due to the sampling issues described in the methods section, the sourced obsidian
does not provide a representative sample of individual households. It does, however,
include enough artifacts to estimate overall source frequencies for each phase. While the
phase-wide samples are not as unbiased as might be hoped for under an ideal sampling
strategy, they do represent a wide range of artifact types within each technological
category, as well artifacts from multiple households during each phase. Corrections to
account for sampling bias by tool type are described below.
163
Technology
Dongu
Blade-Core
Bifacial
Bipolar
Flake-Core
Unidentified
Ninupi
Blade-Core
Bifacial
Bipolar
Flake-Core
Unidentified
Yata
Blade-Core
Bifacial
Bipolar
Flake-Core
Unidentified
ZacualCruz
ZinaEl
Santa
Otumba Ucareo tipan Paredon Negra pecuaro Paraiso Enchisi Teresa
1
10
2
1
4
34
5
9
34
2
1
1
4
3
3
4
4
1
12
2
2
2
1
1
1
1
1
1
1
1
1
1
Table 4.11 Grey obsidian sourcing results with sample results divided by phase and lithic
production technology. Includes both Burke and MURR sample results.
Calculating obsidian source frequencies from the original sample results requires
two data transformations. The first corrects the relative frequencies of grey obsidian
sources for the effects of correlation between sources and production technologies (Table
4.11). This problem occurred because particular technologies were strongly correlated
with particular sources, the frequencies of particular technologies varied by phase, and
because lithic production technologies had not been sampled proportionally to their
occurrence within each phase. I calculated grey source frequencies individually for each
lithic technology and then weighted them by the average household percentage of that
164
lithic technology within all of the DS-1 grey obsidian for the phase. This is the same
procedure used to correct for the presence of stratification by vessel form within the
ceramic INAA and petrography samples. It produces a matrix of values, in this case
unique combinations of technologies and sources, which collectively add up to 100%.
Once this matrix is created, the values for all technologies for each source can be
summed to give a corrected source frequency out of the total grey obsidian. The results of
this calibration step can be seen in Table 4.12.
Second, the sampled artifacts only included a few pieces of green obsidian,
because such pieces can be reliably identified visually and in Central Mexican contexts
almost always source to Pachuca, Hidalgo. As a result, I use the visually identified
percentage of green obsidian as the frequency of the Pachuca source in the assemblage.
Because the proportions of the grey sources can only be calculated at the phase, rather
than household component, level, I use the average percentage of green obsidian in DS-1
household components as the frequency of Pachuca material for that phase. The grey
obsidian source frequencies produced by the previous step are then each proportionally
reduced to express only the non-green percentage of all the obsidian, rather than a full
100% of grey obsidian produced by the previous step. This second step produces the fully
calibrated values for each source for each phase previously seen in Table 4.10.
165
Technology
Dongu
Blade-Core
Bifacial
Bipolar
Flake-Core
Average
% Hh Grey
Obs, DS-1
% of Tech. Sourced Samples
Calibrated % of Grey
Otumba
Ucareo
Other
36.01
17.92
4.02
0.17
2.78
100.00
66.67
100.00
94.44
0.00
33.33
0.00
2.78
0.00
0.00
0.00
1.00
17.92
2.68
0.17
34.01
0.00
1.34
0.00
1.00
0.00
0.00
0.00
Unidentified
Total
Ninupi
Blade-Core
Bifacial
Bipolar
Flake-Core
41.87
44.44
44.44
11.11
18.61
40.38
18.61
53.96
4.65
5.65
41.03
17.47
1.26
0.24
12.20
60.00
50.00
0.00
82.93
13.33
50.00
0.00
4.88
26.67
0.00
100.00
5.00
10.48
0.63
0.00
34.02
2.33
0.63
0.00
2.00
4.66
0.00
0.24
Unidentified
Total
Yata
Blade-Core
Bifacial
Bipolar
Flake-Core
40.00
42.86
42.86
14.29
17.14
33.26
17.14
54.13
5.71
12.61
46.97
10.61
8.84
0.00
23.53
66.67
100.00
0.00
70.59
33.33
0.00
0.00
5.88
0.00
0.00
0.00
11.05
7.07
8.84
0.00
33.16
3.54
0.00
0.00
2.76
0.00
0.00
0.00
Unidentified
Total
33.59
50.00
50.00
0.00
16.79
43.75
16.79
53.48
0.00
2.76
Otumba Ucareo
Other
Table 4.12 Primary obsidian sources by phase, showing correction procedure based on
production technology frequencies by phase to adjust for lithic production technology
related bias in original sample selection.
Obsidian Exchange: Results
The results of the analyses of obsidian show three patterns over time at
Calixtlahuaca, related to major sources, minor sources, and the total volume of obsidian.
First, there is a shift in the relative frequencies among the major sources of Ucareo,
166
Otumba, and Pachuca at the site (Table 4.10). The Ucareo, and to a lesser extent Otumba,
sources decrease over time. Pachuca increases over time. Ucareo and Pachuca are both
high quality sources that can be used to make prismatic blades, so the replacement of
Ucareo by Pachuca is a technologically reasonable substitution. In contrast, Otumba is a
medium quality source used primarily for biface production (Cobean 2002; Hirth 2006a).
The shift from Tarascan-controlled Ucareo to Aztec-controlled Pachuca was likely a
result of shifting political conditions in the Toluca Valley, though it should be noted that
the shift begins prior to the Aztec conquest of the Toluca Valley.
The second pattern concerns the diversity of sources over time. Calixtlahuaca is
an unusual site for Postclassic Central Mexico in that no single obsidian source ever
accounts for the majority of the assemblage, despite the chronological shifts in source
frequency. Most Middle and Late Postclassic sites in the Basin of Mexico and Morelos
are heavily dominated (80+%) by green obsidian (Table 4.13) (Smith 1990), and the rare
exceptions are usually dominated by Otumba. The grey fraction of the assemblage is
usually dominated by either the Otumba or Paredon source areas (Braswell 2003). In
contrast, sites within the Tarascan area rarely have more than an occasional 1-2% of
obsidian from the Pachuca source.
167
Site
Toluca Valley
Calixtlahuaca1
Calixtlahuaca1
Calixtlahuaca1
Santa Cruz Atizapan
Santa Cruz Atizapan
Basin of Mexico
Cihuatecpan2
Chalco
Chalco
Chiconautla
Xaltocan
Xaltocan
Xaltocan
Puebla-Tlaxcalla
Tlaxcallan
Morelos
Yautepec
Yautepec
Yautepec
Yautepec
Cuexcomate
Cuexcomate
Capilco
Capilco
Michoacan
Erongarícuaro
Urichu
Tzintzuntzan
Apatzingan
Oaxaca
Tututepec, House A
Tututepec, House B
Tututepec, House C
Nicayuhu, House 1
Nicayuhu, House 2
1
2
Total
% Obs/1000 Obs/100
Period Obsidian Pachuca
Sherds
Rims Source
MPC
LPC-A
LPC-B
LC
EPI
1,249
1,622
707
4,453
7,974
15.5
34.9
47.5
9.2
6.5
39.3
41.4
24.6
LPC
MPC
LPC
M-LPC
EPC
MPC
LPC
11,319
100.4
365
30.7
72.0
66.7
82.7
68.9
89.4
93.2
1,556
14.0
2,108
4,952
20,126
161
93.4
93.2
94.0
90.7
91.0
92.8
95.8
95.2
LPC
MPC
LPC-A
LPC-B
COL
LPC-A
LPC-B
LPC-A
LPC-B
LPC
LPC
LPC
PC
836
MPC
LPC
M-LPC
LPC-B
PC
838
281
71
27
15
0.7
8.6
0.3
03
43.50
37.73
22.54
Kabata 2009, Figure 4
Evans 1988, Table 1.2
Elam et al, 2008
Elson 1999, Table 5
184 Millhauser 2005,
170 Table 12.8
54
31.9
26.8
25.3
18.6
34.0
32.0
18.0
19.0
Millhauser et al 2015,
Table 1
Smith 2006c, Tables
C4-1 to C4-5
Earle and Smith 2012,
Table 10.6
Rebnegger 2013,
Table 5.17
33.4
13.8
16.4
12.2
0.7
2.3
Kelly 1947, Tables 1,
18.
Levine 2011, Table 3
Perez Rodriguez 2003
Tables 4.5, 5.3-4
Calix Values are for DS-1
Average of Structures/Operations excluding Op 8, Surface collection
3
Described as "all black obsidian". This contrasts with Hester et al 1973 (cited in Braswell 2003),
which sourced 7 of 17 pieces(41%) from the site to Pachuca.
Table 4.13 Comparative total and green obsidian frequencies for Late Classic to Early
Colonial sites in Central Mexico
168
In addition, the diversity of minor sources at Calixtlahuaca is also chronologically
sensitive. The Ninupi phase shows the highest diversity of sources, with 1-3 pieces from
each of six additional sources beyond the major sources noted above. These include at
least one newly located source within the Toluca Valley (Palomas), and one
geographically unknown source group that has only appeared at sites in the Toluca
Valley (San Antonio Enchisi) and is thus likely an additional local source. Both the
preceding Dongu and subsequent Yata phases have one minor source represented among
their samples. Because the Yata phase sample of grey obsidian is only about half the size
of that for the preceding phases, the disappearance of minor sources may not be quite as
severe as it appears, but it is certainly still present. In a metanalysis of Mesoamerican
obsidian sourcing, Golitko and Feinman (2015), showed that Late Postclassic sites (n=61)
have obsidian from an average of 3.74 (±1.98) sources. If this is used as a general point
of comparison, the Dongu and Yata phases fall very close to the macroregional average,
with four sources each. In contrast, the Ninupi phase features a diversity of obsidian
sources far above the macroregional average. The same analysis, which did not include
data from Calixtlahuaca, showed that Late Postclassic sites in West Mexico (i.e.,
Tarascan sites) generally form an almost completely independent obsidian exchange
network from the rest of Mesoamerica. The presence of substantial amounts of Ucareo
obsidian at Calixtlahuaca suggests that this conclusion may need to be reconsidered.
169
Unit
307
315
316
320
323
324
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
Lithic
Total Weight
Sherds
(g)
4,770
357
13,890
115
3,050
147
3,840
149
8,915
550
914
354
Total
Obs./ Green/
%
Lithic N. Total N. Green
1000
1000
Count Obisidan Obsidian Green Sherds Sherds
295
276
29 10.51
57.86
6.08
101
97
26 26.80
6.98
1.87
31
30
2
6.67
9.84
0.66
98
94
15 15.96
24.48
3.91
719
699
59
8.44
78.41
6.62
63
53
6 11.32
57.99
6.56
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
8,951
20,280
3,729
5,030
15,800
1,668
232
787
114
373
245
126
240
610
124
436
174
124
211
584
109
427
173
118
38
235
40
40
64
79
18.01
40.24
36.70
9.37
36.99
66.95
23.57
28.80
29.23
84.89
10.95
70.74
4.25
11.59
10.73
7.95
4.05
47.36
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
10,200
3,094
7,451
9,638
3,438
948
562
129
154
208
218
36
272
73
94
82
188
26
260
70
92
76
185
24
131
41
44
48
36
13
50.38
58.57
47.83
63.16
19.46
54.17
25.49
22.62
12.35
7.89
53.81
25.32
12.84
13.25
5.91
4.98
10.47
13.71
13.28
34.71
48.93
39.26
41.36
24.58
4.28
14.32
10.19
Dongu
Ninupi
Yata
Table 4.14 Green and total obsidian quantities by household component, with obsidian
weights, counts, and standardizations by rim sherd and total sherd counts. DS-1 sample.
170
Figure 4.7 Obsidian source frequencies per 1000 sherds by phase
Third, the quantity of obsidian arriving at Calixtlahuaca can also be addressed on
a more absolute basis by calculating the frequency of obsidian artifacts per 1000 ceramic
sherds (Table 4.14, Figure 4.7). As a primarily locally produced artifact class, the
absolute quantity of ceramics used by a household would have been mostly independent
of the volume of extra-polity trade in any given time period. In addition, ceramics should
have a relatively consistent average breakage rate (though different vessel types may
have different use lives), providing a reasonably consistent rate of accumulation. Given
the high degree of variation in artifact densities among deposits at Calixtlahuaca, sherd
counts are a better standardization factor than excavated volume. When obsidian counts
are standardized in this manner, they show a small increase in the absolute quantity of
obsidian entering the site between the Dongu and Ninupi phases, followed by a sharp
decrease from the Ninupi to Yata phases. When broken down by source, between the
171
Dongu and Ninupi phases, the absolute quantity of obsidian entering the site from
Pachuca and minor sources increases, while the amount arriving from both Otumba and
Ucareo decreases. Between the Ninupi and Yata phases, the absolute quantity of obsidian
arriving from all sources decreases. This final decrease includes even the Pachuca source,
despite the fact that it comes to represent a much higher percentage of the obsidian
assemblage during this phase, indicating that the apparent increase in the frequency of
green obsidian at the site is due more to a decrease in grey obsidian than an actual
increase in green obsidian.
Discussion
Some aspects of the trade patterns seen at Calixtlahuaca parallel those seen in
other provinces of the Aztec Empire, while others stand out as distinctive. First, the
pattern of substantial trade with the Basin of Mexico prior to Calixtlahuaca’s political
incorporation into the Aztec Empire parallels other cases in Central Mexico. For both
ceramics and obsidian, trade with the Basin of Mexico begins to increase during the
Ninupi phase, prior to the Aztec conquest of the Toluca Valley. In addition, while I
generally interpret the Yata phase as the period during which the site was under Aztec
rule, the phase actually begins two decades prior to the historically recorded date of
conquest. In a general sense, the discrepancy in timing between the evidence for changes
in trade and the historical dates for political control demonstrates the relative reach of the
Aztec economic and political nets (Hall 1997). It also demonstrates the ability of
exchange networks to reach beyond the political unit(s) at their cores. This accords well
172
with the historic and archaeological understanding of Postclassic Mesoamerica as a
commercial economy (Berdan 2003; Blanton, et al. 2005).
On the other hand, the specifics of the exchange patterns observed at
Calixtlahuaca also stand out as distinctive in some regards, namely for the volume of
trade (Table 4.15). The overall percentage of imported ceramics is much lower than that
observed at contemporaneous sites in Morelos. In addition, in Morelos, the period under
Aztec rule usually shows a small decrease in the frequency of imported ceramics when
compared to the preceding phase. The opposite occurs at Calixtlahuaca, where the
frequency of imported ceramics continues to rise after the Aztec conquest of the region.
At the same time, the frequency of extraregional imported ceramics is much higher than
at contemporary Aztec sites in the Basin of Mexico. Obsidian does show the more
stereotypical pattern, with a reduction in the absolute quantity of material entering the site
under Aztec rule, as is seen at a number of other sites in both the Basin of Mexico and
Morelos (See obsidian/sherd ratio data for Xaltocan, Yautepec, and Cuextomate in Table
4.13).
Both ceramics and obsidian reach their greatest diversity of sources during the
Ninupi phase, prior to Calixtlahuaca’s conquest by the Aztec Empire. As a result, the
increase in goods from the Basin of Mexico during the Yata phase was a mixed benefit
that came at the expense of the further development of trade connections in other regions.
173
Total % Total
Period Sherds Imported
Site
Toluca Valley
Calixtlahuaca
MPC
Calixtlahuaca
LPC-A
Calixtlahuaca
LPC-B
Cerro Toloche
M-LPC
Basin of Mexico
Chalco
EPI
Chalco
PC
Chalco
MPC
Chalco
LPC
Tlatelolco
LPC
Xaltocan
EPC
Xaltocan
MPC
Xaltocan
LPC
1
Cihuatecpan
LPC
Oxtotipac
MPC
Tepexpan
LPC
Maquixco
LPC
Teacalco
LPC
Xometla
LPC
Puebla
Cholula, UA-1, Midden EPC
Cholula, UA-1, Well 3 EPC
Cholula, UA-1, Well 1 LPC
Morelos
Yautepec
LPC-A
Yautepec
LPC-B
Cuexcomate
LPC-A
Cuexcomate
LPC-B
Capilco
LPC-A
Capilco
LPC-B
1
6,204
Total % Total
Rims Imported Source
0.62
0.93
2.90
1.98
3.83
3.45
7.46
455
1,442
2,288
2,101
12,303
0.11
867
3,731
2,335
96,881
0.13
1,007
1,961
4,000
1,856
562
3,858
262
445
Jaramillo & De la
Peña 2012, Table 2
0? Hodge 2008, Chap. 9
0?
0?
0?
Gonzalez Rul 1988a
4.04 Brumfiel 2005a,
2.22 phases per
0.64 Overholtzer 2014
Evans 1988, Table 1.2
2.48 Parsons 1966, Table
0.97 28
0.53
0.81
0.36
0.96
0.38
2.25
McCafferty 2001,
Tables 3.1, 5.2
5.6 (+3.4% Guinda) 2
3.3 (+2.5% Guinda)
3.4 (+3.9% Guinda)
2.6 (+2.8% Guinda)
3.3 (+2.2% Guinda)
2.3 (+1.7% Guinda)
Earle and Smith
2012, Table 10.8
Average of all excavations, excluding Operation 8, which is a specialized lithic dump
2
Guinda B/R or B&W/R are produced in both the Basin and Morelos, and cannot be visually
distinguished between the two regions. Chemical sourcing shows approximately a 50/50 split
in sources at these sites.
Table 4.15 Comparative frequencies of imported ceramics at Postclassic Central
Mexican Sites
174
Conclusions
The finer chronological control at Calixtlahuaca allows for a nuanced discussion
of the interplay between the political and economic control exerted by the Basin of
Mexico, with implications for the understanding of Aztec rulership strategies. The Dongu
to Ninupi phase trajectory of change in long distance trade establishes a baseline that can
be compared to subsequent, Yata phase, changes.
The pre-Aztec baseline for macroregional interaction at Calixtlahuaca begins at a
lower level than in most surrounding areas of Central Highland Mexico. During the
Middle Postclassic, the occupants of Calixtlahuaca participated in relatively low levels of
exchange with adjacent areas. While the Basin of Mexico is one of these areas, it is not
particularly more culturally similar to the Toluca Valley than any of the site’s other
trading partner regions. The low level of initial interaction differs from the pattern seen in
most of the regions that would become the inner provinces of the Aztec Empire, many of
which did have a higher degree of both intra-regional trade and general cultural
similarity. As a result, the degree of change in trade at Calixtlahuaca should be measured
from the site’s lower baseline, rather than by absolute volume.
The occupants of the Basin of Mexico exerted an increasing level of economic
dominance over Calixtlahuaca over the course of the Late Postclassic. This increasing
influence, is visible as an increased “market share” of the non-local goods reaching
Calixtlahuaca, starting during the Ninupi phase. Given that the earlier portion of this
phase predates the establishment of the Triple Alliance, let alone the conquest of the
Toluca Valley, the data indicate that the increasing dominance of the Basin of Mexico in
175
regional trade networks was not solely the result of imperial actions. The Ninupi phase
also sees the greatest diversity of sources for imported goods at Calixtlahuaca. Based on
this evidence of long distance trade at Calixtlahuaca, the pre-Aztec trajectory at the site
was one of increasing integration into broader Postclassic Central Mexico, with an
increasing diversity (and for lithics, increasing quantity) of goods reaching the site. This
increasing integration was relatively evenly distributed among households, indicating that
commoners were likely acquiring foreign goods primarily through markets, rather than
personal connections. The relatively even distribution of non-local goods is consistent
with a relatively collective form of rulership at the site prior to its incorporation into the
Aztec Empire.
The political influence on economic activity becomes visible later, during the
Yata phase. The quantity of imported ceramics continues to rise, but the absolute quantity
of obsidian arriving at the site drops. For both types of imports, the proportion of nonlocal goods coming from the Basin of Mexico rises. The changes in trade patterns
indicate that incorporation into the Aztec Empire partially disrupted the prior trend
toward increased interregional connection, redirecting it from an expanding mutilateral
web of regional connections to one where all further increases in trade were strongly
centered on the Basin. This indicates that Aztec rule was sufficiently direct to cause some
reorganization of trade in directions favorable to the imperial core. However, imperial
rule was not sufficiently direct to completely sever prior lateral trade connections, even
those reaching across the Tarascan border.
The Yata phase also saw a reduction in the evenness of access to non-local goods
from particular sources, suggesting a breakdown in the prior market system and a higher
176
degree of reliance on patronage or personal ties for access to non-local goods. This
interpretation would be consistent with a relatively non-collective rulership strategy by
the Aztec Empire, with a lower dependence on market revenues and a greater dependence
on patron/client relationships. The reduction in long distance trade at Calixtlahuaca could
also be a side effect of the Aztec relocation of the primary regional market, though this
would still indicate a primary focus on controlling local elites rather than providing
benefits to commoners.
The results of the analyses of long distance trade at Calixtlahuaca support the first
of the two hypotheses presented at the beginning of this chapter. That is, the data support
an interpretation of relatively indirect imperial rule. There is a dampening - but not a
complete negation - in the prior pattern of economic growth, and a reduction - but not a
complete disappearance - of goods from regions other than the Basin of Mexico. In
contrast, the increase in the dominance of goods from the Basin of Mexico and the
increased inequality of access to such goods, are not consistent with relatively collective
imperial rulership of the Toluca Valley.
177
CHAPTER 5
CHAPTER 5. “THE GRIDDLE MAKER IS ONE WHO MOISTENS CLAY”5:
LOCAL CRAFT PRODUCTION
Integration into larger economic systems is a two-way process and this chapter’s
discussion of local production at Calixtlahuaca is the flip side to the previous chapter’s
discussion of exchange networks and imported goods. As such, changes in local craft
production at Calixtlahuaca over time can be used to examine the degree and form of
Aztec control over the local economy as a product of particular imperial strategies of
rule. Some items produced locally at Calixtlahuaca would have exited the local economy
via medium or long distance trade, while the majority would have been used locally.
Despite this chapter’s focus on the production of objects, the relatively low density of
occupation at Calixtlahuaca and the generally low levels of craft production at the site
suggest that the majority of the site’s occupants were either full or part-time farmers. This
contrasts with the idea that Aztec cities were centers of craft production (Charlton, et al.
1993; Fargher 2009). This chapter presents the evidence for the production of ceramic
vessels, small molded ceramic items, obsidian artifacts, textiles and activities associated
with projectile points and scrapers at Calixtlahuaca over time.
I find that craft production at Calixtlahuaca was low throughout the site’s history,
with production occurring primarily for household use and little variation among
households. This indicates that while Calixtlahuaca was becoming increasingly integrated
5
General History of the Things of New Spain. Book 10: The People. Chapter 23.
178
into broader networks of interaction, market integration had not yet reached the level
where it could drive specialized production. Following the Aztec conquest of the site,
there is a modest increase in textile production, likely related to tax (tribute)
requirements, and modest reduction in lithic production, likely related to the increasing
scarcity of material noted in the previous chapter.
Empire and Economic Development
The economics of peasant households – such as Aztec commoners – sit at the
intersection of two major debates in the social sciences. The first is how pre-modern
world-systems were organized, and the degree to which various aspects of modern
international economic relationships can be applied to ancient cases. The second is the
degree to which factors other than economic self-interest influence economic behavior,
which might result in pronounced cultural differences in economic behaviors. Both
questions echo the long-running formalist-substantivist debate about the degree to which
economic behaviors are explained by generalizable laws versus culturally specific logics.
In more recent applications, these two questions pertain to the general question of
whether there was substantial economic growth and/or increases in the standard of living
in antiquity (Morris 2004a; Scheidel 2010; Stark, et al. 2016). If the potential for such
growth can be demonstrated, what political conditions support or impede it?
In the post-1492 world, political and territorial control have been intrinsically
linked to economic control. Colonies or provinces have been used as sources of raw
materials, and as controlled markets for manufactured goods produced in the home
179
country (Wallerstein 1976; Wolf 1982). Under this paradigm, manufacturing activities
are actively suppressed in provincial areas. However, Schortman and Urban (1994) argue
that this type of peripheral economic underdevelopment is only likely to occur in cases
where the core state controls trade within the boundaries of other trading partners, has an
ability to project military force into peripheral areas, and has more advanced technologies
of trade and/or transport. Only the first and second of these conditions are even weakly
met for the provinces of the Aztec Empire, including Calixtlahuaca. The Aztec Empire
could control trade with other areas to some extent, as was demonstrated by the shifting
patterns of imported goods in the previous chapter. However, there is clear archaeological
evidence that even the most supposedly antagonistic boundaries of the Aztec Empire let
goods through in both directions. Aztec sites feature goods from outside the empire, such
as “contraband” Tarascan obsidian and metal items, and sites in the Tarascan heartland
include low, but certainly more than insignificant percentages of Aztec Pachuca obsidian
(Pollard 2000). Cross boundary trade was not limited to the Tarascan frontier. Aztec
goods, including Black-on-Orange ceramics and Pachuca obsidian, also appear at sites in
Tlaxcala (Xiuhtecutli 2014). Militarily, the Aztec Empire did have the ability to mobilize
larger forces than the rulers of Calixtlahuaca, and the influx of Aztec settlers into the
Toluca Valley would have provided a check on independent local military action.
However, the regular occurrence of provincial revolts, possibly including in the Toluca
Valley, demonstrates that Aztec military power was also far from all-controlling. In terms
of technological or transportation advantages, the Basin of Mexico did not differ notably
from other parts of Central Highland Mexico. (The lake system in the Basin provides an
internal transportation advantage, but not one that can be applied to transporting goods
180
outside of the Basin.) There is a potential association between the strength of rulership
and the development of more modern-patterns of world systems-based economic
exploitation. As a result, establishing the directness and social organization of power of
rulership have broader implications for clarifying the validity and time depth of this
relationship.
There is a similar debate as to the degree to which extra-economic factors
influence the rationality of economic choices. In anthropology, this debate has
historically been framed as the argument between moral and political economy
perspectives, where the former perspectives argues that households (or individuals) are
primarily motivated by maintaining appropriate social relationships, while the latter
argues that households are primarily motivated by the rational pursuit of their own selfinterest (Cheal 1989; Wilk and Cliggett 2007). In the social sciences more broadly,
arguments concerning the influence of other factors on economic self-interest make up
part of the debate around rational choice theory (Eriksson 2011; Tucker 2014). This latter
body of work has found that what options are seen as viable and/or appropriate are at
least partially culturally determined (e.g. cultures vary widely in the average offers made
and rejected in the ultimatum game (Henrich, et al. 2004)), but that all human groups
cooperate at higher levels than would be expected from purely self-interested strategies
(Bowles and Gintis 2006). However, cooperation and group-benefitting behaviors are not
particularly inherently irrational, if one considers that humans live in social groups and
often reap collective benefits from the overall wellbeing of their group. On a more
applied level, the fundamental (maximizable) goal of most peasant households is the
continued existence of the household itself. In most premodern cultures (and most non181
first world modern states), insuring that your neighbors are doing reasonably well is a
form of self-interest, because one’s neighbors are one’s social safety net. This same focus
on the perpetuation of the household unit will also likely produce a relatively risk-averse
attitude toward economic choices (Kuznar 2001; Winterhalder, et al. 1999). However,
risk-averse does not mean unchanging, and modern peasant households often shift among
multiple strategies, including subsistence farming, cash crops, wage labor, and crafting,
based on nuanced understandings of risk and potential gain (Cashdan 1990; Wilk 1989).
In addition, risk-averse does not mean that households seek to directly supply all of their
own needs. At a minimum, the differential distribution of natural resources means that
most households have to trade for some necessary items not available locally, such as salt
or obsidian. The degree to which households seek to self-provision is likely related to the
reliability of alternate means of acquiring basic necessities (Hirth 2009a). This means that
household decisions concerning specialization can be used to provide information about
the reliability of alternate means of access to other goods, both at local and regional
scales, under particular political regimes.
Fargher’s (2009) argument for peasants as rational actors provides a set of testable
expectations related to both of the debates presented above. In this hypothesis, under
conditions of high agricultural demand, people in areas amenable to agricultural
intensification will specialize in agricultural activities (usually of staple crops) at the
expense of other types of production. People in agriculturally marginal areas, or areas not
amenable to agricultural intensification, then specialize in producing other types of craft
goods. These craft goods are then purchased by the agricultural specialists to replace
those that they no longer make for themselves. At this basic level, this type of
182
interregional interdependence occurs in both state and relatively high population density
non-state societies (Earle 1987; McGuire 1993; Sanders 1956; Stark 1991). In more
complex societies, both of the preceding groups supply urban areas, which focus on the
political/religious activities, and craft production. The Middle and Late Postclassic Basin
of Mexico conforms well to the predictions of this model, with agricultural intensification
in productive areas, and variable specializations in less productive zones (Charlton, et al.
1991; Fargher 2009). However, the Basin is a much smaller area than the other test cases
from which the model was derived. This has two possible implications. The lack of draft
animals in Mesoamerica may have limited the spatial zones over which the trade of staple
goods resulting from agricultural specialization occurred. Alternatively, the pattern of
subregional occupational specialization observed in the Basin of Mexico may also extend
to adjacent regions, such as the Toluca Valley. Integration over a larger geographic area
would correspond to Sanders’ (1956) idea of a Central Mexican Symbiotic Region.
Aztec rule over the Basin of Mexico was both more direct and more collective
than over most provincial areas, with the ability to cause changes in economic
organization (e.g. founding and moving market locations, and elite sponsorship of luxury
craft production), and the ability to maintain regular and reliable market access to allow
for specialization. If Calixtlahuaca does fall within a broader Central Mexican core area, I
expect the pattern of craft production at Calixtlahuaca to parallel those seen in the Basin
of Mexico. In this case, I would expect Calixtlahuaca to gain importance as an urban
center from the Dongu to Ninupi phases, which would be accompanied by an increase in
craft production. The reduction of administrative functions at the site between the Ninupi
and Yata phases, resulting from the reorganization of local power under Aztec rule,
183
would have required the occupants of Calixtlahuaca to focus on other productive
activities. Given that Calixtlahuaca is marginally located for maize agriculture, this
would likely have meant an intensification in the production of other goods.
In contrast, if Calixtlahuaca falls outside of the more highly integrated core of the
Aztec Empire, it would likely be under more indirect and, at the imperial level, less
collective rule. There is a potential issue of conflation between relatively collective forms
of local social organization, and relatively collective imperial rule. As a result, the preAztec pattern may be one of stable market access at the local level resulting in the
development of specialized craft production, especially given general regional trends
toward economic integration over the course of the Postclassic. However, given the
somewhat lower population density of the Toluca Valley relative to the Basin of Mexico,
I expect an overall lower level of specialization. Subsequently, under a relatively indirect,
network-oriented Aztec rulership strategy, I would expect to see relatively minor changes
to the preexisting pattern of craft specialization at the site.
The Economics of the Aztec Empire
Patterns of craft production over time are highly variable at Postclassic Central
Mexican sites. Archaeologists’ interpretations of changes in local craft production
resulting from the expansion of the Aztec Empire fall into two groups. These
interpretations largely reflect the degree to which researchers interpret the development
of the Basin of Mexico market system as resulting from primarily political as opposed to
primarily economic processes (Nichols, et al. 2009). Unfortunately, two of the sites with
184
the clearest evidence for specialization in particular activities, Otumba and Cihuatecpan,
do not have sufficient chronological refinement to discuss changes over time.
The first group interprets changes in craft production and market development as
heavily influenced by political processes (Brumfiel 1980; Garraty 2007), and often
interprets changes in a negative light. If local production rates drop, it is because the
economy is being swamped by foreign goods, or access to raw materials is being limited
or taxed. If the evidence shows increased craft production, it is a result of imperial
taxation either directly or indirectly extracting wealth out of the provincial community.
At Xaltocan, reductions in local lithic production (Millhauser 2005; Rodríguez-Alegría
2008), and increases in the frequency of spinning tools (Brumfiel 2005a) are both
interpreted as responses to imperial rule that were negative for local communities.
In contrast, the second group of interpretations sees most changes in craft
production after imperial conquest as the result of positive changes resulting from
increased economic opportunities (Charlton, et al. 2000; Hirth 2013; Smith 2003c). If the
evidence for the production of certain crafts increases, then it is a result of greater
opportunities to sell items in an expanded market network. Decreasing evidence for
particular crafts is also seen as a result of greater market dependence. Case studies falling
into this mode of interpretation include Smith’s interpretations of the increase in textile
production tools at the Morelos sites of Cuexcomate and Capilco (Smith and Heath-Smith
1993), and explanations for decreases in bark-beaters and molds for small ceramic items
at various sites in Morelos (Smith 2003c).
As a result, changes in the intensity and diversity of craft production must be
contextualized against broader changes in the standard of living. The patterns of craft
185
production explored in this chapter are compared against changes in household wealth in
the following chapter.
Mesoamerican Craft Production
Craft production in Mesoamerica was generally a part-time occupation for a large
portion of the population, rather than a full-time means of making a living (Brumfiel
1980; Hirth 2009a). This is especially true for the production of commonly used goods,
such as pottery or grinding stones. In contrast, very high value goods, such as metal items
or stone jewelry were often produced by more specialized, urbanized producers (Smith
2003a). As a result, changes in the distribution or intensity of the production of widely
used items are likely to be visible in consumption-based household contexts, while the
study of rarer items depends on encountering the primary contexts of production.
Mesoamerican studies of craft production can be framed around the concepts of
specialization and multicrafting. Specialization addresses the organization of production
of a single craft and can be broken down into multiple variables, as characterized in
Costin (1991). Her dimensions of specialization are context, concentration, scale, and
intensity. Context ranges from independent to attached and describes the independence of
the craftsperson relative to the consumer. Concentration describes the spatial patterning
of specialization, from dispersed to nucleated. Scale describes the size of the production
group, from small family-based groups to factories. Finally, intensity describes whether
crafting is a part time or full-time activity for the producer. The variables can
theoretically be combined in any set of combinations, but certain combinations seem to
186
be more common. In general, state-level social organization in accompanied by at least
intermediate levels of specialization, but this ranges from primarily family-based
workshops to factory/manufactory levels of organization, and from primarily private
enterprise to heavy state involvement in the production of goods.
Most Mesoamerican crafting falls into the independent, dispersed, kin-based, parttime end of Costin’s spectrums of variation, with occasional forays toward attached
specialization for higher-value crafts. This produces an appearance of homogeneity,
masking important variations in the organization of Mesoamerican craft production. As a
result, Hirth (Hirth 2006b, 2009a) and Shimada (2007) proposed two dimensions, the
periodicity of crafting and the number of crafts practiced, that better characterize the
variability of Mesoamerican craft production. The first of these two dimensions
distinguishes between craft activities that a household or individual may perform
discontinuously (such as seasonally, in years with bad harvests, or for an occasional
commission) from those that occur on a fairly continuous basis. The second distinguishes
between households or workshops where there is evidence for a single craft, and
locations where the craftspeople created multiple classes of objects. Multicrafting is
sometimes used to refer to any context with evidence for the production of multiple
crafts, but should more accurately be reserved for cases where there is evidence for the
production of multiple crafts above the level needed for use within the household.
Multicrafting is easier than periodicity to identify archaeologically.
At Calixtlahuaca, I consider Costin’s dimensions of specialization at both the
household and site level, with a particular focus on intensity. At the household level, I
seek to identify whether there is any evidence for a particular craft activity, and whether
187
there is sufficient evidence to consider a particular household as producing for more than
their own needs. The evaluation of the intensity of production is based on both relative
comparisons among households at Calixtlahuaca and comparisons with other Aztec sites.
Inter-site comparisons are also used to evaluate the overall intensity of production for
particular crafts at Calixtlahuaca from a regional perspective, as a level of craft
production that looks high at a local scale may not be unusual at a regional scale. Due to
the limitations of the data both from Calixtlahuaca and comparative sites, not all of the
comparisons described here can be used for all of the crafts under consideration.
Following the discussion of production and specialization by household, I discuss the
evidence for multicrafting.
Ceramic Vessel Production
Ceramic production occurred widely, usually as a part-time specialization in
Mesoamerica. Ceramic vessel production may be identified by direct evidence of
production, such a molds, kilns, or waster sherds, or via indirect methods based on the
frequency of artifact types and/or the characteristics of the produced artifacts themselves.
The production of non-vessel ceramic items is discussed separately in the following
section.
Direct Evidence for Ceramic Production
Direct evidence for ceramic production consists of the raw materials, tools and/or
facilities used to make ceramics or the debris resulting from production. Examples of raw
188
materials include deposits of raw clay, tempering material or pigments to be used for
decorative purposes. These are rarely recovered archaeologically, and even when found
can be difficult to link to ceramic production. In Mesoamerica, ceramic production tools
were simple, consisting of one-piece molds (in some time periods and regions),
burnishing stones (possibly including worked sherd disks), and small grinding stones for
pigments. The only facilities recovered in Mesoamerica are kilns, the use of which was
very rare (for exceptions, see Balkansky, et al. 1997; Castanzo 2004; Healan 1989:
Appendix 2; Santley, et al. 1989). Theoretically, settling basins for clay preparation could
also be identified archaeologically, as they have been in the Southwest (e.g., Abbott
2006). Direct evidence for ceramic production also includes the by-products of
production in the form of improperly fired sherds. These may take the form of highly
vitrified or warped sherds, or simply as unusual concentrations of sherds from a single
type of vessel (Arnold 1991; Curet 1993; Stark and Garraty 2004; Sullivan 2006).
There is no direct evidence for ceramic production at Calixtlahuaca. This is not
unusual in Mesoamerican contexts. Most ceramics were formed by hand or over simple
one-piece molds and then open fired at relatively low temperatures. Ethnographic work
has demonstrated that very small scale producers may fire ceramics in their household
cooking hearth, leaving no specific evidence of firing (Deal 1998). Larger firing areas are
also often located some distance from houses (Arnold 1991), and the Calixtlahuaca house
excavations may not have extended far enough to reach the appropriate locations. A few
heavily vitrified sherds were recovered from a number of excavations (Table 5.1), but
given all four of the excavation units where these sherds were recovered also included
quantities of burned daub, it is unclear whether the sherds were exposed to extremely
189
high temperatures during firing or at some later point in their use-life. According to
Levine, et al. (2015), a strong “local” sourcing signature, but poor correlation among
local INAA groups, local petrographic groups, and macroscopic ceramic characteristics
can be reasonably interpreted evidence of widespread, low-level local ceramic
production. Given the lack of one-to-one correlations between local INAA groups, local
petrographic groups, and specific decorative types at Calixtlahuaca this argument
provides additional support for widespread, low-intensity ceramic production at the site.
Provenience
315-12-8
323-13-1
323-13-6
316-16-2
316-17-2
316-20-1
316-22-2
316-6-1
316-14-2
316-15-2
317-12-4
317-14-4
317-14-7
317-35-5
Phase
Dongu
Dongu
Dongu
Ninupi
Ninupi
Ninupi
Ninupi
Yata
Yata
Yata
Yata
Yata
Yata
Yata
DS-1
X
X
X
X
X
X
X
DS-3
X
X
X
X
X
X
X
X
X
X
X
X
X
X
N. Sherds
1
1
1
1
1
1
1
1
1
4
1
1
1
2
Type(s)
Not recorded
Not recorded
Not recorded
Not recorded
Not recorded
Not recorded
Not recorded
Bowl
Jar, decoration unknown
Not recorded
Unidentifiable
Plain Jar
Not recorded
Not recorded
Table 5.1 Excavated contexts at Calixtlahuaca with overfired sherds. Includes phase and
project DS-1 and DS-3 sample assignments (See Chapter 3 for a description of samples
of contexts used for analysis)
Indirect Evidence for Ceramic Production
Changes in ceramic production can also be inferred from macro- or microscopic
characteristics of the finished pots themselves. Such characteristics provide information
190
about the general organization of producers at a site, rather than the specifics of
production by a particular household. I look at INAA for general sitewide patterns of
local production and attribute data for more detailed information on the production of
specific types.
Vessel
Phase Form
Dongu
Bowls
Jars
Other
Total
Ninupi
Bowls
Jars
Other
Total
Yata
Bowls
Jars
Other
Total
N.
% of DS-1
Samples
Rims
Local
Aztec S/SW Mex.
Imports St. Imports
72
45
16
54.57
34.44
10.99
47.93
32.87
10.99
91.80
2.95
0.00
0.00
2.95
3.69
1.57
0.00
5.25
84
57
21
57.53
34.39
8.07
54.92
32.06
6.24
93.21
1.96
1.17
1.47
4.59
0.65
1.17
0.37
2.19
63
42
17
56.65
33.16
10.2
48.56
31.58
9.60
89.74
6.29
0.79
0.60
7.68
1.80
0.79
0.00
2.59
Table 5.2 INAA based frequencies of local and imported ceramics by macroregion
(Local, Basin of Mexico, South-Southwest State of Mexico), showing adjustment for
vessel form frequencies during the phase.
INAA. The general frequency of local ceramic production can be addressed using
INAA data, which were described in more detail in the preceding chapter. The primary
local source group, Group 1, and its variants Group 2 and Group 3, make up the majority
of the sourced ceramics from all three phases of the site’s occupation. While this cannot
be taken as a precise measure of direct ceramic production at Calixtlahuaca, as some
191
ceramics in these groups likely came from communities in the surrounding area, it is a
broad-brush measure of local production. When the INAA data are adjusted by the
assemblage frequency of each major vessel type (bowls, jars and other vessels) to account
for the effects of the original sampling strategy, the local source groups show very little
change in assemblage dominance over time (Table 5.2). Excluding unattributed sherds,
the local groups account for 91.8% in Phase 2, 93.2% in Phase 4, and 89.7% in Phase 6.
Because most Mesoamerican INAA studies are focused on particular decorative
types, there is little comparative data as to whether the levels of local production at
Calixtlahuaca are particularly high or low. Of the cases that exist, Garraty (2007) shows
that approximately 73-83% of plainwares in the Basin of Mexico were produced in the
same major third of the Basin where they were used. When this figure is combined with
sourcing studies that show that 70-80% of Late Postclassic Black-on-Orange wares
(Nichols, et al. 2002) and 81-97% of Late Postclassic redwares (Minc 2009) were
remaining within similar geographic regions, the net use of locally produced ceramics
appears to be somewhat lower than what is seen at Calixtlahuaca.
Attribute Analysis. Increases in the scale or intensity of production may be visible
as a greater level of standardization among vessels of a given type. This may be a result
of a shift from a larger number of smaller producers to a few larger producers, an
increase in consistency due to increased skill/repetition, or due to the adoption of new
technologies to facilitate rapid production, such as molds (Rice 1987:201-204). However,
comparisons of pottery standardization in different ethnographic contexts have
192
demonstrated that only relatively large changes in production intensity are likely to be
visible in the standardization of the finished products (Roux 2003).
I examine the standardization of three ceramic attributes – the rim form, the rim
diameter, and the vessel wall thickness. These three attributes could be consistently
recorded for almost the entire attribute sample. In addition, all three are likely subject to
reduced variation as a result of production intensification. For rim diameter and vessel
wall thickness, increased standardization would likely be an unintentional product of
increased repetition, while a decreased variety of rim forms might relate to a decreasing
number of producers. I examine these variables in three ceramic types - plain bowls, plain
jars, and biconical censers - using the attribute sample of more intensively recorded
sherds from each household component. These types occur in all households during all
phases. The first two are very common types with large sample sizes. The latter is a
relatively less common but still ubiquitous type, potentially offering a contrasting pattern
of production, both because demand could have been met by a smaller number of
producers, and because the plaster and appliqué decoration of the censers is unique
among ceramic types at the site.
These three ceramic types generally do not show changes in the organization of
ceramic production over time (Table 5.3). The metric variables, rim diameter and vessel
wall thickness, are unimodally distributed for all three vessel types and their distributions
are evaluated using the coefficient of variation. The coefficient of variation measures the
distribution of variation relative to the mean. Because, in the absence of measuring
devices, humans psychologically perceive variation as relative to the total, this measure
provides an appropriate means of assessing changes in the degree of variation, even in the
193
presence of changing mean values (Eerkens and Bettinger 2001). Rim form is a
categorical variable, and is evaluated using the scaled version of Simpson’s C. Simpson’s
C is a diversity measure, and in the version used here, scores assemblages from 0 for
complete dominance by a single type, to 1 for a completely even distribution of types
among categories.
For jars, I include both plain and eroded jars in the analyses. Because most jars
are plain, most eroded jar sherds are likely to originally have been plain. The use of both
types results in a larger sample size. Jars do not show increasing or decreasing
standardization over time. The standard deviation for jar diameters decreases from the
Dongu to Ninupi and Yata phases, but this is accompanied by a general reduction in the
mean vessel diameter. As a result, the coefficient of variation decreases from the Dongu
to Ninupi phases, and then rises to an intermediate level in the Yata phase. For vessel
thickness, the coefficients of variation are very similar across all three phases, suggesting
either consistent production methods, or that thickness is a technological constraint that is
not affected by changes in the organization of production. The distribution of different
rim forms over time shows consistently increasing standardized Simpson’s C values,
indicating that the assemblages are less dominated by one or two rim forms over time.
This may indicate increased diversity of producers over time.
For bowls, I only included plain bowls in the analyses. As a high percentage of
the total bowls are decorated, the original nature of eroded pieces is much harder to
predict. Bowl diameters show a decrease in the coefficient of variation from the Dongu to
the Ninupi phases, followed by an increase to an intermediate level during the Yata
phase. Conversely, the coefficients of variation for bowl thickness show a near-opposite
194
pattern, increasing from the Dongu to Ninupi phases before returning to their original
levels during the Yata phase. The diversity of rim forms remains even during the Dongu
and Ninupi phases, followed by a small increase in heterogeneity during the Yata phases.
The relatively small changes in the amount of variation among phases, as well as the
highly inconsistent patterning among the three variables lead me to conclude that there
was no systematic change in the organization of bowl production over time.
Diameter
Thickness
Rim Forms
Phase
Jars
Dongu
Ninupi
Yata
Std. Co.
N. Mean Dev. Var.
Std. Co.
N. Mean Dev. Var.
N.
278 21.07
355 20.47
208 19.37
6.04 0.29
5.04 0.25
5.05 0.26
344
416
230
8.68 1.54 0.18
8.81 1.53 0.17
8.38 1.58 0.19
344
415
231
0.67
0.74
0.78
Bowls
Dongu
Ninupi
Yata
188 17.69
248 17.73
79 17.24
4.98 0.28
4.48 0.25
5.64 0.33
202
253
88
6.46 1.35 0.21
6.15 1.43 0.23
6.12 1.28 0.21
203
253
88
0.79
0.79
0.77
Censers
Dongu
Ninupi
Yata
36
41
20
6.89 0.30
7.62 0.28
5.67 0.24
39
43
22
7.54 1.29 0.17
8.19 1.74 0.21
8.14 1.39 0.17
39
45
22
0.78
0.79
0.79
23.36
26.90
24.10
Simpson's
C, Stand.
Table 5.3 Standardization measure values for ceramic bowls, jars, and hourglass
censers, by phase. Diameter means in cm, thickness means in mm.
Biconical “hourglass” censers are a much less frequently recovered type than
bowls or jars, though the type does have a high ubiquity among excavated contexts. The
rim diameters for this type show a decrease in the coefficient of variation from each
phase to the next. The coefficient of variation for vessel thickness, however, increases
195
from the Dongu to Ninupi phases, followed by a return to approximately its starting level
during the Yata phase. The Standardized Simpson’s C statistic for the diversity of rim
forms in almost identical in all three phases. For biconical censers, the relatively small
degrees of change in the variables examined for standardization, and the inconsistent
temporal patterning among variables suggests that there were no systematic changes over
time in how censer production was organized.
Molded Ceramic Items
Small non-vessel ceramic items, particularly spindle whorls and figurines, were
produced independently of ceramic vessels. The production of such items can be
identified either through the same markers of direct production used for vessels (prepared
clay, misfired pieces, etc.), or through the identification of the ceramic molds used to
produce the items in question. At other Aztec sites, particularly Otumba (Charlton, et al.
1993; Nichols, et al. 2000), there is clear evidence that the production of these artifact
types was conducted in workshops, and that the two artifact types were likely made by
the same producers. At the same time, molds only occur as a very low frequency item at
Middle and Late Postclassic Yautepec (.07-.19 molds per 1000 sherds) (Smith 2006c:
Table C3-2), and are completely absent at the rural sites of Cuexcomate and Capilco for
the same periods (Smith 2006a:Table B3-2). The only Postclassic site in the Toluca
Valley where molds are reported is Teotenango, with four whorl molds (Velázquez V.
1975). However, given that molds are a very low-frequency item, and most projects in
the Toluca Valley have featured either relatively low-intensity test-pitting or a
196
concentration on the excavation of monumental architecture, the general absence of
molds should not be taken as evidence for a lack of production. Mold-produced items,
especially spindle whorls, are common at sites across the Valley. Toluca Valley maguey
whorls are stylistically distinct enough from those produced in surrounding regions that it
is unlikely that they were being produced outside of the region and imported in bulk.
At Calixtlahuaca, there is little evidence for the production of small molded items
on-site. The excavations recovered four fragments of figurine molds (M-263, 264, 265,
567) and four unspecified mold fragments that may have been part of maguey whorl
molds (M-272, 504, 576, 582) (Figure 5.1). The molds are scattered among household
components and phases, occurring in six excavation units and all three phases (Table
5.4). At .02 molds per 1000 sherds in the DS-1 sample, the overall frequency of molds
falls below those noted above for the site of Yautepec. The sitewide survey located two
figurine molds in addition to those described above, a frequency which also does not
support intensive production of molded artifacts in non-excavated portions of the site.
Piece #
M-576
M-264
M-265
M-272
M-567
M-263
M-504
M-582
M-262
M-451
Mold Type
Other/Unknown
Figurine
Figurine
Other/Unknown
Figurine
Figurine
Other/Unknown
Other/Unknown
Figurine
Figurine
Provenience
314-5-1
315-17-2
315-22-10
303-1-7
311-5-1
307-14-1
317-13-4
317-18-3
302-4-1
300-831-2
Context
Excavation
Excavation
Excavation
Excavation
Excavation
Excavation
Excavation
Excavation
Survey
Survey
DS-1 DS-3
X
X
X
X
X
X
X
X
X
X
X
Phase
Dongu
Dongu
Dongu
Ninupi
Ninupi
Yata
Yata
Yata
-
Table 5.4 Molds for small ceramic items at Calixtlahuaca by type and provenience. See
Chapter 3 for a description of DS-1 and DS-3 samples of contexts.
197
Figure 5.1 Examples of figurine molds and spindle whorl molds excavated at
Calixtlahuaca
The frequency principle also fails to support the notion of specialized production
of these artifact types in any of the excavated households. None of the excavated
households have truly exceptional frequencies of either finished artifact class, which
might suggest on-site production (See Table 5.7 for whorl frequencies by household
component, and Figure 7.2 for figurine frequencies by phase.) Additionally, the molded
figurines and spindle whorls are very diverse. No duplicates from the same mold were
identified among the figurines. There is only a single case of two maguey whorls with
tops that appear to have come from the same mold, but their sides differ. The diversity of
198
finished artifacts suggests that there may have been a correspondingly high number of
intermittent producers, either at the site or elsewhere.
Due to recent changes in INAH policy, examples of spindle whorls and figurines
could not be exported for INAA as originally planned. As a result, I can only say that
figurines appear to include examples made from Basin of Mexico clays, locally produced
variants of Basin-style fine clays, and several groups of brown pastes that broadly overlap
those seen in vessels produced within the Toluca Valley. Figurines are discussed in more
detail in Chapter 7. Of the spindle whorls, some of the cotton whorls show the orange
shifted paste colors more characteristic of the Basin of Mexico, while the remainder are
consistent with local paste colors. On grounds of macropaste alone, none of the maguey
whorls are inconsistent with a production location within the Toluca Valley, though
stylistic traits suggest that at least some of them are likely imported from the Basin of
Mexico.
Lithic Production
Stone tools, especially those made from obsidian, make up the second most
common artifact class in Postclassic Mesoamerica households. Stone tool production was
often differently organized than other craft activities (Hirth 2006a), generally featuring a
higher degree of specialization than ceramic production. This is likely due to the more
limited spatial distribution of appropriate raw materials and relatively lower annual
consumption rates (Sanders and Santley 1983; Sheets 1992). The higher relative degree
of necessary training may also play a role. Despite this general trend toward more
199
specialized production, the organization of Mesoamerican lithic production was highly
variable over time and space (Cobean 2002: Chapter 5). On one end of the production
spectrum are workshops, such as those found at Epiclassic Xochicalco (Hirth 2008,
2009b), Early Postclassic Tula (Healan, et al. 1983) or Postclassic Otumba and the
surrounding Teotihuacan Valley (Parry 2001). Workshop-level production can usually be
identified even during survey, due to the very high concentrations of debitage that it
leaves behind. In addition, the lithic assemblage from a workshop is likely to show
relatively little evidence of use-wear (Hirth and Castanzo 2006). Based on reported data
from Middle Postclassic Xaltocan, obsidian to rim sherd ratios were approximately seven
times higher in a workshop dump context than in other contexts (1135 vs. 170 obsidian
fragments per 100 rims) (Millhauser 2005). Similarly, a deposit at Late Postclassic
Cihuatecpan interpreted as an obsidian workshop dump had an obsidian to total sherd
ratio approximately 18.5 times higher than the average of other excavated contexts (1868
vs. 100 obsidian pieces per 1000 total sherds) (Evans 1988). Below the workshop level,
lithic production can also take place as one of several household crafts in multicrafting
situations, such as at Classic period El Palmillo (Haines, et al. 2004), or Postclassic
Erongaricuro (Rebnegger 2010). In such cases, there is a reasonable amount of evidence
for lithic production, such as artifacts from several stages of a production sequence, but it
is generally lower in density than what is found in workshop contexts and may be mixed
with a higher proportion of domestic refuse. In addition, in multicrafting situations, lithic
production may be oriented toward producing tools for use in another craft, such as
scrapers for producing xerophytic plant products (Evans 1988) or gourd working (Hay
1978). At the most basic level, most Postclassic households show some evidence for the
200
occasional production or modification of tools for household use, and the mere presence
of production-related lithic forms should not be taken as evidence of specialization.
The degree of specialization in lithic production also varies by the type of artifact
being produced. Prismatic blade production tends to be a more specialized industry than
biface production, even at the same site. This is likely a product of the specialized skills
required to produce prismatic blades, as well as the fact that they require higher quality
raw material than bifacial artifacts (Clark 1987; Parry 1994).
Because lithic production is a reductive technology, the production of stone tools
leaves behind direct evidence in the form of the debitage produced by the knapping
process. In some cases, tools associated with lithic production, such as hammerstones,
can also be identified, but functional assignments often suffer from issues of equifinality.
As a result, the Calixtlahuaca Archaeological project did not separate lithic production
tools from other similar artifacts, and I focus only on the chipped stone material itself in
the following discussion.
Lithics found in archaeological contexts represent both the remains of the
production of stone tools, and tools discarded after use. Evidence of local production
must be distinguished from other potential causes for a concentration of lithic material,
such as another craft that required the use of large numbers of stone tools, or the
opportunistic exchange of production byproducts with novel but functionally valuable
forms, such as core rejuvenation artifacts or plunging blades.
201
Evidence for Lithic Production at Calixtlahuaca
Lithic production may be identified by very high concentrations of lithic material,
particularly production debitage. At Calixtlahuaca, Dr. Bradford Andrews classified
artifacts both by technology, and by the functional stage of lithic production or
consumption that they likely represented. The assignment of lithic artifact types to
“production” or “consumption” categories is a rough approximation, as many of the
artifact types created as production byproducts may be repurposed as cutting edges and
used. In order to be considered evidence of intensive, specialized lithic production, I
require that a household have both a high percentage of production-related lithics
(evidence of production), and a relatively high lithic-to-sherd ratio (evidence of
intensification relatively to other households activities). The full list of lithic artifact
codes, and their technologies and production/consumption codes can be found in
Appendix E. The discussion of lithic craft production included here draws heavily on
Andrews’ work, which will be available in more detail in the upcoming final project
report to INAH.
The stone tools used at Calixtlahuaca fall into three technological categories:
core-blade, bifacial, and bipolar (Andrews 2013). In the DS-1 sample, 66% of the lithic
artifacts are technologically diagnostic. Of these, 76% are core-blade, 19% are bifacial,
and 5% are bipolar. There was no evidence for lapidary production in the DS-1 sample.
However, of the lithic artifacts directly related to production, the proportions are
different, with the majority of the technologically identifiable production-related pieces
pertaining to bifacial production, followed by core-blade, and finally bipolar production
(Table 5.5). Part of the difference between the production-related and the total lithic
202
percentages is due to differences in the quantity of identifiable production-related pieces
each technology produces. However, the discrepancies here are high enough to suggest
differences in how various types of lithic artifacts were reaching consumers at
Calixtlahuaca.
For all three production technologies, the quantities of production-related artifacts
are low enough to raise questions of sample size, especially in the smaller components.
As a result, I require that household components fall more than two standard deviations
from the mean before being considered unusual, for both the frequency of percentage of
production related lithics for a given technology, and for the obsidian to sherd ratio. I also
generally interpret outlier households conservatively, with a focus on the broader,
sitewide pattern over time.
203
204
Table 5.5 Total and production related lithic frequencies for core-blade, biface and
bipolar lithic production artifacts by house and phase. DS-1 sample.
Core-Blade Production All Core/1000 Blade %
Lithics
N. % Lithics sherds
38.35
1.05
1.79
10
57.41
0.36
4.63
6
46.67
0.37
3.33
3
37.63
0.51
2.15
52
37.62
5.78
7.13
53
54.55
3.36
5.45
11
Biface Production
All
%
/1000 Biface %
N. Lithics sherds Lithics
14.70
7.99
38 13.62
7.41
0.44
5.56
6
26.67
2.61
7 23.33
11.83
2.53
10 10.75
18.74
13.59
122 17.06
9.09
5.61
9.09
5
Bipolar Production
All
/1000 Bipolar %
Lithics
N. % Lithics sherds
0.72
0.42
0.72
2
6.48
0.51
6.48
7
0.00
0.00
0.00
0
4.30
1.01
4.30
4
5.31
4.08
5.03
36
3.64
2.24
3.64
2
Unit
307
315
316
320
323
324
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
Total
Lithics
279
108
30
93
715
55
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
351
602
87
429
183
103
25
19
16
26
10
10
2.56
2.66
3.45
3.96
5.46
0.97
1.00
0.79
0.81
3.41
0.63
0.60
48.72
54.82
48.28
47.32
57.92
74.76
43
60
6
79
21
0
12.25
9.97
6.90
18.41
11.48
0.00
4.78
2.96
1.34
15.24
1.20
0.00
13.96
11.63
6.90
19.58
14.21
0.97
5
0
9
0
1
7
1.42
0.00
10.34
0.00
0.55
6.80
0.56
0.00
2.42
0.00
0.06
4.20
1.42
0.00
10.34
0.23
0.55
6.80
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
289
71
90
77
175
18
15
5
2
10
11
3
3.81
5.63
1.11
2.60
4.57
16.67
1.08
1.29
0.13
0.21
2.32
3.17
62.28
66.20
56.67
74.03
51.43
88.89
21
3
9
2
3
0
7.27
4.23
10.00
2.60
1.71
0.00
2.05
0.97
1.20
0.21
0.87
0.00
9.69
7.04
12.22
2.60
2.29
0.00
1
0
4
4
33
0
0.35
0.00
4.44
5.19
18.86
0.00
0.10
0.00
0.53
0.42
9.55
0.00
0.35
0.00
4.44
5.19
18.86
0.00
Dongu, Mean
Ninupi, Mean
Yata, Mean
All Phases, Mean
4.08
3.18
5.73
4.33
1.91
1.21
1.37
1.49
45.37
55.30
66.58
55.75
13.24
9.83
4.30
9.12
5.46
4.26
0.88
3.53
14.74
11.21
5.64
10.53
3.36
3.19
4.81
3.78
1.38
1.21
1.77
1.45
3.41
3.22
4.81
3.81
Dongu, Std Dev
Ninupi, Std Dev
Yata, Std Dev
All Phases, Std Dev
2.05
1.51
5.58
3.50
2.22
1.09
1.19
1.52
8.94
10.41
13.41
13.70
6.31
6.12
3.72
6.42
4.79
5.63
0.74
4.49
7.11
6.48
4.77
6.98
2.52
4.35
7.26
4.85
1.53
1.73
3.82
2.44
2.56
4.32
7.26
4.84
Core-Blade Production. Potential evidence for production of core-blade artifacts
consists of artifacts relating to core shaping, rejuvenation and recycling, as well as initial
series pressure blades and flakes or blades resulting from production errors. Most
artifacts in this category relate to the production of prismatic blades. One or more of these
artifact classes are present in all of the excavated household components, but the
frequencies are generally extremely low. With two exceptions, artifacts related to coreblade production account for 1-6% of the total lithics in a household component, and at a
relative frequency of less than three and a half pieces per 1000 ceramic sherds. The first
exception to this pattern is household component 327-Ph6, which is a very small sample
(n=18 obsidian artifacts), of which the identifiable artifacts consist exclusively of core
shaping artifacts (n=3) and prismatic blades. The relatively low obsidian to sherd ratio in
this component, however, suggests that the lack of artifact type diversity is due to sample
size rather than actual craft production. The second exception to the general pattern is
323-Ph2. In this component, artifacts related to core-blade production account for both an
unusually high percentage of the lithics and the highest frequency relative to the ceramic
count of any component. This component includes artifacts from a wide range of stages
of production, with all categories noted above except core recycling present. However,
while this household component does have higher frequencies than any other household,
they are not outside of the two-standard deviation range for all components at the site, nor
are they the order of magnitude higher characteristic of specialized lithic production.
The general pattern of core-blade artifact provisioning is one in which the
overwhelming majority of prismatic blades entered circulation in the city as completed
artifacts. This is supported by the much higher frequency of consumed blades relative to
205
the frequency of artifacts related to blade production. Given the technical skill necessary
for prismatic blade production, the very low but widespread presence of production
related artifacts is more consistent with their sale and secondary use as cutting edges,
rather than widespread household blade production. The evidence for blade production is
reasonably constant over time. If the outlier households are excluded, evidence for coreblade production accounts for an average of 3-3.5% of the total lithic assemblages over
time. Similarly, it accounts for 1-1.5 obsidian artifacts per 1000 sherds in all three
periods. This suggests little change in the organization or intensity of prismatic blade
production (or, rather, the absence thereof) at Calixtlahuaca over time.
Biface Production. Potential evidence for bifacial artifact production includes
debitage from the production of bifacial, unifacial, and flake-core artifacts, and at
Calixtlahuaca primarily includes material in the former two categories. Finished tools are
not included as evidence of production. Bifacial artifact production requires less technical
skill than core-blade production, and produces less useful byproducts. As a result,
relatively lower total quantities of production related artifacts are required to argue for
bifacial production than for blade production. At Calixtlahuaca, the evidence for bifacial
artifact production shows a higher level of intercomponent variability than the evidence
for prismatic blade production, ranging from 0-23% of the total lithic assemblage and
frequencies of 0-15 artifacts per 1000 sherds. However, no household component falls
outside of the two standard deviation range for both the percentage of biface production
artifacts and their frequency per 1000 sherds. (Interestingly, the two households which
completely lack evidence for bifacial production – 322-Ph4 and 327-Ph6 - are the only
206
two components located on the smaller hill to the east of the main site. This may be an
area with lower than average biface production, but the sample sizes are too small to
evaluate this conclusively.) As a result, I do not consider the excavated sample of
households to have included any cases of specialized biface production.
Based on this patterning, the production of bifacial artifacts likely occurred in
most households (except those on the east edge of the site) during all time periods. Some
households produced more than others, but none reach a level of production that could be
considered a workshop. Bifacial tool production decreases consistently over time, both as
a percentage of the total lithic assemblage and as a relative frequency per 1000 sherds.
While the exact magnitude of the drop depends on whether cases falling between one and
two standard deviations are included, the inclusion or omission of these components does
not change the overall pattern.
Bipolar Production. The third major category of lithic artifacts found at
Calixtlahuaca was produced using bipolar techniques. These are produced by supporting
the piece of obsidian on an anvil while striking it, allowing force to be transmitted from
both ends (Flenniken 1981). It is an efficient way to use small fragments of stone that
might otherwise not be knappable, or a means of recycling tools that had lost their edge,
particularly blade sections. (This reuse of blade sections at Calixtlahuaca is unusual for
Mesoamerica.) Because this technology represents a final effort to get a usable edge from
a small fragment of stone, the act of bipolar production presumably took place within the
site. Evidence for bipolar production includes almost all artifacts produced using the
technique, and as a result, this category may include a higher degree of equifinality
207
between evidence of production and evidence of consumption. The average evidence for
bipolar production is similar to that for core-blade production, accounting for a phase
average of 3-5% of the total lithics. However, the evidence for bipolar production is
much more variable among households, ranging from 0-18% of the total lithics, and 0-9
pieces per 1000 sherds. There are five components with no evidence for bipolar
production, spread across all three phases, meaning that this was not an activity that all
households participated in. There is also one component, 324-Ph6, which shows an
unusually high level of bipolar production, with both the percentage of bipolar artifacts
and their frequency per 1000 sherds falling outside of the two standard deviation range,
both as a percentage of the lithic assemblage and as a frequency per 1000 sherds.
Bipolar production was most likely an occasional activity performed by some
households at Calixtlahuaca, primarily to recycle/produce tools for their own use. The
much patchier distribution of evidence for bipolar production (which is almost
coterminous with any evidence for the production or consumption of bipolar artifacts),
may be partially due to sample size and the relative rarity of bipolar artifacts overall.
However, it does suggest that bipolar artifacts were produced within the household on an
as-needed basis, rather than bought and sold in the marketplace. 324-Ph6, with its more
intensive bipolar production, may have been a poorer household with more need to
conserve obsidian. The overall level of bipolar production is approximately stable during
the Dongu and Ninupi phases, followed by an increase during the Yata phase. Given that
the absolute quantity of obsidian entering the site drops during this final period, an
increase in bipolar production, which “recycles” previously used obsidian, is expected.
However, at an individual household level, there is little correlation between the quantity
208
of bipolar production and the obsidian/sherd ratio. As noted in the previous chapter
(Table 4.13), the overall obsidian to sherd ratios at Calixtlahuaca are low relative to those
seen in the Basin of Mexico, but comparable to those in Morelos or Oaxaca.
In general, a low degree of lithic production occurred at Calixtlahuaca, but the
details of production vary by technology. Prismatic blades were widely utilized, but there
is next to no evidence for the preparation of cores or the abandonment of exhausted cores.
This suggests that most blades were produced by itinerant specialists, who arrived,
produced blades, and then moved on, taking their partially used cores with them, rather
than local individuals. In contrast, most households show some evidence for biface
production, and a few cases may have been producing for use in other crafts. Similarly,
many, but not all, households show evidence of occasional bipolar production, with one
possible case of more intensive production. Evidence for blade production remains fairly
constant over time, while biface production drops, and bipolar production slightly
increases.
Cloth Production
The evidence for the production of textiles at Aztec sites often patterns differently
than that for other crafts at the same sites. This is most likely due to the use of textiles as
a form of currency, for both tax payments and independent trading purposes. As a result,
the Aztec Empire could conceivably promote local textile production even as it
suppressed the production of other crafts in order to create more demand for products
from the imperial core. With the technology – drop-spun thread, and backstrap looms 209
available in Postclassic Mesoamerica, textile production is not particularly amenable to
economies of scale. Female members of a household could invest more time in textile
production, or engage more members of the household in various stages of fiber
processing, but the net output per work hour would probably not have gone up
substantially (Hicks 1994). (The same problem confronted textile producers in England
immediately prior to the industrial revolution, leading to the “putting-out” system (e.g.
Millward 1981)).
Mesoamerican textile production involved a series of stages and a variety of
fibers. The stages of production include fiber extraction, spinning, weaving, sewing,
dyeing, and embroidering. The two primary fibers were cotton and maguey (agave), with
an additional variety of less well documented fibers such as rabbit fur, palm, and metallic
threads (McCafferty and McCafferty 2000). Cotton and maguey have different
technological constraints in their production, resulting in different types of associated
production artifacts. The archaeological evidence for maguey fiber production consists of
tabular “desfibrador” scrapers, usually made of basalt, and larger spindle whorls (>10g)
(Parsons and Parsons 1990). The evidence for cotton fiber production consists of smaller
spindle whorls (<10g) and the small bowls used to support them (Parsons 1972; Smith
and Hirth 1988). The weight division between cotton and maguey whorls shows some
regional variation, but Parson’s original values remain valid for central highland Mexico.
Additional tools, such as copper or bone needles, bone awls, bone or greenstone weaving
battens, and stamps for decorating cloth are rare enough that they cannot usually be
evaluated in a systematic manner.
210
The easiest way to compare the intensity of textile production among sites is to
standardize cloth production artifacts by some factor. Due to the variability of
archaeological reporting, different researchers have used excavated volume, surface area,
rim sherd count, and total sherd count (King 2011). I use spindle whorls per 1000 sherds,
as these are the most commonly reported (Table 5.6). I also include sites reporting whorls
and rim sherds, applying a conversion factor of ten total sherds per rim unless an estimate
of the rim sherd/total sherd ratio was included. Based on the forty-one Postclassic and
Early Colonial site components for which I could find data (Table 5.6), the frequency of
spindle whorls relative to total sherd count forms a strongly right-tailed distribution
(Figure 5.2). About two thirds of the sites fall between a complete absence of whorls and
about 1.5 whorls per 1000 sherds. Then there are a smaller number of sites with
frequencies up to just over four whorls per 1000 sherds. I interpret sites in the first range
as those producing primarily for domestic use, with a possible low degree of
specialization for market sale toward the higher end of the range. I interpret sites in the
higher cluster as those specializing in cotton production. All but one of them are located
in known cotton cultivating areas, and the exception (the Late Postclassic portion of
O’Neill’s excavations at Chalco) is quite close to cotton producing regions and features
exclusively cotton whorls. Other than Calixtlahuaca itself, none of the sites has more than
about 60% maguey, as opposed to cotton whorls, making it difficult to assess what a
specialization in maguey fiber would look like. The best candidate is Cihuatecpan, which
both falls at the upper end of the proposed “non-specialized” range, and is one of the few
sites with more maguey than cotton whorls.
211
Site
Toluca Valley
Calixtlahuaca
Per 1000
Source
Total
Sherds Cotton Maguey Total Total
Period (Rim=10 Whorls Whorls Whorls Rims Sherds
MPC
LPC-A
LPC-B
Huamango
EPI-EPC
Cerro Toloche LPC
Morelos
Capilco
MPC
LPC-A
LPC-B
Cuexcomate
LPC-A
LPC-B
Xochicalco & MPC
LPC-A
Coatetelco
LPC-B
Yautepec
MPC
LPC-A
LPC-B
COL
Basin of Mexico
Chalco
MPC
LPC
Chalco
MPC
LPC
Tlatelolco
LPC
Cihuatecpan LPC
Huexotla*
MPC
LPC
Xaltocan
EPC
(Testing)
MPC
LPC
Xaltocan
EPC
(Houses)
MPC
LPC
Xaltocan*
MPC
LPC
Xico*
MPC
LPC
Coatlan Viejo* LPC
0.16
0.34
0.87
1.20
0.48
2
5
13
1.43
1.90
2.99
1.53
2.35
2.08
2.63
3.15
0.97
1.17
1.16
1.47
2
42
86
31
210
2
7
5
29
143
227
13
3
2
2
0
25
3
2
0
1
9
10
0
0.48
1.05
1.20
4.10
0.73
1.21
0.59
0.29
0.31
0.20
0.14
0.42
0.46
0.81
0.35
0.91
0.12
0.04
2.26
8
20
6
13
3
2
2
0
53
10
37
2
6
7
4
7
10
6
17
5
1
119
65
11
44
1
7
2
17
6
2
7
4
1
0
13
9
18
22
3
11 6,382 67,345
23 7,317 67,988
35 4,585 40,129
20
16,689 Pina Chan 1981, p. 92,118
3
6,204 Jaramillo & De la Peña
2012, Table 2
5
3,496 Smith 2006 Table B3-2
44
23,115
88
29,444
31
20,286 Smith 2006 Table B3-2
235
99,828
5
2,402 Smith & Hirth 1988, Table
9
3,417 2
5
1,589
30
30,795 Smith 2006, Table C3-2
152
130,206
237
204,243
13
8,866
11
22
8
13
9
118
21
81
3
13
9
21
13
12
13
21
6
1
122
Hodge 2008, Chap 9,13;
Table B-4
O'Neill 1962, cited in
Smith & Hirth 1988
12,284 Gonzalez Rul 1988a, p.186
97,252 Evans 1988
Brumfiel 1991, Table 8.1
3,582
27,720
Brumfiel 2005, Tbls 4.2-3,
965
14.1-2, Phases per
6,661
Overholtzer 2014
6,418
DeLucia and Overholtzer
2014, Tables 1, 2
2,288
2,101
3,740
2,298
5,062
2,247
5,408
Brumfiel 1991, Table 8.1
Brumfiel 1991, Table
8.1(citing Mason 1980)
Table 5.6 Cotton and maguey spindle whorl frequencies at Calixtlahuaca and other
comparative sites
212
Per 1000
Source
Sherds Cotton Maguey Total Total
Total
Period (Rim=10 Whorls Whorls Whorls Rims Sherds
Site
Oaxaca
Nicayuju
PC
Rio Viejo
EPC
Tutuepec, ResA MPC
Tutuepec, ResB LPC
Tutuepec, ResC LPC
Other
Tepetitlan
EPC
Cholula, UA-1 PC
0.00
0.31
1.30
2.34
2.72
0
86
79
39
16
0
0
0
1
0
0.23
0.79
0
46,719 Perez Rodriguez 2006
86 36,677 276,300 King 2011, Table 2
79 6,297 60,816 Levine 2007, Table 7.01
40 1,740 17,090
16
639 5,883
Cobean & Alva Mastache
13
55,730 1999, Figures 2.1, 8.39a
129 16,396
McCafferty 2001, McCafferty & McCafferty, 2000
*Survey Project
Table 5.6 (continued) Cotton and Maguey spindle whorl frequencies at Calixtlahuaca
and other comparative sites
Figure 5.2 Histogram of spindle whorl frequencies per 1000 sherds at Postclassic and
Early Colonial sites
213
Cotton Textile Production
The return per land unit on cotton cultivation may have been higher than that for
food production, making it more economically logical to specialize in cotton cultivation
rather than the intensified production of food crops, in regions where cotton cultivation
was possible. However, after cotton is picked, the remaining highly labor intensive
cleaning, carding, spinning, and weaving may take place elsewhere. Cotton could not be
grown in the Toluca Valley for climatic reasons, and would have had to be imported from
warmer regions, such as Morelos or the Balsas drainage. Thus, the presence of cotton
spinning artifacts is only evidence for the textile production stages from spinning onward.
The float samples from Calixtlahuaca recovered a small number (n=14) of plant
fragments from Family Malavaceae, but none of those that could be identified to genus
level (n=3) were Gossypium (cotton). This lack of identifiable cotton seeds in the
macrobotanical samples from Calixtlahuaca suggests that cotton was imported precleaned. In addition, many of the tools used to spin cotton at the site may also have been
imported. As noted in the preceding section of this chapter on figurine and whorl
production, some of the cotton whorls show orange-toned pastes more consistent with
manufacture in the Basin of Mexico, and very few have definitively local pastes.
Spinning bowls are also disproportionately imported; approximately 2/3 of the examples
in the DS-2 sample were coded as imported Black-on-Orange, and all but one of the
remainder were identified as locally produced imitations of the same. Given that spinning
bowls imported during the Dongu phase predate the widespread importation of Aztec
Black-on-Orange serving vessels at the site, it is likely that they were imported as part of
214
a new craft technology, rather than for stylistic appeal. A sample of cotton spindle whorls
from Calixtlahuaca can be seen in Figure 5.3.
Figure 5.3 Decorated (A-D) and plain (E-I) cotton spindle whorls from Calixtlahuaca,
showing the range of variation in shapes.
Evidence for cotton production at Calixtlahuaca increases steadily over time, from
.13 cotton spinning artifacts (small whorls or spinning bowl fragments) per 1000 sherds
during the Dongu phase, to approximately .66 cotton spinning artifacts per 1000 sherds
by the Yata phase (Table 5.7). The increasing frequency is driven primarily by a rising
frequency of spinning tools within particular households, rather than a rise in the ubiquity
of spinning among households. There are three Dongu phase households with cotton
spinning tools, as compared to four households in each of the subsequent phases,
215
indicating little change in the ubiquity of cotton production. Despite this, only one
household component, 307-Ph2, is an extreme outlier when compared to others within its
phase. The artifact assemblage for this household is unusual in a number of ways, which
are discussed more extensively in the next chapter. While the internal evidence for an
increase in cotton production is strong, comparisons with spinning tool frequencies at
other sites show that even at its highest, cotton textile production at Calixtlahuaca
remained on the low end for Postclassic Mesoamerica.
Total Cotton Maguey Spin. Maguey Freq C Freq M
Freq
Sherds Whorls Whorls Bowls Scrapers Whorls Whorls SpnBwl
5,810
1
1
3
0.17
0.17
0.52
16,775
3
2
0.00
0.18
0.00
4,710
0.00
0.00
0.00
12,189
2
1
2
0.00
0.16
0.08
26,947
1
3
1
0.04
0.11
0.00
914
0.00
0.00
0.00
Unit
307
315
316
320
323
324
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
303
307
308
311
316
322
Ninupi 9,043
Ninupi 22,330
Ninupi 4,359
Ninupi 7,838
Ninupi 22,563
Ninupi 1,855
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
10,257
4,217
10,091
10,860
3,438
1,266
2
2
1
9
2
2
1
4
5
8
6
2
6
4
3
1
2
1
2
6
9
3
3
6
4
1
1
1
Dongu
Ninupi
Yata
Freq Cotton Maguey
MS Tools Tools
0.00 0.69
0.17
0.12 0.00
0.30
0.00 0.00
0.00
0.16 0.08
0.33
0.04 0.04
0.15
0.00 0.00
0.00
0.00
0.09
0.46
0.13
0.00
0.00
0.11
0.18
0.00
0.64
0.35
0.00
0.00
0.00
0.23
0.26
0.27
0.00
0.00
0.09
0.00
0.00
0.00
0.00
0.00
0.09
0.69
0.38
0.27
0.00
0.11
0.27
0.00
0.64
0.35
0.00
0.88
0.47
0.00
0.18
0.00
0.00
0.58
0.47
0.59
0.37
0.87
0.79
0.88
0.71
0.30
0.55
0.00
0.00
0.39
0.24
0.00
0.09
0.00
0.79
1.75
1.19
0.30
0.74
0.00
0.00
0.97
0.71
0.59
0.46
0.87
1.58
0.03
0.11
0.26
0.10
0.21
0.61
0.10 0.05
0.13 0.01
0.41 0.25
0.13
0.24
0.66
0.16
0.23
0.87
Table 5.7 Cotton (cotton-weight whorls and spinning bowls) and maguey (maguey-weight
whorls and tabular basalt scrapers) textile production tool frequencies per 1000 total
sherds by household component
216
Maguey Textile Production
In contrast to cotton, maguey is climactically suited to the Toluca Valley and was
widely cultivated there in antiquity. In addition to the fiber-related tools discussed below,
the daub wall fragments excavated at Calixtlahuaca show frequent impressions of split
maguey stalks (Karabowicz 2009). Maguey plants are commonly mentioned in ColonialPeriod Nahua language wills from the surrounding area (García Castro 2000; Pizzigoni
2013), and even today much of the hill where the site is located is planted in magueys to
support terraces and prevent erosion. Due to the wide range of products that can be made
from maguey, including foods, medicines, building materials, and textiles, not all the
maguey cultivated at the site was probably used for textile or cordage production, but
there would have been no shortage of raw plant material, either.
As might be expected, evidence for maguey textile production at Calixtlahuaca is
higher than that for cotton production, with 2-3 times more maguey whorls than cotton
whorls (Table 5.7). Maguey textile production also increased over time, showing
approximately the same degree of increase as cotton production. This increase is
particularly dramatic by the Yata phase, for which the maguey processing artifact
frequencies do not overlap either of the preceding periods at one standard deviation.
There are no extreme outlier households during any phase.
The increase in maguey processing tools occurs for both basalt scrapers and large
spindle whorls (Figure 5.4, Figure 5.5). Ethnographically, the scraping stage of maguey
fiber processing often occurs near a water source, which is not necessarily near the house
(Camposeco M. 1994; Mendoza Cerón and Canger 1993; Parsons and Parsons 1990). As
a result, this household-focused sample likely under represents this artifact class. Large
217
spindle whorls, which represent the thread-spinning phase of production, increase about
sixfold, to about .61 whorls per 1000 sherds by the Yata phase. When compared to
frequencies at other sites (Table 5.6), this suggests a move from below-expected levels of
maguey textile production to slightly above-expected. However, even at its highest, the
frequency of maguey whorls at Calixtlahuaca is only half as high as that observed at
Cihuatecpan (Evans 1988), a Basin of Mexico site which has been proposed to have
specialized in maguey products.
Figure 5.4 Examples of maguey spindle whorls from Calixtlahuaca. A-D, F and G are
stylistically typical of the Toluca Valley. E stylistically matches the Basin of Mexico.
218
Figure 5.5 Tabular basalt maguey scrapers (“desfibradores”) from Calixtlahuaca
Discussion
Calixtlahuaca shows relatively little evidence for specialized craft production
compared to most contemporaneous sites in Central Mexico. With the exception of biface
production, there is little variation among households in the number of crafts practiced or
the intensity of craft production, when the size of the excavation is taken into account
(Figure 5.6,
Table 5.8). Excavation size and the number of crafts represented are correlated at .52.
The three smallest excavations account for three out of four lowest numbers of crafts
represented. When cases with unexpectedly high levels of evidence for particular crafts
are considered, they are, with one exception, all in the smaller half of the components.
The pattern of smaller components having evidence for fewer crafts, but higher
219
frequencies for the ones that do appear suggests that both patterns are a product of sample
size. There is one possible exception to this general pattern of low frequency, widely
distributed production. Component 323-Ph2 is the largest excavation, but also has higher
than average values for both core-blade and bifacial artifact production.
Figure 5.6 Relationship between component size (sherd count) and the number of craft
activities present. Each point is a household component.
The total average household evidence for craft production decreases steadily over
time (
Table 5.8). However, this is a very rough proxy measure for total craft production, as
different crafts leave very differing amounts of archaeological evidence at the same level
of specialization. The observed decrease is also heavily driven by the two components
with higher evidence for biface production, and averages calculated without these
components or excluding all biface production produce alternate patterns. There is also a
220
high degree of variability in the patterning of individual crafts over time (Figure 5.7).
Local ceramic vessel production shows little variation across time. Textiles – both cotton
and maguey – increase from one period to the next. In contrast, the three lithic production
technologies – core-blade, bifacial, and bipolar lithic production – produce three different
patterns of relative rankings across the three phases. This variability is likely a matter of
changes in the relative importance of the three production technologies, though the net
effect is of decreasing evidence for lithic production over time.
221
Figure 5.7 Temporal trends in craft production at Calixtlahuaca, by craft. (A) Lithics (B)
Textiles
222
Table 5.8 Summary of craft production artifact frequencies by household component.
Dark grey cells are two standard deviations outside mean for all components. Light grey
cells are borderline anomalous.
Possible Alternatives
223
Unit
307
315
316
320
323
324
303
307
308
311
316
322
307
309
316
317
324
327
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Yata
Yata
Yata
Yata
Yata
Yata
Dongu
Ninupi
Yata
Ceramics
Total
Sherds Molds
5,810
16,775 Y
4,710
12,189
26,947
914
9,043 Y
22,330
4,359
7,838 Y
22,563
1,855
10,257 Y
4,217
10,091
10,860 Y
3,438
1,266
Lithics
Textiles
CoreBlade Bifacial Bipolar
0.42
7.99
1.05
0.51
0.44
0.36
0.00
2.61
0.37
1.01
2.53
0.51
4.08
5.78 13.59
2.24
5.61
3.36
0.56
4.78
1.00
0.00
2.96
0.79
2.42
1.34
0.81
0.00
3.41 15.24
0.06
1.20
0.63
4.20
0.00
0.60
0.10
2.05
1.08
0.00
0.97
1.29
0.53
1.20
0.13
0.42
0.21
0.21
9.55
0.87
2.32
0.00
0.00
3.17
1.91
1.21
1.37
5.46
4.26
0.88
1.38
1.21
1.77
Cotton Maguey
0.17
0.69
0.30
0.00
0.00
0.00
0.33
0.08
0.15
0.04
0.00
0.00
0.11
0.00
0.27
0.09
0.00
0.69
0.64
0.38
0.35
0.27
0.00
0.00
0.97
1.75
0.71
1.19
0.59
0.30
0.46
0.74
0.87
0.00
1.58
0.00
0.13
0.24
0.66
0.16
0.23
0.87
Total
Total
Craft
Crafts "Highs"
Present
Evidence
10.32
0.5
5
1.61
0
5
2.98
0
2
4.46
0
5
23.64
2
5
11.21
0
3
6.45
0
5
4.11
0
4
5.25
0
4
19.67
1.5
5
2.52
0
5
4.80
0
2
5.96
1
6
4.16
0
4
2.76
0
5
2.03
0
6
13.61
1
4
4.75
1.5
2
4.17
4.17
4.50
0.42
0.25
0.58
9.04
7.13
5.55
Prior to considering the broader implications of craft specialization at
Calixtlahuaca, I touch briefly on several ways in which the preceding patterns might be
misleading relative to the actual pattern of activity specialization at the site. I do not
consider any of these alternatives particularly probable, but good scientific data
evaluation requires the consideration of potentially confounding factors (Smith 2015c).
A first hypothesis is that the occupants of Calixtlahuaca specialized in some
activity that does not preserve archaeologically. The Toluca Valley is known for the
production of marsh-related plant and animal products both prehistorically (Sugiura
Yamamoto 2009) and historically (García Sánchez 2008), a form of specialization which
would not be directly apparent in the lines of evidence discussed in this chapter.
However, a specialization in marsh-related products would be at least indirectly visible as
net weights, blowgun pellets, patterned faunal remains, salt vessels for preserving, or
ground stone for cane-working. As none of these artifact classes occur as more than
occasional background pieces, an unrecognized specialization in marsh products is
unlikely. In addition, Calixtlahuaca is not located particularly close to the marshy
portions of the Upper Lerma drainage.
A second hypothesis is that worked sherd disks (tejos) are the byproduct of a
sitewide specialization in an unknown craft. Sherd disks occur in all household contexts
during all phases at Calixtlahuaca, in far higher frequencies than at contemporaneous
sites in the Basin of Mexico or Morelos (Warren 2015). Regionally specific artifact types
– such as bark beaters for making amate paper in Morelos – are good candidates for
markers of regional craft activities. However, the patterning of disks at Calixtlahuaca
does not correspond to any of the four currently proposed uses for sherd disks: spinning
224
weights, net weights, gaming pieces (Phillips 2002), or exchange tokens (Manzanilla
2011). Only a handful of the approximately 3000 disks recovered at Calixtlahuaca have
the central perforations or edge notching required for use as spinning or net weights. The
disks are found in both household and non-household contexts, which is not particularly
consistent with use as gaming pieces. They also lack the clear division into standardized
fractions (halves, quarters) that Manzanilla argues may indicate use as a standardized
medium of exchange. With the majority of pieces under 5cm in diameter, the disks are
too small to have been used as plates or lids. As a result, until a reasonable use can be
proposed, sherd disks remain unlikely evidence of craft specialization.
A third hypothesis is that the occupants of Calixtlahuaca were agricultural
specialists. The extensive terraces at the site were a substantial investment in
agriculturally oriented infrastructure. This investment occurred despite the presence of a
wide surrounding alluvial plain, where maize is widely cultivated today. Based on the
number of test pits and terrace trenches excavated by the CAP that did not encounter
house remains, there was substantial open space within the city. In addition, the terrace
soils had been heavily modified to improve drainage and soil quality (Smith, et al. 2013).
This combination of factors suggests the widespread presence of kitchen gardens, if not
full-fledged milpas, within the site. Because farming leaves little evidence, it is difficult
to quantify directly. However, the intensity of agricultural production would have been
fairly consistent over time, as no major reduction in most other craft activities occurred.
(Previous research at the urban site of Huexota showed that decreased craft production
did accompany increased agricultural specialization at that site (Brumfiel 1980)). At a
225
minimum, this suggests that Aztec rule did not result in a dramatic increase in
agricultural activity at Calixtlahuaca.
Conclusions
Calixtlahuaca is, in many ways, an example of a city made up of stereotypical
autonomous peasant households. Craft production at Calixtlahuaca was generally low
intensity, with only occasional production for use above the household level. Evidence of
production is also widespread; most crafts for which there is any production evidence
occur in most excavated household components (
Table 5.8). This generalized pattern parallels the evidence produced by the sitewide
survey, which did not locate workshops or other zones of intensive production (Smith
2006b).
The lack of specialization at Calixtlahuaca stands in direct opposition to Fargher’s
(2009) model’s predictions of cities as hubs of craft specialization, and contrasts with
most contemporaneous sites in Central Mexico. This suggests two things. First, prior to
the Aztec conquest of the Toluca Valley, either the internal population levels and/or the
degree of local market integration was too low to produce a strong internally-driven
specialization of labor in the region. While precise population estimates are not available
for the Postclassic Toluca Valley, Sugiura’s survey of the Toluca Valley (Sugiura
Yamamoto 2011) shows a notably lower Postclassic site density than the Basin of
Mexico surveys (Sanders, et al. 1979). When paired with the evidence for increasing
market development seen in the previous chapter, the lack of craft specialization at
226
Calixtlahuaca suggests that markets were present and increasing in reliability over the
Middle and Late Postclassic-A periods, but that they had not reached the level of intraregional reliability seen in the Basin of Mexico. The lack of market development does not
provide strong evidence for either a network or corporate orientation among the preAztec rulers of the Toluca Valley. In the crafts considered in this chapter, there are
neither the independent specialists that would be expected under strongly collective,
market-dependent social organization, nor the attached specialists that might be expected
under less collective regimes.
Second, the continued lack of specialization in the Toluca Valley under Aztec rule
suggests that it falls outside of the area directly affected by the consumption demands in
the Basin of Mexico, either due to geographic distance or cultural boundaries. If
Calixtlahuaca had fallen within this zone throughout the periods under consideration, the
degree and organization of craft production should have paralleled the pattern of
increasing sub-regional specialization seen at contemporaneous sites in the Basin.
Alternatively, if Calixtlahuaca had become fully integrated into the Basin market system
after its conquest, the Yata phase should show a marked increase in the intensity of
production of one or more crafts. The continued lack of specialization at the site during
the Yata phase supports relatively indirect rule by the Aztec Empire. The changes in craft
production which do occur are relatively minor (a small increase in textile production,
and a small decrease in lithic production, likely driven by material scarcity). Aztec
economic exploitation was neither pronounced enough to produce regional specialization
in reaction, nor Aztec rule penetrating enough to enforce such specialization by fiat.
227
CHAPTER 6
CHAPTER 6. “MECÍA TEICUH BOUGHT AN OLD JÍCARA FOR 15 CACAO
BEANS6”: MEASURING WEALTH
One of the most central questions in discussions of imperialism, both Aztec and
otherwise, is the effect that incorporation into a state or empire had on the quality of life
of conquered peoples. On the one hand, becoming part of a larger state offered greater
opportunities for trade, geographic mobility, and social connections with distant areas.
On the other hand, foreign rule was often accompanied by increased taxation, imported
goods that undercut local markets, and the imposition of foreign customs. As the debates
in each of the previous two chapters demonstrate, the same change in trade or craft
production can be interpreted in both positive and negative lights. While wealth and
quality of life are not synonymous (discussed in more detail below), wealth is the most
easily measured dimension of quality of life in many archaeological contexts. This
chapter provides a context for the relatively small changes in long distance exchange and
local craft production observed at Calixtlahuaca. It measures the degree to which the
overall level and distribution of household wealth changed over time, the implications
that this had for quality of life, and how this relates to larger questions of rulership
strategies.
At Calixtlahuaca, the Dongu to Ninupi periods represent a time during which the
site was politically independent. As a result, changes in household wealth between these
6
Cline, S. L. and Miguel León-Portilla
1993
The Testaments of Culhuacan (modified first edition). UCLA Latin American Center Nahuatl
Studies Series, No. 1: Document 13/f. 35r
228
two phases were most likely the result of greater integration into the network of markets
in Postclassic Central Mexico (Berdan 2003). This integration is visible as the
widespread but modest increase in non-local goods at Calixtlahuaca between these two
periods. While it is unclear what goods Calixtlahuaca may have provided in exchange,
agricultural products remain a strong possibility. In contrast, Ninupi to Yata phase
changes are likely due to a combination of market integration and political factors. As a
result, the contrast between the two patterns can be used to separate the results of Aztec
imperial control from broader regional changes in economic activity. The differences
among time periods can also be used to differentiate the effects of local and Aztec
imperial rulership strategies.
Under relatively collective forms of rulership, commoners have a greater ability to
negotiate for the provision of public services and these services are available to the
majority of the population (Blanton and Fargher 2008:133-136). Relatively collective
rulership is also likely to promote economic growth (Blanton and Fargher 2008:280-282).
Due to these two factors, this form of rulership should promote an increasing standard of
living for the majority of the population. In contrast, relatively non-collective rulership
should provide a more limited range of public goods and fewer opportunities for
economic growth through market participation. Changes in the standard of living should
also be more unevenly distributed, since opportunities are distributed on a patronage
basis. At Calixtlahuaca, I evaluate the shift in collectivity of rule between the periods
under local and Aztec rule, based on changes in the average household wealth per phase
and in the interhousehold variation in wealth.
229
The question of the directness of rule applies primarily to the rulership strategies
of the Aztec Empire. More direct rule will produce greater changes in provincial areas,
but the direction of change will primarily be a result of the collectiveness of imperial
rulership. As a result, I expect indirect rule to produce few changes in the standard of
living at the site, while more direct rule has the potential to produce more significant
changes, in either positive or negative directions. I evaluate the directness of rule using
the differences in the absolute wealth levels at the site among phases, as well as Yata
phase shifts away from the prior trajectory of change.
My analyses in this chapter show that at Calixtlahuaca, most average measures of
wealth increase dramatically between the Dongu and Ninupi phases. This is followed by
approximately even wealth levels between the Ninupi and Yata phases, though these
results are sharply divided between slightly decreasing non-local items (obsidian,
bronze), and neutral to increasing primarily local items (ceramics, ground stone). At the
same time, the interhousehold variation in wealth values increases from each phase to the
next. Based on the combination of these two dimensions of household wealth, I argue that
the pre-Aztec rulership of Calixtlahuaca was relatively collective, but that Aztec rule was
both relatively indirect and non-collective.
Quality of Life, Wealth, and Status
Quality of life, wealth, and status are three related, but non-identical concepts
(Smith 2015d). Quality of life/wealth and wealth/status are often conflated, particularly in
archaeological discussions. Quality of life is a broad measure of an individual’s material,
230
physical and psychological wellbeing. Wealth, in contrast, measures the resources
(including both material and non-material resources) controlled by an individual or larger
unit. Status is a measure of social prestige or membership in ranked social groups. The
latter two concepts have a relatively long history of archaeological investigation, while
the former is just beginning to be considered (For exceptions, see many of the pieces in
the November 2013 issue of the SAA Archaeological Record).
Quality of Life
Quality of life is a difficult concept to quantify. For modern populations, most
formulations are based on Sen (1993), who proposes that quality of life be seen as a
combination of economic and social factors. An individual’s quality of life is based on
the portion of the culturally relevant possible “bundles” they are able to acquire. This
perspective has not been widely considered archaeologically and archaeological
correlates for Sen’s factors are generally undeveloped (However, see Smith 2015d for a
recent exception).
Archaeological correlates for the economic factors are quite similar to those used
to measure wealth, described in more detail below. Due to the material nature of
archaeological data, both the proportion of possible bundles and total number of bundles
for a given artifact class or classes can be relatively easily quantified. The bundle
perspective also provides a strong argument for looking at the degree of variation present
in any given measure, in addition to measures of central tendency, as cultural
perspectives on how well off one is are heavily influenced by how much room an
individual sees for improvement. In addition to looking at households, broader measures
231
of economic standing should also take into account investment and access at levels above
the household, such as neighborhood-constructed terraces, district temples, or city ballcourts. Economic investment above the household level has been less considered
archaeologically, though Pérez Rodríguez’s (2006) study of suprahousehold terrace
construction in the Mixteca provides an exception.
It is more difficult to develop archaeological correlates for social factors. On the
household level, social connectivity can be estimated from the number of local and
foreign sources of goods a household had access to (Smith 2015d). This aspect of quality
of life is primarily discussed in the preceding chapter on trade, but is also integrated into
the discussion at the end of this chapter. At the suprahousehold level, the presence of
community facilities, such as plazas, markets, and temples can be used as indicators of
interaction within a community. Archaeological work in this area has focused on analyses
of accessibility and visibility (e.g. Stockett 2005). However, because the first three out of
four construction phases for the monumental architecture at Calixtlahuaca cannot be
firmly dated (Sergheraert 2011), I am not able to consider superhousehold level economic
investment or social connectivity in this chapter.
Bioarchaeological research has also generated a large body of data on measuring
quality of life, focusing primarily on an individual’s physical condition. Physical quality
of life incorporates aspects of genetics, as well as the economic and social aspects of
quality of life. Measures of physical well-being can provide a complement to several of
the levels of analysis described above by providing information on health, nutrition, and
mortality. Examples of such studies at the general level include Steckel (2008) and
Steckel and Rose (2002). Depending on the burial practices of a group, bioarchaeological
232
data may be applicable at the household level (when significant numbers of individuals
are buried around house structures) or at the community level (when individuals are
buried in cemeteries or other non-house-associated patterns). Because of the limited
number of burials encountered in the household excavations at Calixtlahuaca (n=6), and
the fact that most of them cannot be assigned to a phase, I do not include
bioarchaeological data in this chapter’s analyses.
Economic history has made some attempts at broad, cross-cultural or diachronic
regional evaluations of the standard of living over time, based on both archaeological and
documentary evidence (Allen, et al. 2011; Morris 2004a, 2004b; Scheidel 2010). While
standard of living is a more economically focused measure than quality of life, it is also a
more archaeologically approachable measure. This body of work has demonstrated that
both ancient and proto-modern societies underwent periods of economic growth, and that
there was substantial temporal and regional variation in pre-modern standards of living.
Wealth
Wealth is a measure of valued goods, both social and material, that an individual
or larger social unit can control (Smith 1987b). It can be measured at a variety of
different scales, including the society, the community, and the individual household,
depending on the level of analysis. Traditional archaeological measures of wealth have
focused on three primary lines of evidence – household architecture, artifact assemblages,
and burial goods. As the latter more commonly actually measures status, I address it
under the next subheading.
233
Houses are likely to be the single largest investment that a family makes and are
thus likely to be a good indicator of the long term prosperity of a household. House size
is positively correlated with income, both in Mesoamerica (Tax 1953; Wilk 1983), and
elsewhere around the world (Bodley 2003; Ellis 2000; Yang 1945), though the number of
people in the household may also play a large role. In archaeological contexts with good
surface preservation, household size and/or volume can often be estimated without the
need for excavation, or extrapolated from a limited excavation sample (e.g. Smith, et al.
2014), especially if there is little known variation in construction methods within a
community. In cases with more extensive excavation, more detailed estimates of labor
investment in construction are possible (Murakami 2010). Brown et al. (2012) use both
absolute (area) and relative (Gini indices and Pareto distributions) distributions of house
sizes to compare wealth levels and structures at several Maya sites.
A second approach to wealth estimation uses larger assemblages of artifacts to
develop wealth indices by assigning various types of weighted value scores to different
types of material culture and then measuring the resulting net value. This method can be
applied to a wider range of archaeological data, as it does not require the preservation and
excavation of entire residential structures. For central Mexico, the comprehensive work
on using ceramics in wealth indices has been done by Garraty (2000). He applied six
different indices to ceramics from Aztec-period Teotihuacan. Other culturally appropriate
indices using a broader range of artifact types have also been developed by Olsen (2001).
Such indices measure economic standing at the level of the archaeological unit of
analysis, often the household (excavation) or neighborhood (survey). I use artifact based
methods to measure household wealth at Calixtlahuaca, including both analyses of
234
individual artifact classes (ceramics, lithics, ground stone, and rare items) and multiclass
indices combining all of the previous artifact classes.
When both architecture and other artifact classes are available, the former can be
used as an independent variable for assessing measures of household wealth based on the
latter. In such cases, a discriminant analysis of artifacts from contexts with clear
architectural differentiation can be used to assign artifact assemblages from more
ambiguous contexts to one or more wealth/status groupings. Such techniques also help
identify which artifact types co-vary the most closely with architecture. Olsen and Smith
(2016) apply these techniques to excavated households at several sites in Morelos to
reinforce previous interpretations of commoner, elite, and probably intermediate
(commoner barrio head) households.
Status
Archaeologists often conflate status and wealth, which are different, though not
independent variables. While this is conceptually problematic, the archaeological
applications are less so. This can be illustrated with the historical counter-examples of the
impoverished noble and the nouveau riche industrialist. The first has far more status than
wealth, and the latter the opposite. However, over the multigenerational timescales that
make up archaeological time periods, the industrialists’ daughters marry into the nobility,
once again merging wealth and status. Thus, while there may be short-term discrepancies
between the two factors, the longer scale pattern will be one of convergence.
In addition, in premodern societies most surplus value was generated by
agricultural production. However, in many cases, including Postclassic Central Mexico,
235
land was not a commercialized good, and could not be bought and sold. This likely
strongly limited the number of cases where commoners could successfully achieve a
discrepancy between their social status and their economic status. One possible
exception, the pochteca (merchants), is discussed in more detail below.
Archaeologically, status has been investigated both at the level of the individual
and at the level of the social class. Investigations of individual status are usually based on
mortuary contexts, and can be nuanced explorations of multiple dimensions of status,
including social class, social subgroup (much as sodality or moiety) membership, age,
gender, and/or expertise in particular skills (e.g. Crown and Fish 1996; Cucina and
Tiesler 2003). At the level of the social class, status can be investigated either through
analysis of lifestyles, or the investigation of mortuary contexts. In both cases, the
researcher usually seeks the presence of clearly differentiated sub-populations as
evidence of social statuses. Investigations of lifestyles are usually de facto measuring
wealth (e.g. Hirth 1993; Steere and Kowalewski 2012). In contrast, investigations of
mortuary contexts are more likely to identify status, based on the elaborateness of
funerary architecture and/or mortuary goods, or the presence of goods known to
correspond to particular restricted social roles. Analysis of social status based on burials
is most useful in cultures that invest heavily in mortuary practices.
Postclassic Central Mexican mortuary practices were relatively simple, limiting
this line of inquiry at Aztec sites in general. Excavations of commoner households
sometimes recover subfloor or courtyard burials, frequently of infants or children,
accompanied by a few objects. Adult skeletons are less common, and very rarely account
for even the minimum number of adult residents of a house (De Lucia 2014; Overholtzer
236
2012). Clear elite burials are also next to unknown; none of the Aztec Emperors’ tombs
have ever been located. The Toluca Valley may be an exception to this pattern, as almost
every excavated site, including Teotenango (Zacarías B. 1975), Tlacotepec (McVicker, et
al. n.d.), Metepec (Carbajal Correa and González Miranda 2003), Valle de Bravo
(Murillo Rodríguez 2002), and Calixtlahuaca itself (García Payón 1941b), has included a
large number of plaza burials (Castillo Romero 1996). The data on specific associations
between buried individuals and grave goods are not available for the García Payón
excavations. The current CAP only recovered six burials, all of which had none to
minimal grave goods. As a result, I do not consider burial evidence in this work.
At Calixtlahuaca, I consider status primarily as something to be controlled for,
since any elite households in a primarily commoner sample would result in considerable
bias. There is no a priori evidence, such as differential architectural construction or
restricted artifact types, to suggest that the sample includes any elite household
components.
Wealth Variation in Postclassic Mesoamerica
Most research into wealth in Postclassic Mesoamerica has focused on social class
related wealth variation. Ethnohistoric documentation of Mesoamerican status patterns
describes a clear two-class system of elites and commoners (Hicks 1996; Smith 2003a:
chapter 6). There is also documented variation within each class. Elites ranged from the
Huehuetlatoani (Emperor) in Tenochtitlan to minor provincial nobles in villages. Where
census documentation is available, elites account for approximately 1-2% of households
237
or 5% of the population (Carrasco 1964; Smith 1993). Commoners included individuals
in lower leadership positions, such as calpulli (neighborhood) leaders, as well as
individuals owing various degrees of obligation to the elite leadership. Commoners
included both farmers and part- and full-time craft specialists. Descriptions of elites and
commoners present a clear picture of wealth differentiation as well as status divisions,
though the degree to which this reflects cultural perceptions as opposed to material
realities will require further archaeological research (Olson and Smith 2016). In addition,
there is an ambiguous middle-status group of pochteca, or long distance merchants. The
pochteca generally operated as a self-regulated guild, at least some of whom were
wealthy enough that they needed to be discrete to avoid getting in trouble with the
nobility (Nichols 2013).
Archaeological investigations of status and wealth in Postclassic Mesoamerica have
produced varying results. Some researchers do find a clear division in households, which
would correspond to an elite/commoner distinction (Charlton and Nichols 1992; Olson
and Smith 2016). In contrast, other researchers find a more continuous range of variation.
In some of the latter cases, researchers assume the presence of a two-class system and
assign top 5% (or similar value) of households to an elite class, regardless of the degree
to which these households are materially distinct (Garraty 2000). Others, upon finding a
relatively continuous distribution of wealth markers, question the two-class model
(Brumfiel and Robin 2012; Steere and Kowalewski 2012), without extensively
questioning how much “fuzziness” is likely to occur in the association of wealth and
social class. There has been some discussion of the amount of variation within the elite
class (Elson and Covey 2006), but little consideration of the amount of variation within
238
commoner households. (But see Carballo (2009), Hirth (1993) and Brown et al (2012) for
studies of commoner or site-average household variation from earlier periods.) The
analyses in this chapter serve both to establish the range of variation among commoner
households in the Aztec world and relate this variation to broader economic and political
changes over time.
Analyses of Wealth at Calixtlahuaca
Due to the nature of the data recovered at Calixtlahuaca, the bulk of the potential
analyses are most directly measures of wealth, based on artifact assemblages. Most
houses were too incompletely preserved to measure area or volume, and too few burials
were encountered during the modern excavations to evaluate the range of mortuary
practices. As a result, I present a variety of wealth indices based on individual artifact
classes. Following Hirth’s (1993) argument that evaluations of household wealth are
strongest when drawing on multiple lines of evidence, I then discuss the degree to which
these indices identify the same households as being at the top or bottom on the wealth
distribution. This is followed by the presentation of a method to incorporate multiple
artifact classes into a single index, using prices from historical documents to set the
relative values of different artifact classes. After the presentation of the analyses, I
discuss their larger implications for wealth and quality of life at Calixtlahuaca and how
changes in these relate to broader issues of rulership strategies.
239
Ceramics
Ceramics are the single most common artifact type in most Mesoamerican
household artifact assemblages, both by number and by weight. They include utilitarian
items that all households would have needed as well as a wide range of more elaborate
vessels for specialized functions or display. Ceramic vessels break and are replaced on a
regular basis, and thus, the ceramic assemblage of a household component serves as
evidence of what a household was regularly able to acquire.
Ceramic-based measures of wealth generally fall into two categories. First, there
are measures based on relatively simple ratios or frequencies. This category includes
bowl/jar ratios, decorated/plain ceramic ratios, and slightly more complex measures
incorporating two or three factors or ratios. Second, there are labor-based evaluations of
value, based on a step-by-step assessment of the complexity of production for each
ceramic type. I present several measures from each category below, and then discuss the
degree to which they agree with each other.
Because I use ceramics as the baseline type for determining the frequency of other
artifact types, absolute ceramic frequencies cannot be standardized against some other
artifact type to provide a measure of absolute quantity. Due to the variation in component
contents – some are mostly midden, some are refuse scatter, and some are a combination
– standardization against excavated volume is not appropriate. All of the ceramic wealth
measures are based on rim sherds from the DS-1 sample only.
Frequency-based Ceramic Indices. The three basic measures calculated for
ceramics are the percentage of serving vessels (bowls, copas, and pitchers), the frequency
240
of decorated ceramics, and Index 2 from Olson and Smith (2016), which includes
weighted values for local decorated and imported ceramics (Table 6.1). The first is a
measure of the amount of wealth a household invested in vessels intended for presenting
food to others, rather than those needed for day-to-day food storage and preparation. This
measure is likely related to a household’s ability to host feasts or other gatherings. The
second, the percentage of decorated vessels in an assemblage (or the ratio of decorated to
undecorated ceramics), is a closely related measure, since most serving vessels are also
decorated (Feinman, et al. 1981; Garraty 2000). Previous studies of Aztec households
have demonstrated that the frequency of decorated ceramics in household assemblages
tends to increase across Postclassic phases prior to a site’s conquest by the Aztec Empire,
but drop during the period of Aztec dominion (Brumfiel 2005a). A third index, based on
Olson and Smith (2016) incorporates aspects of the previous two indices, combining the
frequencies of local decorated ceramics and twice the frequency of imported ceramics (%
local decorated ceramics + 2*% imported ceramics). When the artifact assemblages from
elite and commoner Aztec households (as determined based on architecture) in Morelos
are compared, these artifact classes show the greatest differentiation between classes
(Olson and Smith 2016). This index is closely related to prior formulations proposed in
(Smith 1992a) and (Garraty 2000).
At Calixtlahuaca, the frequency of serving vessels does not differ significantly
between any pair of phases, ranging from 54-57% of average household rim sherds. The
coefficients of variation for all three phases are very low, indicating that there is little
variation among households based on this measure. Similarly, no component falls more
than two standard deviations from the mean for all components, indicating that there are
241
no extremely anomalous cases among the households in the sample. There is also a high
degree of overlap in the range of values present during each phase and the next.
Unit
307
315
316
320
323
324
303
307
308
311
316
322
307
309
316
317
324
327
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Yata
Yata
Yata
Yata
Yata
Yata
Dongu Average
Ninupi Average
Yata Average
%
%
Decorated
Serving
Vessels Rank
Vessels Rank
54.99 3
22.14 6
51.06 6
27.66 5
53.79 5
32.41 2
54.43 4
30.13 4
56.24 2
32.25 3
57.14 1
35.06 1
61.81 2
25.97 6
62.49 1
36.77 2
61.59 3
30.09 3
52.98 5
27.72 5
52.90 6
29.88 4
54.59 4
41.23 1
60.37 2
32.12 3
60.24 3
32.83 2
56.09 4
35.61 1
61.04 1
30.16 4
51.55 6
25.00 6
54.64 5
28.87 5
Olson &
Smith
Index 2 Rank
24.33 6
29.04 5
33.79 3
32.66 4
36.69 2
49.35 1
28.16 6
38.18 2
34.04 4
32.89 5
34.12 3
50.24 1
43.89 3
51.36 1
46.15 2
42.28 4
30.06 5
28.87 6
54.61
57.73
57.32
29.94
31.94
30.77
34.31
36.27
40.43
Dongu St Dev
Ninupi St Dev
Yata St Dev
2.12
4.69
3.84
4.56
5.84
3.66
8.51
7.56
9.04
Dongu CoVar
Ninupi CoVar
Yata CoVar
0.04
0.08
0.07
0.15
0.18
0.12
0.25
0.21
0.22
Table 6.1 Simple ceramic-based wealth measures (% serving vessels, % decorated
ceramics, Smith index of decorated and imported ceramics) by household component.
DS-1 sample, rims only.
242
The frequency of decorated vessels is approximately the same across all three
phases. This index features the combined frequency of decorated bowls, decorated jars,
copas, and pitchers. The presence of decorated jars in the Toluca Valley means at
Calixtlahuaca this measure is less closely linked to the previous one than in many other
parts of Postclassic Central Mexico, where almost all decorated vessels are bowls. The
inter-household variation in this measure is low. The coefficient of variation remains low
for all three phases and there in only one extreme outlier case (322-Ph4) falling more
than two standard deviations from the mean for all components.
A measure based on a combination of decoration and imports, Olson and Smith’s
(2016) Index 2 shows modestly more interphase variation than those produced by the
previous two indices. The values produced by this index are comparable between the
Dongu to Ninupi phases, rising slightly between the Ninupi and Yata phases. The
coefficients of variation for all three phases are very low, indicating little overall
variation among households. There are no extreme outliers more than two standard
deviations from the mean for all household components. (Note – in order to be
comparable to the other ceramic wealth measures in this section, this index was
calculated using rim sherds only. This produces higher values than the total sherd counts
used in the original article.)
The results of the three simple ceramic indices generally show little change in
ceramic wealth over time at the site level, as well as little variation among households.
Two of the three measures, vessel form and proportion of decorated ceramics, do not
have phase means that differ significantly among phases. The third, the Olson and Smith
Index, does show a modest increase over time, driven by the increasing frequency of
243
Aztec imports noted in Chapter 4. All three indices have low coefficients of variation for
all phases, and there is only one case of an outlier household falling more than two
standard deviations from the mean. The rankings of individual household components
within each phase are variable among the indices. Combined with the low overall degree
of intrahousehold variation, this suggests that the variation in these indices is too low to
be behaviorally meaningful.
Production Step Ceramic Indices. Production step indices are a measure of the
labor investment in a given artifact. For ceramics, they will assign a point value to
various aspects of production, such as size, presence of decoration, complexity of
decoration, presence of appendages, and/or transport costs. While this is a more specific
measure of labor input than those described above, it does not take into account
culturally-specific notions of value, which may differ from those based on labor input
alone (e.g. Voss 2012), and thus remains a general proxy for value. The following
discussion presents four variations of a production step index, each using rim sherds from
the DS-1 sample (See Table 6.2 for the complete production step scoring rubric). The
first variation is based strictly on the decorative complexity of the piece, with points
assigned for interior and exterior base color, additional interior and exterior linework
paint colors, and the complexity of decoration. The second index variation maintains the
decoration-based score and adds a component for vessel form, which includes points for
vessel size and the presence of appendages. The third index variation includes the
decoration-based score and adds a transportation distance factor, assigning additional
points to non-local ceramic types. The fourth variation combines the weighting for all
244
three types of data – decoration, form, and distance from the place of production. In each
variation, the points assigned to various factors were added up to provide a weighting
value for the ceramic type. For each component, the rim sherd count for each type was
multiplied by the weighting factor for the type. The weighted results for each type were
summed, and then divided by the total number of rim sherds, to provide an average
production step score for the household component.
State
Value
Decoration
Background color, exterior
1
Background color, interior
1
Decorative color, exterior
1 per color
Decorative color, interior
1 per color
Elaborate Decoration, exterior
1
Elaborate Decoration, interior
1
Form
Appendages, some examples of type
1
Appendages, always on type
2
Vessel Size (XS/S/M/L/XL)
0-4
Source
Non-local type
1
Table 6.2 Ceramic production step value scoring
The production step values for all types decrease slightly from one phase to the
next for all four variants of the index (Table 6.3), when all ceramic types are included and
the overall differences between the Dongu and Yata phases are significant at the .90 level
for all four variants. On the surface, this would suggest that the ceramic assemblage is
being simplified, if very slowly. However, when eroded types, which were assigned low
production step values, are removed from the sample, the average production step values
245
for the three phases are not significantly different for any of the four variants of the
index. The high degree of agreement between each set of four analyses suggests that the
additional variables (vessel form, source region) reinforce rather than oppose the broader
pattern based on general decoration. The general trend is one of stability in investment in
ceramic decoration over time, a finding which would support relatively indirect rule by
the Aztec Empire due to the lack of observable effects of imperial rule.
While mean trends provide information about the directness of rule, the
distribution of variation within each phase provides information about the collectiveness
of rule. Due to the similarities among the production step index variants, I only use the
fourth, incorporating vessel decoration, form, and source, to look at distributions. The
coefficients of variation, which are a relative measure of dispersal of values around the
mean, are both low and similar for all three phases, with the differences among phases
driven primarily by two low outliers (Table 6.3). These are 307-Ph2 and 307-Ph6, which
fall more than two standard deviations below the mean. All of the other household
components fall within one standard deviation of the mean. This suggests that despite
changes in the preferred types of decorated ceramics and the sources of traded goods over
time, the overall amount of labor investment in a household’s basic ceramic assemblage
varied little among households, a position consistent with a relatively collective social
organization of local power, both before and during Aztec rule.
246
All Types
Excluding Eroded Types (0, 10, 30, 60)
N. Decora Dec & Dec &
tion Form Source
Sherds
411
1.72
4.24
1.74
1,371
2.01
4.82
2.02
290
2.36
5.12
2.38
395
2.05
4.82
2.07
834
2.47
5.16
2.52
77
2.58
5.12
2.73
All
4.26
4.84
5.13
4.84
5.21
5.26
N. Decora Dec & Dec &
tion Form Source
Sherds
315
2.07
4.42
2.10
1,019
2.52
5.29
2.54
277
2.44
5.16
2.45
242
2.97
5.45
3.00
788
2.59
5.28
2.64
77
2.58
5.12
2.73
Unit
307
315
316
320
323
324
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
All
4.45
5.31
5.17
5.49
5.33
5.26
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
817
2,124
328
570
1,847
207
1.79
2.51
2.09
1.86
2.07
2.37
4.43
5.10
4.70
4.51
4.81
5.16
1.81
2.52
2.13
1.91
2.11
2.45
4.45
5.12
4.73
4.56
4.84
5.25
466
1,928
250
434
1,459
185
2.79
2.70
2.62
2.28
2.50
2.58
5.21
5.23
5.14
4.79
5.19
5.34
2.83
2.71
2.66
2.34
2.55
2.68
5.25
5.25
5.19
4.85
5.24
5.44
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
1,128
332
920
1,114
355
97
1.72
1.72
2.16
1.64
1.79
2.02
4.36
4.36
4.83
4.28
4.56
4.77
1.83
1.89
2.26
1.74
1.83
2.02
4.46
4.53
4.93
4.38
4.60
4.77
930
180
813
584
257
73
1.88
2.76
2.39
2.79
2.26
2.60
4.24
5.12
5.05
5.30
4.91
5.44
2.01
3.08
2.50
2.98
2.32
2.60
4.36
5.44
5.16
5.50
4.97
5.44
Dongu Average
Ninupi Average
Yata Average
2.20
2.12
1.84
4.88
4.78
4.53
2.24
2.16
1.93
4.92
4.82
4.61
2.53
2.58
2.45
5.12
5.15
5.01
2.58
2.63
2.58
5.17
5.20
5.15
Dongu St Dev
Ninupi St Dev
Yata St Dev
0.33
0.28
0.20
0.35
0.30
0.23
0.36
0.28
0.19
0.37
0.31
0.21
0.29
0.18
0.34
0.36
0.19
0.42
0.30
0.17
0.40
0.37
0.19
0.43
Dongu CoVar
Ninupi CoVar
Yata CoVar
0.15
0.13
0.11
0.07
0.06
0.05
0.16
0.13
0.10
0.08
0.06
0.04
0.11
0.07
0.14
0.07
0.04
0.08
0.12
0.06
0.16
0.07
0.04
0.08
Table 6.3 Average production step index values by household component. DS-1, rims
only
247
Lithics
Postclassic Central Mexican lithic assemblages consist primarily of obsidian (e.g.,
Hodge 2008; Millhauser 2005; Smith 2006a, 2006c). As a lightweight, high value good,
obsidian was traded across large distances. The lithic assemblage at Calixtlahuaca
conforms to the expected regional pattern, consisting primarily of obsidian, with limited
amounts of basalt and the occasional piece of chert. XRF sourcing has demonstrated that
the obsidian at the site comes from both Central Mexican and West Mexican sources,
primarily Otumba, Pachuca, and Ucareo, with occasional pieces from other sources in
both regions (Glascock 2012). See Chapter 4 for a more detailed analysis of obsidian
importation over time. The lithics data in this section are from the basic lithic
classification performed by the project lab staff, rather than the more specialized analyses
of production done by Dr. Bradford Andrews, and uses the DS-1 sample. This version of
the lithic classification was used because its simpler format made extracting necessary
information easier. Two potential wealth markers are calculated for each household
component: the frequency of green obsidian per 1000 sherds and the frequency of all
obsidian per 1000 sherds (Table 6.4).
The most common non-ceramic wealth indicator used in Central Mexico is the
proportion of green obsidian in a lithic assemblage. Within Central Mexico, green
obsidian comes from a single source, the Pachuca volcano. Pachuca green is a high
quality obsidian and would have been functionally equivalent to or better than obsidian
from other widely traded sources in the region (Cobean 2002). Because blue/green was a
symbolically valued color, archaeologists assume that the visibly green obsidian from the
Pachuca source would have been the preferred if consumers had a choice. For most of
248
Central Mexico, which forms a single trade sphere for obsidian (Braswell 2003), this is
probably a valid assumption. However, the results of the sitewide survey at Calixtlahuaca
show that Central (including Pachuca) and West Mexican obsidians have different
distributions within the site (Novic 2015). Given the shifts in overall source frequencies
over time at the site, it is unclear whether the survey pattern is due to differential lengths
of occupation of different portions of the site, or differential access to particular sources.
As a result, I include the percentage of green obsidian in my evaluation of household
wealth, but give it a secondary importance relative to the total obsidian frequency.
The first pattern measures the frequency of obsidian relative to ceramics over
time, which is a proxy for the absolute quantity of obsidian used by each household. The
frequency of lithic artifacts is even between the Dongu and Ninupi phases, at 39-41
pieces of obsidian per 1000 sherds. It then drops dramatically during the Yata phases,
averaging about 25 lithic pieces per 1000 sherds. However, due to the high standard
deviations for this measure, there is only approximately a 70-75% probability that this
difference is non-random. There is a high degree of intercomponent variability in the
lithic/ceramic ratio, with households ranging from 7 to 93 lithic artifacts per 1000 sherds.
The variation is continuous, without clear breakpoints or modalities that would suggest a
distinction such as craft workshops. This agrees with the previous chapter, which showed
no evidence for intensive lithic production at the site. When measured more formally, the
coefficient of variation increases slightly from each phase to next.
249
Obs. per
Green/
1000 Green per Sherd
sherds 1000 sherds Rank
57.86
6.08
3
6.98
1.87
5
9.84
0.66
6
24.48
3.91
4
78.41
6.62
1
57.99
6.56
2
Unit
307
315
316
320
323
324
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
Total
All
Green
Sherds Obsidan Obsidian % Green
4,770
276
29
10.51
13,890
97
26
26.80
3,050
30
2
6.67
3,840
94
15
15.96
8,915
699
59
8.44
914
53
6
11.32
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
8,951
20,280
3,729
5,030
15,800
1,668
211
584
109
427
173
118
38
235
40
40
64
79
18.01
40.24
36.70
9.37
36.99
66.95
23.57
28.80
29.23
84.89
10.95
70.74
4.25
11.59
10.73
7.95
4.05
47.36
5
2
3
4
6
1
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
10,200
3,094
7,451
9,638
3,438
948
260
70
92
76
185
24
131
41
44
48
36
13
50.38
58.57
47.83
63.16
19.46
54.17
25.49
22.62
12.35
7.89
53.81
25.32
12.84
13.25
5.91
4.98
10.47
13.71
3
2
5
6
4
1
Dongu Average
13.28
39.26
4.28
Ninupi Average
Yata Average
34.71
48.93
41.36 7.71 (14.32) 1
24.58
10.19
Dongu St Dev
7.33
Ninupi St Dev
Yata St Dev
20.00
15.46
29.51
2.57
29.34 3.52 (16.49) 1
16.05
3.86
Dongu CoV
1.81
1.33
1.67
Ninupi CoV
Yata CoV
1.74
3.17
1.41
1.53
2.19 (0.87) 1
2.64
1
Excluding(Including) component 322-Ph4
Table 6.4 Obsidian-based wealth indices (Total obsidian/1000 sherds, green
obsidian/1000 sherds, % green obsidian) by household component
250
The second pattern is the frequency of green obsidian per 1000 sherds, which
provides a standardized measure of the absolute quantity of green obsidian used by each
household. There is one outlier household for this wealth measure, 322-Ph4, which has
approximately fifty pieces of green obsidian per 1000 sherds, compared to less than
fifteen for all other components. This component is one of two located on the smaller
secondary hill to the west of the site, where the survey showed a significantly higher
frequency of eastern obsidian sources (including Pachuca). Overall, the resulting pattern
shows that green obsidian is rare during the Dongu phase, and then generally increased
over time, if the outlier household 322-Ph4 is excluded. The differences between the
Dongu phase mean and those for both subsequent phases are statistically significant at the
.90 level, while the difference between the Ninupi and Yata phases is not (with or without
the outlier). The interhousehold variation in the frequency of green obsidian per 1000
sherds, as measured by the coefficient of variation, increases from each phase to the next.
The total range of variation, between the highest and lowest scoring households, also
increases from each phase to the next.
In summary, the obsidian data show improving access to lithics between the
Dongu and Ninupi phases, with comparable overall levels of obsidian and increasing
access to green obsidian. This is then followed by a drop in the total volume of obsidian
entering the site, and a lack of further improvement in the average level of access to
green obsidian. Despite the fluctuating quantities of obsidian entering the site, the
coefficient of variation for all measures shows small increases from each phase to the
next, suggesting that access to lithics became increasingly differentiated over time. This
increased variation is unlikely to be due to changes in the organization of craft production
251
over time, as there was very little overall evidence for such production and it decreased
over time (See Chapter 5).
Rare Items
In addition to everyday items, such as ceramics and lithics, the occupants of
Calixtlahuaca also had to make choices about acquiring rare, highly valuable classes of
objects. At Calixtlahuaca, these are bronze items7 (including bells, tweezers, needles, and
earspools) and stone jewelry (including pendants, beads, lip plugs, and earspools, made
from crystal, obsidian, or turquoise). While the circulation of some rare types of objects
may be limited by sumptuary laws in some cultures, all of the examples included in this
category have been previously demonstrated to occur in commoner households in other
Aztec cases (e.g., Brumfiel, et al. 1993; Smith and Heath-Smith 1993). Because the
recovery of such items is likely to be highly influenced by random chance and excavation
size, I consider them primarily on a presence/absence basis.
Metal artifacts in Mesoamerica are often described as being “copper”. Given that they are usually made
from carefully manipulated alloys of copper, tin and/or arsenic, the more accurate term is “bronze”. See
Hosler (1994) for the basics of metallurgy in Mesoamerica.
7
252
Jewelry
0
0
0
0
0
0
Total
Rare
Items
1
0
0
0
0
0
0
13
0
1
0
0
0
3
1
1
0
0
0
16
1
2
0
0
0.00
0.72
0.23
0.26
0.00
0.00
A
HF
P
P
A
n/a
10,257
4,217
10,091
10,860
3,438
1,266
2
1
1
0
0
0
0
0
0
0
1
0
2
1
1
0
1
0
0.19
0.24
0.10
0.00
0.29
0.00
P
P
P
A
P
n/a
175,462
19
6
25
0.14
Unit
307
315
316
320
323
324
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Dongu
Total
Sherds Bronze
5,810
1
16,775
0
4,710
0
12,189
0
26,947
0
914
0
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
9,043
22,330
4,359
7,838
22,563
1,855
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
Total
Dongu Average
Ninupi Average
Yata Average
1
Freq./
1000 Presence/
Sherds Absence 1
0.17
P
0.00
A
0.00
n/a
0.00
A
0.00
A
0.00
n/a
0.03
0.20
0.14
0.10
n/a=Unknown due to small sample size, A=absent, P=present, HF=High Frequen
Table 6.5 Rare items counts by household component with frequencies per 1000 sherds.
DS-2 sample.
Rare artifacts are coded on a four part scale: n/a due to small sample size, absent,
present, or present in high frequencies (Table 6.5). The category n/a due to sample size is
used for components without rare items, which also had fewer sherds than the average
number of sherds per rare item in all excavations. This category is due to the fact that the
253
size of the excavation will play a large role in the probability of recovering extremely
rare artifacts. In the DS-2, there is approximately one bronze or jewelry item for every
7,000 excavated sherds, so for components smaller than this, it is difficult to say whether
rare items are missing due to sample size or due to true absence. Absent is used for
coding components that are above the threshold size, but which do not have any rare
artifacts. Present is used for components with a rare item to sherd ratio within one
standard deviation of the mean, and High Frequency is used for components with a
frequency of more than one standard deviation above the mean.
The frequency and ubiquity patterns show similar results. The frequency of rare
items begins at a very low level in the Dongu phase and then rises to a much higher level
during the following two phases. The patterning between the Ninupi and Yata phases is
dependent on whether household component 307-Ph4, an extreme outlier, is included.
Based on an independent t-test of means, the Dongu to Ninupi phase (including the
outlier) difference has a p-value between .85 and .90, while the overall Dongu to Yata
phase difference is significant at the .90 level. The Ninupi and Yata phases are not
significantly different, regardless of whether the outlier case is included or excluded.
Similarly, the ubiquity of rare items increases from a single household during the Dongu
phase, to three households during the Ninupi phase and four households in the Yata
phase. As the larger and smaller components are fairly evenly distributed across the
phases, the change in ubiquity is unlikely to be a result of variation in sample size.
While the small sample sizes for rare items mean that the observed frequency for
any given household may be result of random variation, the overall pattern is one of
increasing access to rare items between the Dongu and Ninupi phases, followed by
254
similar levels of both sitewide access and interhousehold variation between the Ninupi
and Yata phases.
Summary of Individual Index Results
The previous analyses of individual artifact classes focused on three variables: the
magnitude of change from each phase to the next, the directionality of that change, and
the degree of variation within each phase. Of these three markers, the magnitude of
change is used as an indicator of the directness of Aztec rule, while the directionality of
change and the degree of intra-phase variation are used as indicators of the collectiveness
of governance under local and Aztec imperial rule.
The majority of the wealth indices for different goods presented here show a
consistent pattern at the phase level, but a higher level of variability at the level of the
individual components. Of the five indices that are relatively independent of each other
(serving vessels, decorated ceramics, average number of production steps, green obsidian
to sherd ratio, and rare item to sherd ratio), three (serving vessels, decorated ceramics,
production steps) are basically the same across all three time periods, and two (green
obsidian, rare items) show a pattern of a relatively large increase in the wealth measure
between the Dongu and Ninupi phases, followed by a leveling off with similar values for
the Ninupi and Yata phases (Figure 6.1). I consider this latter pattern to be the most
representative measure of the relative wealth of Calixtlahuaca’s residents over time. This
demonstrates a trend toward increasing access to goods prior to the Aztec conquest of the
site, followed by a cessation of this growth under Aztec rule.
255
The temporal patterning of inter-household variation at the site also differs among
the indices. All of the ceramic-based indices have very low coefficients of variation, with
similar values across all three phases. In contrast, the obsidian and rare item based indices
generally have steady increase in coefficients of variation from each phase to the next.
Figure 6.1 Summary of selected indices by phase
When the two highest and lowest scoring components for each phase for each
measure are plotted, however, they do not consistently identify the same households
(Figure 6.2). While there are general trends for particular households during each phase
to score toward the top or bottom of multiple measures, only one household component
(307-Ph4) manages to score in the top third for all five measures. Additionally, only two
components do not fall within the top third for at least one measure. Because of the
number of components with missing data on rare items, I did not assign low-scoring
256
components for this index. Half of the components include both a low score according to
one measure, and a high score according to another. The consistency among the sets
indices that track together at the phase level – the three ceramic indices vs. green
obsidian/rare items – are not noticeably better or worse than among all five indices. This
suggests that individual households were choosing to invest in different types of wealth
items, within the larger patterns of availability dictated by temporal trends in trade and
the necessities of differing household activities. As a result, a wealth index that can
include multiple artifact types will likely be a more accurate measure of the relative
rankings of individual households.
Dongu Phase
Ninupi Phase
Yata Phase
307 315 316 320 323 324 303 307 308 311 316 322 307 309 316 317 324 327
Measure
Servingware
Decoration
Production Steps
Green Obs/Sherd
Rare Items (High only)
?
?
?
?
Figure 6.2 Summary of highest and lowest scoring components for each phase for
selected indices.
Master Wealth Index
A single wealth measure incorporating multiple artifact types will provide a more
accurate measure of overall household wealth for three reasons. First, it evens out small
variations in individual artifact classes resulting from random differences in artifact
recovery among households. Second, alternatively, it may allow for the identification of
257
subtle patterns across artifact types, which only become significant when combined.
Third, it provides a means of compensating for differences in household consumption
choices, whether these were based on specialized activities (such as needing more
obsidian tools for a particular craft activity) or social-group membership (such as
ethnicity). In such cases, investment in a single artifact class may be due to particular,
household-specific reasons, but this investment will simultaneously reduce the amount of
household assets which can be used to acquire other classes of artifacts.
Any attempt to combine multiple artifact classes into a single measure of wealth
must first establish their relative values. Two possible methods for assigning relative
values are possible. The first is to compare the energetic cost of production and transport
to the length of use. Another method is to use historical prices to establish relative values.
The first approach can be used to establish very basic energetics-based
calculations of value, such as those applied to the ceramic production step index above.
However, it cannot take into account factors of cultural preference or scarcity-based price
inflation. As a result, it is primarily useful as a check against which other methods of
calculating value can be compared, or as a means of identifying classes of goods with
anomalous values.
The second approach, of using documented prices to establish relative values is
primarily used by historians. Within the United States, historical archaeologists have
made wide use of documentary prices (e.g. Deetz 1977), and applications have been
spreading both temporally and geographically. Wills in English speaking countries often
include probate inventories, allowing for a wide range of studies of the wealth
distributions, household occupations, and gender issues (Bragdon 1988; Brown 1988;
258
Green Carr and Walsh 1980; Shackel 1992). These types of documents often provide
information on the value of exactly the range of small, everyday items of interest to the
household archaeologist, which has also resulted in their use in historical archaeology.
Document-based studies of wealth or value also draw on advertising material and store
ledgers to provide time-of-sale values for household items. In the latter vein, Miller
(1988) developed a pearlware index, measuring the relative values of ceramic types in
“CC” units, or the number of times more than a comparable vessel of the cheapest
possible ware other types cost, on the premise that the cheapest type would have such a
low profit margin that the value would remain constant (The abbreviation “CC” comes
from historical merchants’ records of such pottery and stands for “Cream Colored”). This
method can also be extended to provide comparisons between classes of goods, using an
item type with little potential for fluctuations in value as the base measure, and this is the
approach I follow in establishing the relative values of different artifact classes.
Historical Values of Domestic Goods in Mesoamerica
The corpus of everyday documents from Colonial Latin America has been much
less well explored than its English language equivalent. Spanish legal practice did not
require systematic probate inventories following a death, and thus the documentary
corpus includes far less systematic information about household goods. As a result, I
draw on price data from four different classes of documents: wills and probate
information, other legal documents, town council records, and histories produced by the
chroniclers. Where possible, I have drawn on sources likely to represent indigenous
views of the value of goods; most of the wills were written in indigenous languages, the
259
legal documents were entered as evidence in courts cases by Native individuals (though
often against Spaniards), and the town council records are from Native towns. Values
given by Spanish chroniclers or other non-native writers are used only when a value for
the type of item could not be found in another source. The resulting list of items and
prices are listed in Table 6.6.
There are three collections of published Native-language wills from single
locations, the Testaments of Culhuacan (Nahuatl) (Cline and León-Portilla 1993), the
Testaments of Toluca (Nahuatl) (Pizzigoni 2007), and Life and Death in a Maya
Community (Yucatec) (Restall 1995). There is also a three-volume published compilation
of colonial indigenous wills held in Mexico’s Archivo General de la Nación (Rojas
Rabiela, et al. 1999-2000), and various publications of single or small groups of wills
(Anderson, et al. 1976; Gasco 1992; Wood 1997). With a few exceptions, the wills do not
include systematic lists of the values of the goods owned by the deceased. As a result,
historians have focused more on land ownership and inheritance patterns in their studies
of these documents (García Castro 2000; Lockhart 1992; Pizzigoni 2013). However, it is
not uncommon for the values of a few items to be specified in a will, and these can be
compiled to provide values for a range of domestic goods.
I also draw price information from legal documents, primarily the Codex of the
Potters of Cuauhtitlán (also known as the Códice de los alfareros de Cuauhtitlán and the
Pieza de contaduría de una fábrica de cerámica indígena, posterior a la Conquista)
which is a pictorial document with Spanish glosses entered as evidence in a lawsuit in
AD 1568 (Cuauhtitlán 1568). The lawsuit was presented by a number of native potters
who were suing Juoan Suarez de Peralta, Alcalde Mayor, for failure to pay for goods
260
received. The document is held in the French National Library as Mesoamerican Codex
109 (Barlow’s identification of the piece as #107 is incorrect), and a transcription and
good quality images have recently been made available through the Amoxocalli project
(2009). It has been previously published in Barlow (1951), though the accompanying
images are of poor quality. Charlton and Fournier (2011) discuss the historical context of
the codex, including further information about Suarez de Peralta. The codex lists
contemporaneous prices for a wide range of ceramic vessel forms, produced in a single
town.
The third category of documents, town council records, comes primarily from two
compilations and translations of a wide range of types of native language documents
(Anderson, et al. 1976; Restall, et al. 2005). Among other things, these works provide
meeting minutes, receipts, and council decrees from multiple communities. Examples of
council records with price information include the widely cited decree of prices for
various goods at the Tlaxcala market (Anderson, et al. 1976:document 34), lists of
expenditures for celebrations, and lists of contributions to religious institutions. Such
reports most often give prices for foodstuffs, but do include some archaeologically
recoverable items.
The fourth category of documents, reports by Spanish chroniclers, was generally
only used in cases where they provided the only price for a given category of good, or
they were the original source for a value widely cited in the secondary literature. I
consider prices in large-scale summary works on Central Mexico to more likely reflect
perceived “normative” values for a given item over a longer period of time, as opposed to
the time-specific prices recorded in the other types of sources.
261
Table 6.6 Documentary values for domestic items recovered in Mesoamerican
archaeological contexts
262
Item
Ceramics
Simple Vessels
Complex Vessels
Tinaja (Large Vessel)
Michoacan-style Tecomate
Three broken tecomates
Jug
Collapsed tecomate,
cracked
Clay jar
Comal
Earthenware tub
Porcelain basin
Small pitcher or jar
Water jar (new)
Small Majolica cups
(Puebla)
Porcelain cups
Porcelain cup
Water jar (Tehuantepec)
Ground Stone
Two metates
Metate w/ mano
Metate
Mano (for a metate) (new)
Metate with mano
Metate w/o mano
Grinding stone (metate?)
Year
Given Price
1568
1568
1568
.19-.33 tomin each
.5 tomin each
3 tomines each
Given
Price,
Tomines
Inflation
Adjustment
1 year1 5 year1 Location
Source
Location in
source
0.25
0.5
3
2
0.09
0.19
1.14
0.46
0.12 Cuauhutitlan
0.23 Cuauhutitlan
1.39 Cuauhutitlan
0.44 Culhuacan
n.d. (1580) 2 .5 tomin
n.d. (1580) 2 .5 tomin
0.5
0.12
0.11 Culhuacan
Test. 13
0.5
0.12
0.11 Culhuacan
Test. 13
n.d. (1580) 2 15 cacao
0.15
0.03
0.03 Culhuacan
Test. 13
2
0.60
16
10.5
2
2
2
1.5
4.71
3.53
0.57
0.57
0.57
0.28
0.59 Santa Maria
Tezcatzonco
2.06 Amecameca
2.47 Soconusco
0.52 Soconusco
0.52 Soconusco
0.52 Soconusco
0.31 Soconusco
2
8
2-3
0.37
1.47
.37-.55
8
5
3
8
24
16
32
2.40
1.15
0.69
2.26
6.79
4.53
9.06
n.d. (1580) 2 2 tomines4
1587
2 tomines
1625
1687
1724
1724
1724
1740
2 pesos
1 peso, 2.5 reales
2 reales
2 reales
2 reales
1.5 reales
1740
1740
1740
2 reales
1 peso
2 or 3 reales
1597
1599
1599
1724
1724
1724
1724
1 peso
5 tomines
3 tomines
1 peso
3 pesos
2 pesos
4 pesos
Codice de los
Alfareros de
Cuauhutitlan
Cline & LeonPortilla 1993
Rabiela et al
1999-2000
Gasco 1992
v. 2; p. 272
v. 3; p. 168
Table 3
Table 3
Table 3
Table 3
Table 3
Table 3
Table 3
Table 3
0.41 Soconusco
1.66 Soconusco
.41-.62 Soconusco
2.00 Tecamachalco
1.29 Quauhtitlan
0.77 Quauhtitlan
2.09 Soconusco
6.27 Soconusco
4.18 Soconusco
8.36 Soconusco
Test. 13
Rabiela et al
1999-2000
Gasco 1992
v. 1; p. 156
v. 2; p. 334
v. 2; p. 334
Table 7
Table 7
Table 7
Table 7
263
Table 6.6 (continued). Documentary values for domestic items recovered in
Mesoamerican archaeological contexts
Item
Lithics
20 obsidian blades
Metal/Jewelry
Cozcatl (jewel)
String of beads
Feather bracelet
Emerald
Year
Given Price
1565-15753
1 real
1551
1569
1577
5 pesos
3 pesos
26 pesos
3 pesos
n.d. (1580) 2
Given
Price,
Tomines
Inflation
Adjustment
1 year1 5 year1 Location
Source
Location in
source
0.47 Central Mexico Mendieta 1945 p. 58
1
0.52
40
24
208
24
40.00
24.00
47.94
7.19
40.00 Moyotlan
24.00 Xochimilco
42.32 Tulantzinco
5.99 Culhuacan
Rabiela et al
1999-2000
Cline & LeonPortilla 1993
Gasco 1992
v. 2; p. 90
v.2; p. 152
v.2; p. 186
Test. 23B
Table 6
Blue and white glass beads 1654
4 reales
4
3.47
2.82 Soconusco
1654
1654
1724
1740
1740
1740
1 real
1 real
.5 real
.5 real
1 peso
8 pesos
1
1
0.5
0.5
8
64
0.86
0.27
0.14
0.12
2.41
54.79
0.77 Soconusco
0.25 Soconusco
0.13 Soconusco
0.11 Soconusco
2.24 Soconusco
49.09 Soconusco
Table 6
Table 10
Table 10
Table 6
Table 6
Table 6
1740
5 pesos, 3 reales
43
11.73
10.75 Soconusco
Table 10
Silver ring with no stone
Small tin medallion
Metal medallion
Ring from China
Crystal earrings/pair
Gold earrings with 34
pearls
Gold cross weighing 3/8 of
a peso
1
Prices standardized for inflation using "BBB" values for Central Mexico from the Global Price and Income Group 2013,based on the year
given for the document ("1 year") and the average of the preceeding five years ("5 year")
2
Date estimated based on position in original source document
3
Date estimate based on Hirth 2013
4
Tomines and reales are synonymous terms. Both are worth 1/8 of a peso.
Price Standardization
The information on historic prices contains internal variation from two sources:
reported currency and year of recording. I describe each issue and the methods used to
compensate for the resulting variation below.
First, prices are given in a number of different currencies. These include native
units, such as cacao beans and lengths of cotton cloth, and Spanish coinage, such as pesos
and tomines. The exchange rate between native and Spanish currencies changed over
time. Coins also came to contain lower percentages of silver over time, but the effects of
this devaluation are included in the compensation for inflation discussed under temporal
issues. In “Price, Tomines” column of Table 6.6, prices are standardized into tomines, as
this is the most common given denomination. Prices originally given as cacao beans are
converted using internal information in each document about the local relative cacao to
peso value.
Second, the sources also cover a wide temporal span, from AD 1551 to 1740. This
period saw substantial price inflation due to the amount of precious metals entering
circulation from New World mines. This requires some method of standardizing prices to
account for inflation, as one peso had wildly varying real purchasing power in different
decades. To do so, I use the concept of the “Bare Bones Basket” (BBB), measure of the
basic cost of living borrowed from economic history. The BBB has been used for
comparisons of the quality of life and real purchasing power between world regions
(Allen 2001; Allen, et al. 2011) and factors in the price of a staple grain (maize, for
Mexico), a small amount of other foodstuffs (such as meat, oil, or beans), and other basic
commodities (clothing, fuel, soap, and candles/lamp oil). It has also been applied to
264
ancient states where written records are available, such as Rome (Allen 2009) and Egypt
(Scheidel 2010), which generally show very low incomes for unskilled laborers. The
Global Price and Income History Group, based at University of California Davis (Global
Price and Income History Group 2013), has compiled data on the prices of basic
commodities in various Latin American countries, including Mexico, from the Spanish
conquest through independence (Arroyo Abad, et al. 2012). As part of their project, the
research group calculated the price of a BBB for most years between AD 1520 and 1810,
which can then be used to standardize prices in “AD 1551 tomines” based on the relative
increase in the basic cost of living. I also include a standardization based on the average
of the BBB value for the five years preceding the date of a document as a standardization
factor for the price actually written in the source. I use an average for the preceding years
because households would be more likely to purchase items beyond those required for
basic subsistence in “good” years with relatively low prices for basic staples. Prices from
documents lacking a precise date are standardized using the average BBB value for the
original author’s best estimate of the range of publication dates.
Calculation of Value Ratios
When the historic prices are standardized to account for the two issues raised in
the previous section, price ratios can be developed for the various categories of goods
recovered archaeologically at Calixtlahuaca. I include four categories of artifacts:
ceramics, lithics, ground stone, and rare copper or jewelry items.
I use one whole plain bowl as the basic unit of value for establishing subsequent
value ratios among artifact types. Following the logic used to create the CC index, plain
265
bowls are likely to have a stable value over time. From this baseline, most local ceramics
are assigned value up to five times that of a simple bowl, and imported ceramics are
given values twice that of locally produced pieces. For other artifact classes, four
obsidian blades or four formal tools are considered equal to one plain bowl. Small ground
stone artifacts are weighted as equal to five bowls and large ground stone items are
weighted as being equal to ten bowls. Copper and jewelry items are weighted as equaling
30 bowls. Fragmentary ceramic and lithic items are summed into whole item equivalents,
while ground stone and rare items are treated as whole items. These ratios are based on
the prices given in documentary sources, and the specific rationale for the values assigned
for each category is discussed in more detail under the following subheadings.
The total value of a household assemblage can potentially be calculated and
standardized several ways, based on either excavated volume or a ceramic based-factor.
Because of the formation process based variation in artifact density at Calixtlahuaca, I
have chosen to standardize value calculations per ceramic vessel equivalent (total rim
sherds/average rim arc for the component). This results in a numerical value that can be
considered an average total value per vessel for each household component. It is
calculated as follows:
𝐴𝐶 ∗ 𝐴𝐶𝑣𝑎𝑙𝑢𝑒
)
𝑖𝑛𝑑𝑒𝑥 = 𝐴𝑉𝑉 + ∑ (
𝐶𝑒𝑟𝑉𝑒𝑠
𝑡𝑦𝑝𝑒=1
Where: AVV=Average Ceramic Vessel Value; AC=artifact count for any other artifact
class; ACvalue=value per item, in “plain bowl” units; CerVes=number of ceramic vessel
equivalents; type=uniform value artifact class or subclass.
266
Scoring Specific Value Categories
In this section, I discuss the specific reasoning behind each of the relative value
scores for different artifact classes presented above. I use two general cross-cutting
principles to address the variation in historical sources. First, I consider the earlier
sources more likely to represent prehispanic value ratios. Second, I consider sources that
contribute values for more items to generally be more important, since these can be
compared in multiple dimensions. While I am aware that there is a substantial subjective
component in setting these value ratios, I feel that I have reasonably represented the value
ratios present in the original historical documents.
Ceramics
Ceramics are the most variable class of artifacts included in this analysis and have
the largest set of historical values. In order to structure this variability, I focus on two
dimensions. First, what are the primary causes of variation in value of locally produced
ceramics? I answer this question using data from the Codex of the Potters of Cuauhtitlán.
Second, to what degree does non-local origin (or style) increase value? I answer this
question using a wider range of data sources, with a focus on specific documents that
included values for both local and non-local ceramic vessels.
Local Ceramics. As noted previously, the Codex of the Potters of Cuauhtitlán is
highly useful for archaeological studies of value because it provides a set of prices for
different ceramic types. The prices are contemporaneous, and the pieces should not have
267
any variation due to differential transport costs. As a result, the document can be used to
establish both the range of values present in a ceramic assemblage – is the most
expensive pot worth twice or ten times the cheapest one – and potentially, particular
attributes – such as decoration - that might increase the cost of an item.
The vessels shown in the document fall within a relatively narrow range of prices
(Table 6.7). With one exception on each end of the scale, the pots cost between .18 and 1
tomin each, with the most common value being .5 tomin for a vessel. The single
exception to this range is a line of very large jars (“tinajas”), which cost either 3 tomines
(according to the gloss) or 4 tomines each (according the images). This suggests that the
locally produced vessels within an archaeological assemblage are likely to have a similar
spread of values, with most vessels falling within five times the value of the simplest
item, and a few high outliers. These values are, on average, on the lower end of the range
of prices in the other sources examined, likely because higher value ceramics were more
likely to be specifically mentioned in wills.
268
Doc. Vessel
Order 1 form
1
1
40
1
41
1
6
2
14
3
22
3
25
3
28
3
29
3
32
4
13
5
4
6
5
6
10
7
11
7
9
8
19
10
18
11
26
11
38
12
39
12
42
13
43
13
55
14
56
14
7
15
8
15
23
15
24
15
35
16
46
16
20
17
21
17
2
18
3
18
Name
molcajete
jarros
jarro
jarro
alcarraza
alcarraza
alcarraza
alcarraza
alcarraza
Price given
1 tomin for 4
One small coin for 12
20 cacao for 23
3 tomines for 6
.5 tomines for 1
4 tomines for 21
.5 tomin for 1
2 tomines for 8 (6) 4
1 tomin for 4
2 tomines for 4 (6)
.5 tomines for 1
4 tomines for 22(21)
2 tomines for 22
2 tomines for 6 (8)
2 tomines for 9
2 tomines for 6 (8)
One small coin for 1
.5 tomines for 2
1 tomin for 3 (4)
1 tomin for 2
2 tomines for 4
1 tomin for 2
6 granos for 1
3 tomines for 6
1.5 tomines for 3
2 tomines for 4
2 tomines for 4
2 tomines for 4
1 tomin for 2
1 tomin for 1
1 tomin for 2 (1)
2 tomines for 5
1.5 tomines for 5
2 tomines for 4
1 tomin for 2
Tomines
per pot
0.25
0.25
.75?
0.5
0.5
0.19
0.5
.25 (.33)
0.25
.5 (.33)
0.5
.18(.19)
0.09
.33(.25)
0.22
.33(.25)
0.25
0.25
.33 (.25)
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
1
.5 (1)
0.4
0.3
0.5
0.5
Table 6.7 Ceramic vessel forms and prices in the Codex of the Potters of Cuauhtitlan
(Cuauhtitlan 1568). Vessel form numbers correspond to Figure 6.3.
269
Doc. Vessel Name
16
18
30
18
31
18 jarros como picheles
36
18 jarros como picheles
37
18
47
19
17
20
49
21 jarros que le dimos la
hechura como negritos
50
21
57
21
58
21 jarros
59
21
60
21 jarros
44
22
45
22 jarros como caracoles
33
23
34
23 jarros
53
23
54
23 jarro
12
24
51
25 tinaja
52
25
27
11 &3 jarros
48 16 & 19
15
1
9 &10
Price given
2 tomines for 4
3 tomines for 6
1.5 tomines for 3
1 tomin for 2
2 tomines, 6 granos for 5
1 tomin for 2
1 tomin for 2
1 peso, 2 tomines, and 6
granos for 21 (20)
2 tomines, 6 granos for 5
1 tomin for 2
1 tomin for 2
2 small coins for 1
3 tomines for 6
2 tomines for 6
1 tomin for 3
3 tomines for 6
3 tomines for 6
1 tomin for 2
.5 tomin for 1
1 small coin for 1
6 tomines for 2 (1.5)
6 tomines for 2
1 tomin for 4
1 tomin for 3
Tomines
0.5
0.5
0.5
0.5
0.5
0.5
0.5
.5 (.53)
0.5
0.5
0.5
0.5
0.5
0.33
0.33
0.5
0.5
0.5
0.5
0.25
3 (4)
3
0.25
0.33
2 tomines for 8
0.25
Left to right within each block of the document, blocks read top to bottom
2
The small coin is never named in the glosses. Context suggests that it equals
1/4 tomin.
3
Contextually, six cacao are used as equivalent to half a tomin. A 100
cacao/peso rate would equal 6.5 cacao per half tomin, so the author may
simply be ignoring the resulting fraction of a cacao.
4
Prices in parentheses note discrepencies between pictoral images and
Spanish glosses and/or edits to the pictoral images themselves
Table 6.7 (continued) Ceramic vessel forms and prices in the Codex of the Potters of
Cuauhtitlan (Cuauhtitlan 1568). Vessel form numbers correspond to Figure 6.3.
270
The document shows a range of vessel forms (Figure 6.3). It is uncertain exactly
how accurately the scribe was depicting the vessels under consideration. On one hand,
there is a wide range of variation in forms, and variation within general form classes (e.g.
the same body shape, with and without handles), demonstrating that the scribe was not
just repeating a couple of stock images. On the other hand, in at least one case, it appears
that the scribe was not exactly clear on what a vessel looked like – the gloss reads “jarros
como caracoles”, and the accompanying image shows rows of shells. In either case, the
vessels include both open (bowl) and restricted (jar) forms, of both Aztec and European
derivation. Many of the forms correspond to archaeologically recovered colonial
Redware vessels (e.g. Charlton 1996; Charlton, et al. 1995). Prices can be calculated for
both a number of general variables (type of form, size, decoration, name in gloss), and
specific attributes (presence/absence of handles, lids, bases).
Most of the vessels can be assigned to one of three general form categories –
bowls (open vessels), jars (closed vessels), and pouring vessels with spouts. The majority
of the assignments here are obvious, but a few require more explanation. Barlow (1951)
previously identified #24 as a larger denomination coin. Based on the position of the item
within the arrangement of the codex, and the fact that it is not colored yellow like the
other coins, I consider it a vessel, probably a plate. It is treated as a bowl/open form for
analytical purposes. Form 23 may be another variety of anthropomorphic jar, similar to
northern European bellarmine jars. It is treated as a jar for the purposes of this
comparison. Vessel forms 6 and 22 are not included in the comparison of vessel forms,
because they could not be confidently assigned to a form category. Bowl forms are
generally the least expensive, with prices between .19 and .5 tomines per piece; over half
271
of the bowls cost less than .5 tomines. Jars are slightly more valuable on average, ranging
from .22-.5 tomin, (with the exception of forms 16 and 25), with the majority of the
examples of the form costing .5 tomines. Spouted vessels uniformly cost one .5 tomines
each. This comparison suggests that vessel size may be more important than vessel form
in determining price. Jars and spouted vessels are often similarly sized, while bowls are,
on average, somewhat smaller. This position is supported by the fact that form 25 is both
the largest image depicted, and the most expensive.
Figure 6.3 Vessel forms depicted in the Codex of the Potters of Cuauhtitlan (Cuauhtitlan
1568)
Vessels are decorated in two ways – painting and molding. There are four vessel
types that show clear evidence of painted decoration: 1, 16, 24, and 25. The prices for
272
painted bowls or plates are about average for open form vessels. However, there are no
bowls without some form or decoration (scalloping or painting), so the value of a truly
unornamented bowl may be lower than the range presented here. In contrast, the painted
jars are not only the most expensive jars, but also the only two vessels to cost more than
.5 tomines each. For form 25, the tinaja, the higher price may be in part a function of the
vessel’s larger size. However, form 16 is twice as expensive as form 14, which is an
undecorated version of the same shape.
Molded decoration takes two forms: simple scalloping, and more complex
designs. Scalloping does not appear to significantly increase the price of a vessel. There
are three pairs of simple and scalloped versions of the same basic vessel shape – 7/10,
8/11, and 15/17. In the first two, the scalloped variant is midway within the range of
prices for the plain variant. In the third case, the scalloped variant is slightly more
expensive. The vessel forms with more complex molding, 21, 22, and 23, are priced
comparably to jars and spouted vessels in general, though all of them fall in the upper
half of the range for these vessel types.
There are also two minor traits that can be tested for their effect on the price of
the vessels – handles, and bases. In both cases, there are pairs of vessels with and without
the trait in question. For handles, these are 10/11 and 14/17. In the first case handles do
not affect the price, while in the second, the version with handles is actually less
expensive. There are three states for bases – none, a simple pedestal, and a saucer-type
base. The second two may co-occur. The pairs are 7/8, 9/10/11, and 12/13. The presence
or absence of a base does not affect the price of the vessel. These minor traits suggest that
small additions to a vessel, such as handles, or a base, do not affect the value of the piece.
273
In summary, the two most important variables for determining the price of a
vessel in the Codex of the Potters of Cuahutitlan are size and decoration. This
corresponds well to most “common sense” or “rule of thumb” methods that
archaeologists use to assign relative values for ceramics. Small variations are unlikely to
change the value of a piece, suggesting that many archaeological production step indices
may create more gradations of value than the producers originally perceived in an
assemblage. Based on the general premise that ceramic vessels range within one and five
times the value of the cheapest piece, each local ceramic vessel type at Calixtlahuaca was
scored as worth one (small and plain), three (large and plain, or small and decorated), or
five (large and decorated) times the value of a plain bowl. A few extremely large types,
such as temple braziers, were scored as being worth ten times the value of a plain bowl.
Imported Ceramics. The second major question in assigning value to ceramics is
to what extent imported ceramics were considered more valuable than locally produced
items. There are three documents with ceramics that can be inferred to be both local and
imported: the undated testament from Culhuacan, and the 1724 and 1740 wills from the
Soconusco (Table 6.6). In the first case, the imported vessel is worth four times the value
of the only other unbroken ceramic vessel listed. In the second case, the presumably
imported (porcelain) vessel is listed at precisely the same value as the other two vessels,
though the fact that they represent three different vessel forms of varying sizes makes this
difficult to interpret. In the third case, the majority of the ceramic values are for sets of
cups, leaving it unclear whether these sets included the same number of items. However,
the porcelain cups are listed with higher values (.3-5 times) than the set of majolica cups.
274
The majolica may or may not have been produced in the immediate area (In the 17th
century, majolica production was concentrated in a number of major cities in Latin
America), but in either case would have been substantially more local than the Asian
porcelain. Both the majolica and the less expensive porcelain have similar values to the
remaining piece, a local water jar, though the latter was likely a large vessel. This
suggests that there should be some overlap between the prices of larger/more complex
local vessels and the smaller/simpler end of the range of imported vessels.
The combination of the variation in local values and the differences between local
and imported pieces can be used to generate a general set of value ratios for ceramics. For
the purposes of this analysis, I assigned non-local ceramic types a value twice that of a
locally produced pieces of the same general size and decoration. This is approximately
the halfway point in the variation between local and imported pieces seen in the historical
data, and does create a range of overlap between the values of local and imported pieces.
Once relative values were assigned to all local and non-local ceramic types at
Calixtlahuaca, I calculated the average value per rim sherd for each household
component (Table 6.8). This procedure is similar to that used in calculating the average
number of ceramic production steps in the first half of this chapter. I also calculated the
number of ceramic vessel equivalents at this time. This measure was calculated by
multiplying the total number of rim sherds in the component by the average fraction of
the vessel rim arc for that component, based on the attribute sample for the component.
The use of vessel equivalents rather than rim sherds standardizes for differences in sherd
size among household components, as well providing an easier point of comparison with
the whole vessels described in the documentary sources.
275
276
Table 6.8 Excavated ceramic assemblage value calculations, based on historic price
ratios. DS-2 sample.
Total Total,
N. Local Rims
small, small large, large, XL,
Rim nonUnit Phase Sherds erod. Unkn. plain , dec. plain dec. plain
5 10
3
3
1
Weighting
38
434 105 162 109 114
539
307 Dongu
315 Dongu 1,670 1,219 451 267 399 409 122
42
47 116 130 117
412
459
316 Dongu
1
624 549 116 281 111 100
320 Dongu 1,173
323 Dongu 2,454 2,220 234 452 954 557 192
5
19
16
26
0
77
77
324 Dongu
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
829
2,356
383
941
2,557
229
471
2,068
298
692
2,055
200
358
288
85
249
502
29
141
566
86
190
427
23
157
782
111
206
622
63
89
464
65
183
668
72
65
191
24
70
240
20
307
309
316
317
324
Yata
Yata
Yata
Yata
Yata
1,141
426
1,218
1,274
355
664
230
1,079
685
257
477
196
139
589
98
85
22
209
55
67
236
63
328
204
63
101
34
345
141
82
100
41
90
126
30
327
Yata
147
86
61
14
34
26
12
1
1
1
3
N. Imported Rims
small, small, large, large,
plain dec. plain dec.
10
6
6
2
0
3
6
2
0
0
19
3
1
0
6
0
0
1
12
2
0
1
58
6
0
1
9
1
Average
# Vessel
Known Average EquivaValue 1 Rim Arc
lents
2.49
2.81
2.70
3.02
2.84
2.83
6.78
6.09
5.32
4.59
5.76
10.42
36.53
101.73
24.44
53.80
141.27
8.02
0
2
2
2
5
1
17
49
1
26
71
20
2
14
9
14
21
1
0
0
0
0
0
0
2.80
2.73
2.68
2.83
2.95
3.28
4.90
5.71
4.40
6.61
5.32
3.66
40.65
134.54
16.85
62.21
136.04
8.39
7
2
6
8
2
71
47
74
97
8
63
21
27
49
5
0
0
0
2
0
3.65
4.04
3.05
3.89
2.86
5.16
4.65
5.12
5.77
4.94
58.91
19.81
62.34
73.52
17.55
0
0
0
0
2.95
5.132
7.54
1
Average Known Value = ∑ (N sherds in category * weighting factor)/Total non-eroded sherds
2
327 Value is Phase 6 average, due to lack of attribute data
Lithics
I encountered only one documentary value for lithic artifacts, the 20 obsidian
blades/real price given in Geronimo de Mendieta’s Historia Eclisiastica Indiana (1945
[ca. 1571-1596]). Given that this is a summary work by a Spaniard, rather than a nativeperspective price and/or a price dating to a specific year, this is not as reliable of a source
of evidence as those used for other artifact classes. It is also contextually unclear whether
the figure refers to whole blades or some fraction thereof. Most lithicists interpret the
phrase as referring to whole blades (Hirth 2013), and I do the same for the purposes of
this analysis.
This leaves a problem of establishing equivalencies between the lithics actually
recovered during excavations, which are fragmented into pieces far smaller than an entire
blade, and include both blade and non-blade artifacts. The average blade (pressure or
percussion) fragment at Calixtlahuaca measures 1.9-2.1cm, making it about a fourth to a
fifth of the 8-11cm core length commonly used in Mesoamerica (Parry 2002). Due to
formation processes, most non-blade artifacts are similarly sized. As a result, I treat most
of the recovered lithics as having the value of 1/5th of a blade. The exception to this is
formal tools, which would have required more work, and are generally sufficiently
unique that each recovered fragment represents a unique original piece. As a result,
formal tool fragments are assigned a value equal to one whole blade. For purposes of
weighting lithics relative to other artifact classes, twenty complete blades (or the
equivalent thereof) are treated as equal to five times the value of a plain ceramic bowl, or
a four blade to one bowl ratio.
277
As a note, this procedure for assigning lithic values works at Calixtlahuaca
precisely because there is almost no evidence for lithic production at the site (See Chapter
5). As a result, the majority of lithics recovered in household contexts can be considered
as evidence of consumption, not production.
Ground Stone
The documentary sources provided information on two types of ground stone
items, manos and metates. Based on information from two wills with prices for both
items, a metate was generally worth twice as much as a mano. The domestic excavations
at Calixtlahuaca recovered a much wider range of ground stone items. For the purposes
of this analysis, the excavated ground stone was coded as “large” (metates, molcajetes,
supports from one of the two preceding forms, plaster smoothers, and anvils), and
“small” (all other ground stone items, such as manos, pestles, bark beaters, etc). Pebble
tools were excluded from the analysis on the basis that they were likely opportunistic
tools, rather than specially purchased items. Artifacts from each excavation unit had been
informally checked for reconstructible pieces both during their initial cataloging and
during later analysis, meaning that there is little probability that any of the pieces listed
under different catalog numbers were originally from the same object. As a result, each
ground stone fragment is valued as if it were a whole item.
There is a high degree of variability in the standardized prices for manos and
metates, but when each is compared to relatively simple ceramics from the same region
and time period, this variation is less than it originally appears. I use pieces with the
278
earliest dates to approximate a value of five simple bowls equaling a small piece of
ground stone and ten equaling a large piece of ground stone.
Rare Items
Small, rare, high value items are the hardest category to estimate values for. The
documentary sources present a very wide range of items and associated values for items
in this category. This category is probably the most likely to have been influenced by
European contact, which dramatically changed metal production technologies in the New
World. The items recovered archaeologically at Calixtlahuaca, which include both bronze
and semi-precious stone artifacts, likely correspond to the simpler end of the spectrum
represented in the historical documents. None of the excavated artifacts is particularly
elaborate and none was recovered in a context that would suggest that it was part of a
more complex, multicomponent piece, such as an inlaid piece of jewelry.
As a result, I weight the archaeologically recovered items as being worth 30
simple bowls each. This is subjective decision, placing the archaeological value in the
middle to upper portion of the less expensive items on the list of historical values.
Master Wealth Index: Results
The Master Wealth Index, which calculated the per-vessel value of ceramics,
obsidian, ground stone, and bronze/jewelry items, shows a modest but statistically
significant (at the .90 level) increase in average household wealth between the Dongu and
Ninupi phases, followed by statistically equivalent values between the Ninupi and Yata
279
phases (Table 6.9, Figure 6.4). When the index is broken down by artifact type, there is
variation in how the individual artifact types pattern over time, which is not unexpected
in light of the variation in the individual analyses in the first half of this chapter.
Paralleling the results of the individual artifact type indices (though with slightly different
numbers due to the use of whole vessel equivalents, rather than number of rim sherds as
the standardizing factor), obsidian and copper/jewelry rise between the Dongu and
Ninupi phases before dropping during the Yata phase, while ground stone features the
opposite pattern, dropping and then rising, and the average ceramic value rises from each
phase to the next. The overall pattern, of modestly increasing household wealth values
prior to Aztec rule, followed by a cessation of further growth while part of the empire, is
consistent with a modest degree of Aztec imperial control over local affairs.
The overall range of within-phase variation in wealth is generally low, but does
consistently increase over time. Both the absolute range of within-phase variation and the
coefficient of variation show an increase from each phase to the next. However, with the
exception of one outlier household, 324-Ph2 (discussed in more detail below), the
distribution of household scores for each phase is similar. Each phase has one component
falling between one and two standard deviations above the mean, and one falling between
one and two standard deviations below the mean. All other components fall within one
standard deviation of the phase mean. As a result, while the amount of variation among
households increases over time, the patterning of that variation is remaining relatively
similar, as a continuous spread of values. The low degree of interhousehold variation is
generally consistent with relatively collective forms of governance, both before and
during Aztec rule.
280
Figure 6.4 Master Wealth Index results by phase, with contribution of each artifact class
281
N.
Unit Phase Vessels
Weighting Factor
307 Dongu
36.53
315 Dongu 101.73
316 Dongu
24.44
320 Dongu
53.80
323 Dongu 141.27
324 Dongu
8.02
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
Lithic
Frags/ Lithic GS, GS, Rare
100 Tools Small Large Items
5 5
5
10
30
2.7 9
1
2
1
1.03 6
8
0
0
0.29 3
4
2
0
2.55 9
4
4
0
7.34 27
10
2
0
0.54 1
10
3
0
Mean Lithic
Vessel Value/
Value 1 Vessel
Ground
Rare
Stone
Item
Value/ Value/
Vessel Vessel
Master
Wealth
Index
Total
2.49
2.81
2.70
3.02
2.84
2.83
1.60
0.35
0.67
1.07
1.22
0.96
0.68
0.39
1.64
1.12
0.50
9.97
0.82
0.00
0.00
0.00
0.00
0.00
5.59
3.55
5.01
5.21
4.55
13.76
40.65
134.54
16.85
62.21
136.04
8.39
3.35
5.69
0.84
4.63
2.06
1.02
16
33
3
29
15
1
5
14
2
5
9
1
2
3
0
1
3
0
0
16
1
2
0
0
2.80
2.73
2.68
2.83
2.95
3.28
2.38
1.44
1.14
2.70
0.63
1.20
1.11
0.74
0.59
0.56
0.55
0.60
0.00
3.57
1.78
0.96
0.00
0.00
6.29
8.48
6.19
7.06
4.13
5.08
58.91
19.81
62.34
73.52
17.55
7.54
2.82
0.67
0.84
0.75
1.74
0.38
11
4
6
2
2
0
11
5
7
13
0
0
3
1
4
3
0
0
2
1
1
0
1
0
3.65
4.04
3.05
3.89
2.86
2.95
1.17
1.18
0.55
0.19
1.07
0.25
1.44
1.77
1.20
1.29
0.00
0.00
1.02
1.51
0.48
0.00
1.71
0.00
7.28
8.50
5.29
5.37
5.63
3.21
Dongu,
Ninupi,
Yata, Mean
2.78
2.88
3.41
0.98 .86(2.38) 2
1.58
0.69
0.73
0.95
0.14 4.78(6.28) 2
1.05
6.20
0.79
5.88
Dongu, StD
Ninupi, StD
Yata, StD
0.18
0.22
0.52
0.44 .51(3.75) 2
0.80
0.21
0.46
0.76
0.34 .79(3.73) 2
1.43
1.51
0.74
1.83
Dongu, CoV
Ninupi, CoV
Yata, CoV
0.06
0.08
0.15
0.45
0.50
0.63
2.45
1.36
0.94
1
See Table 6.8 for calculation of this value
2
Excluding (Including) component 324-Ph2
0.59
0.31
0.80
0.17
0.24
0.31
Table 6.9 Master wealth index values for ceramics, lithics, ground stone, and rare items,
by excavated component, based on historical price ratios
282
This analysis also identified one extreme outlier, household component 324-Ph2,
which has a Master Wealth score approximately fifty percent higher than that of any
other component during any phase. The score is approximately three times the average
for the remaining components dating to this phase. This component may be the result of
specialized ritual activities. The component consists exclusively of material from a single
pit sealed under a stone pavement, only half of which was excavated due to time
constraints. As a result, the component is much less of an “average” than those which
include material from more mixed contexts and/or multiple contexts. The ceramics from
this excavation are a fairly standard mix of types and forms, but the component has the
largest sherd size of any household component at the site (both based on average sherd
weight and in the attribute sample, rim arc) and an unusually high frequency of partially
reconstructable vessels, including a complete crude unfinished bowl. The high score for
this component is primarily driven by the amount of ground stone (3 pieces of large items
and 10 of small items) recovered in an otherwise small excavation. As a result, this
component is excluded from both interhousehold correlations among artifact types and
the calculations of general temporal trends in the Master Wealth index.
The individual artifact types included in the Master Wealth index (Ceramic,
Lithic, Ground Stone, and Rare Items) show poor correlation with each other at the
individual household level, though the latter three are significantly correlated with the
household score on the Master Index (Table 6.10). Of the six possible correlations
between the four artifact classes, only ceramics and ground stone are significantly
(positively) correlated at the p=.10 level. Two variable pairs are very weakly correlated:
Ceramics/Lithics (negative) and Rare Items/Lithics (positive). The household component
283
values for the remaining three variable pairs (Ceramics/Rare Items, GS/Lithics, Rare
Items/Ground Stone) are essentially random relative to each other, which is not surprising
given the small sample sizes for ground stone and rare items. The findings of the lack of
correlation between many artifact types supports the use of a multi-factor wealth index in
order to compensate for both recovery bias in rare artifact classes and variation in
consumer choices by ancient people.
Pearson's r
Master Ceramic
Master
1.00
.
Ceramic
0.31
1.00
Lithic
0.56
-0.26
GS
0.48
0.54
Rare
0.75
-0.07
Lithic
.
.
1.00
0.01
0.28
GS
.
.
.
1.00
-0.05
Rare
.
.
.
.
1.00
Chi-squared p-value
Master Ceramic
Master
1.00
.
Ceramic
0.23
1.00
Lithic
0.02
0.31
GS
0.05
0.03
Rare
0.00
0.79
Lithic
.
.
1.00
0.96
0.28
GS
.
.
.
1.00
0.86
Rare
.
.
.
.
1.00
Table 6.10 Household component level Pearson's r correlations and Chi-squared
significances for the Master Index and its component factors (Excludes 324-Ph2)
Conclusions
This chapter analyzed changes in household wealth over time at Calixtlahuaca. It
included analyses based individual artifact classes (ceramics, obsidian, and rare items)
284
and a master index which weighted various artifact classes (ceramics, obsidian, ground
stone, and rare items) based on their relative values in early colonial historical
documents. When contextualized within the broader range of findings at the site, the
analyses in this chapter provide information on household wealth, and more indirectly
quality of life, over time. I then relate this to changes in rulership practices at this site.
First, most measures analyzed in this chapter show increasing wealth between the
Dongu and Ninupi phases, followed by a leveling out between the Ninupi and Yata
phases. Changes are unevenly distributed among different artifact classes, with primarily
non-local items (obsidian, rare items) paralleling the general pattern, while primarily
locally produced items either remain relatively unchanged (ceramics) or increase during
the final Yata phase (ground stone). The timing of the overall patterning shows that the
majority of the economic gains at the site occurred prior to its incorporation into Aztec
Empire. This suggests that these pre-Aztec economic gains were the result of increased
participation in the Central Mexican market system, and a local political organization that
supported widespread participation in this system. The more ambiguous later results
suggest that the residents of Calixtlahuaca maintained many of their earlier economic
gains under Aztec rule, but that further growth slowed or stopped. This may parallel the
lack of growth in trade with areas other than the Basin of Mexico during the Yata phase.
Moving beyond simple measures of wealth, I argue that the quality of life at
Calixtlahuaca improved between the Dongu and Ninupi phases, followed by a decrease
during the Yata phase, when the site was under Aztec rule. This conclusion draws on
several factors in addition to the average wealth trends. First, prior to the Aztec conquest
of the site, the quantity and diversity of imported goods at the site was steadily
285
increasing, likely leading to a perception by households that they were getting more
“bundles”. Following the Aztec conquest, the loss of trading partners likely created a
perception of resource bundles that were known, but now unavailable. This would have
been particularly notable for obsidian, where the evidence for bipolar “recycling” goes up
sharply during the Yata phase. Second, for those crafts that increased over time,
especially textile production, the larger increase in production occurred between the
Ninupi and Yata phases, suggesting that the increase in production did not result in
increased household wealth, but rather, that increased production was necessary to
maintain the status quo under Aztec rule. Finally, the amount of variation in access to
culturally relevant “bundles” also likely reached its highest level during the Yata phase,
leaving many households with the self-perception that they were even less relatively well
off than they may appear in a longer contextual view. This variation in access is visible
both in the less uniform access to ceramics from various imported sources discussed in
the previous chapter, as well as in the higher standard deviations and coefficients of
variation in the master wealth index during this time period.
The changes in the quality of life at the site provide evidence for the nature of
Aztec rule at Calixtlahuaca. The independent, pre-Aztec social organization of power at
the site was relatively collective. This position is supported by the steady economic
growth between the Dongu and Ninupi phases, as well as by the relatively even
distribution of changes in access to non-local goods and overall wealth among commoner
households. Subsequently, under Aztec rule, further economic growth stalled,
demonstrating that imperial policies and/or actions had sufficient force to penetrate and
influence local economic networks. This interruption of economic growth indicates that
286
Aztec rule, while still relatively indirect, should not be considered completely hands-off.
The increasing unevenness of commoner wealth distributions and the variability in
individual household’s sources of non-local goods under Aztec rule implies a reduction in
the collectiveness of the social organization of power, with households depending more
heavily on individual network connections for household provisioning.
287
CHAPTER 7
CHAPTER 7. “BY THEMSELVES THEY CELEBRATED THE FEAST DAY8”:
CULTURAL CHANGES IN RITUAL PRACTICES
The shift from local to imperial rule has consequences for commoner households
beyond the economic. Imperial rule potentially provides new avenues for negotiating
status, marking identity, and interacting with non-local people. However, the degree and
ubiquity of changes in cultural practices in provincial areas are also strongly influenced
by imperial rulership strategies. Cultural interaction under imperial rule can occur in
various ways. In the next two chapters, I look at changes in ritual practices (this chapter)
and foodways (Chapter 8) and how these relate to the directness and collectiveness of
local and Aztec rule at Calixtlahuaca.
As with the previous chapters addressing economic issues, I focus on identifying
the direction, magnitude, and intra-household variation in changes in cultural practices at
Calixtlahuaca over time. However, unlike the previous chapters, which were primarily
concerned with the distribution, production, and acquisition of physical objects, the
discussion of cultural practices requires a focus on the activities of ancient people. As a
result, I use a two-part strategy, looking at both the use of non-local objects and non-local
use patterns.
One potentially confounding factor in looking at cultural practices is whether such
changes were due to the immigration of new people into the area or due to the local
8
Florentine Codex: A General History of the Things of New Spain. Book 10:The People. Chapter 29,
Section “The Quaquata, The Matlatzinca, The Toloque”.
288
adoption of new cultural practices. At Calixtlahuaca, the original Middle Postclassic
population may have been partially replaced by Nahuatl populations from the Basin of
Mexico at various points during the site’s history. The next two chapters also evaluate the
probability of immigration as opposed to the local adoption of Aztec practices, based on
the appearance of Aztec traits in high- and low- visibility contexts. These two scenarios
have very different implications for the interpretation of the observed cultural and
economic shifts.
I find that ritual practices at Calixtlahuaca were both distinct from those seen in
other parts of Central Mexico and relatively homogenous among households at the site
during the Dongu and Ninupi phases. During the Yata phase, overall ritual practices shift
toward moderately more Aztec pattern. This is accompanied by an increase in
interhousehold diversity, with some households remaining strongly local, while others
conform to patterns expected of emulation and immigration. The overall pattern in
consistent with relatively collective local rule, paired with indirect and relatively noncollective imperial rulership.
Rulership and Cultural Change
Individuals mark and reinforce particular cultural identities under circumstances
in which doing so is beneficial to themselves (Laitin 1998). Demonstrating a common
cultural identity with another often results in increased trust and preferential interaction
with those sharing a common identity (Bowles and Gintis 2004; La Ferrara 2003). Both
the directness and collectiveness of rule carry implications for the degree of cultural
289
variability within the resulting state and the conditions under which shifting toward a
more imperial affiliation will be beneficial. In both cases, cultural changes may be the
targeted result of state policy, the indirect result of increased interaction between the core
and province, or the result of intermediate positions between these two degrees of state
intervention. The first position, state-enforced cultural change, is primarily top-down and
depends on the power and desire of the state to enforce such change. In contrast, the
second position, cultural change due to increased interaction, is a relatively bottom-up
process by provincial people adopting some aspects of foreign cultures for their own
reasons.
Relatively indirect rulership will generally result in few top-down cultural
changes to provincial areas. Under relatively indirect rule, an empire will lack the
penetration into local societies to enforce changes in cultural practices. In addition, it is
not in the state’s interest to promote beliefs or practices which would destabilize the local
rulers on which the state relies for local control. While there may be a limited increase in
interaction with the imperial core, overall interaction patterns are expected to follow their
preexisting trends.
In contrast, under more direct rule, empires have greater penetration into local
societies. This offers more opportunities for the state to successfully introduce or
suppress cultural practices in provincial areas (e.g.Hechter 2013:Chapter 4; Parker 2003).
Under more direct rule, provincial people also have more indirect interaction with the
imperial core, at a minimum due to the increased presence of state agents from the core
stationed in the provinces, such as bureaucrats, tax collectors, military garrisons, or statesponsored merchants. Such people bring their own cultural practices with them,
290
providing provincial people the opportunity to observe or participate in a wider range of
core-style practices than would generally be targeted by official state policy.
The collectivity of both local and imperial rule will also influence how provincial
culture changes under imperial rule. The literature on collective action in modern
societies generally predicts that higher group social homogeneity (but not necessarily
economic homogeneity) will improve collective action outcomes (Ostrom 2007:190).
More homogenous groups will have more shared norms concerning appropriate behavior
and sanctioning, leading to better group cohesion. As a result, groups with a strongly
collective pre-imperial organization should have more success resisting forcibly imposed
cultural changes. As far as voluntary adoption of new cultural traits, more collectively
organized groups should probably sanction individual cultural experimentation that
threatens group identity. If imperial practices are adopted, the adoption should occur
relatively evenly across the group in order to maintain internal coherence.
On the ruling side, relatively collective imperial rulership should provide
substantial motivation for local people to adopt imperial practices and identities. Under
this form of rulership provincial people have opportunities for participation in the
imperial system, provided that they conform to its norms. Relatively collective rulership
practices also involve large portions of the population in public events, such as religious
ceremonies and feasts (Carballo 2015), promoting widespread exposure to imperial-style
practices. In contrast, under less collective forms of local rule, personal connections are
an important means of social advancement. As a result, some individuals (or households,
or particular social subgroups) are likely to seek to develop and publically express ties to
the imperial state.
291
Relatively non-collective imperial rulership will focus on promoting a shared
culture among elites (Scheidel 2006). Any promotion of imperial values to the population
at large will likely to be linked toward promoting elite differences, such as cults of the
emperor as a divinity. If local elites adopt imperial practices, these may trickle down due
to the local emulation of local elites, but such adoption is likely to be uneven across the
population, due to differing levels of personal ties to local elites.
Aztec Policies Toward Acculturation
Aztec policies toward the acculturation of subject populations varied along two
lines: social status, and geographic distance. Aztec imperial policy included a suite of
actions which served to promote a common culture, heavily influenced by central
Mexican practices, among ruling elites within the empire (Berdan and Smith 1996). Such
actions included marriages between the ruling dynasties of the Triple Alliance and those
in provincial areas, ritual and feasting events in the imperial capital where attendance was
mandatory for provincial leaders, and requiring that children of provincial leaders live in
the imperial capital of Tenochtitlan. These practices built on pre-imperial traditions of
elite interaction in Central Mexico. As a result of both imperial policy and prior trends,
public architecture, including both elite residences and temple complexes, followed a
single relatively standardized architectural cannon across central Mexico by the Late
Postclassic (Evans 2006; Smith 2008). Aztec imperial actions vis-à-vis provincial elites
should be considered a network–oriented strategy.
In contrast to imperial actions serving to integrate elites across the empire, Aztec
imperial policies toward the acculturation of commoners are mostly absent outside of the
292
Basin of Mexico, and uneven within the Basin. According to the Relaciones
Geographicas, the worship of the Mexica patron god Huitzilopochtli (or close cognates)
had been adopted by some other communities in the Basin or which had migrated out of
the Basin relatively shortly prior to the Spanish conquest (Umberger 1996). However,
based on analyses of figurines in the Basin of Mexico, commoner domestic ritual had few
to no ties to imperial state religion (Brumfiel 1996; Klein and Victoria Lona 2009). Aztec
goods, including both items produced in the Basin of Mexico and locally produced
imitations, become progressively more common in provincial areas over the Late
Postclassic (Smith 1990). However, in areas with sufficient chronological control,
including Calixtlahuaca, this increase begins prior to the historic dates for the beginning
of Aztec rule of the province (Huster and Smith 2015; Smith 1987a). This strongly
suggests that the use of Aztec goods and/or associated cultural practices in provincial
areas was voluntary, rather than imposed by state policy.
Geographically, areas toward the center of the Aztec Empire generally show more
adoption of practices and objects characteristic of the Basin of Mexico (Smith 1990;
Umberger 1996). However, there are some regional differences in the degree to which
areas at similar geographic distances adopted Aztec practices. Additionally, in many
areas, the adoption of Aztec style objects and/or practices begins prior to the historical
date of Aztec rule for the region, strongly suggesting that their adoption was due to local,
bottom-up processes rather than imperial policy.
293
Causes of “Aztec-ization”: Travelling Pots, People or Ideas
Given that the appearance of Aztec-style items in other parts of Mesoamerica is
not likely a direct product of imperial rule, the mechanism for increasing “Aztec-ization”
bears further exploration. Archaeological explanations for the appearance of non-local
forms of material culture fall into three categories – the movement of objects, the
movement of people, and the movement of ideas.
Physical objects can move from one region to another through trade and gifting.
The movement of objects is usually the easiest explanation to evaluate, as sourcing
techniques can definitively establish whether an object travelled from one region to
another. Explaining why trade patterns changed is more complicated, and often
incorporates political or economic factors. As discussed in Chapter 4, Calixtlahuaca
shows a clear increase in goods imported from the Basin of Mexico over time. This may
simply be due to changing political circumstances, and an increased availability of goods
from the Basin as the Toluca Valley became more incorporated in the Late Postclassic
market network.
The second explanation, the movement of people, postulates that changes in
material culture are due to changes in the population producing and using the artifacts. In
this case, the immigrant population continues making goods from their homeland, or
using locally produced goods in ways more consistent with practices from their
homeland. At Calixtlahuaca, there is a historical suggestion of immigrants from the Basin
of Mexico, either under the umbrella of Tepanec control during the Ninupi (LPC-A)
phase, or under Aztec control during the Yata (LPC-B) phase (García Castro 1999).
294
The third explanation is that the idea of a particular type of material culture
travelled, causing the local production of a foreign-style object. The causes for the spread
of an idea are diverse. Most overtly, a technology may spread because it is useful and
relatively easily replicated in a new area (e.g. the popularity of log cabins, originally of
Fenno-Scandinavian origin, among settlers on the American frontier (Burmeister 2000)).
Functional explanations are unlikely to apply at Calixtlahuaca, as technologies were
broadly comparable across Postclassic highland Central Mexico. More subtly, an idea can
spread because it implies an affiliation with another group. This may lead to its spread
due to a direct desire to integrate oneself with a specific group for political gain, or a
more generalized desire to reference groups or places seen as cosmopolitan or powerful.
Different subgroups within a single population may choose to adopt foreign styles for
different reasons. Over time, such originally foreign styles may become integrated into
local traditions.
Migration
People move across the landscape in diverse ways, and the more permanent
and/or long distance of these are generally called migration. Migration has a long
conceptual history in archaeology, though it has only relatively recently moved from an
explanation in itself, to a process to be studied (Anthony 1990; Burmeister 2000; Rouse
1986). One useful conceptual model of migration is provided by Tilly (1978), who
categorizes migration in terms of the distance moved and the degree of social breakage
involved in the move. This model has been considered in archaeological terms by Smith
(2014). For pre-modern states, this model provides three useful forms of migration to
295
consider: local migration, chain migration, and career migration. A fourth form of
migration, circular migration, is considered too difficult to identify archaeologically at
the chronological resolution currently available in Postclassic Central Mexico. These
categories are useful heuristic concepts, but any real world case, whether archaeological
or modern, is likely to contain evidence for multiple forms of migration. Thus, the wellknown Oaxaca Barrio at Teotihuacan may have been founded as an example of career
migration (a large group moving at once), but saw ongoing chain migration from Oaxaca
after that.
The first type of migration under consideration is local migration, which involves
relatively low levels of both geographic distance and social rupture. It involves
circulation within preexisting social, economic, or marriage networks. This is a welldocumented phenomenon in medieval and early modern Europe (Osborne 1991). It is
rarely considered archaeologically, though this level of background mobility should
probably be considered the normal state of affairs in antiquity, rather than the exception
(Smith 2014). Because most local migration happens over relatively short distances and
within existing social networks/cultural areas, it is almost invisible archaeologically. It
does provide a reasonable interpretation for households where DNA or biodistance
analyses show a lack of continuous descent over multiple generations, but few
individuals with non-local isotopic signatures (e.g. Mata-Míguez, et al. 2012; Miller
2015). At Calixtlahuaca, this type of migration could be associated with the low degree of
household continuity over time; two thirds of households occupied in one phase are not
occupied in the subsequent phase. In addition, a locally mobile population would help
explain the survey pattern at the site, where distinct ceramic assemblages are highly
296
spatially intermixed (Novic 2015). Such a pattern would be consistent with families from
a variety of nearby communities moving in and out of the site on a regular basis, under
some sort of conditions that prevented immigrants from settling near others from their
prior home community. This type of migration is taken to be the default background state
of affairs at Calixtlahuaca, likely accounting for some of the inter-household variation in
cultural practices during all phases of the site’s history.
On a broader scale, chain migration involves the ongoing movement of
individuals or households from one community to another, across larger social and
geographic distances than what is observed in local migration. Movements are facilitated
by preexisting contacts between the source and destination communities, such as family
members, trading partners, or previous immigrants from the source community. This
scale of migration is likely to be visible bioarchaeologically as scattered individuals in a
population with non-local bone isotope values or other evidence for an origin outside of
the surrounding area. In Mesoamerica, Bullock Kreger’s (2010) dissertation findings on
burials from Cholula could be considered an example of this form of migration. She
found that 18-20% of the burials in her commoner sample had non-local strontium or
oxygen values, despite not being distinguished by unusual burial practices. Because this
type of migration also crosses greater social boundaries, it may be visible in domestic
material culture. In the US Southwest, Clark’s (Clark 2001; Clark, et al. 2009) work on
immigration into the Hohokam San Pedro Valley provides cases where a strong argument
can be made for immigration by households, which then integrated into particular
destination communities. While this scale of immigration is not likely to explain an initial
297
influx of settlers at Calixtlahuaca, it could play a role in the subsequent maintenance of
immigrant households at the site.
The third category, career migration, is characterized by a high degree of both
social rupture and geographic distance. While originally used to describe individuals who
made definitive moves in response to opportunities for new positions in government,
military, or trade organizations (Tilly 1978), Smith (2014) expands the category to
include “most cases of big-groups-of-people-moving-from-one-place-to-another.” Under
this expanded definition, most archaeologically documented cases of migration fall into
this category. This is especially true in cases of site unit intrusions, where an entire site or
neighborhood within a site features a broad range of traits from a single non-local origin.
In Mesoamerica, the best known archaeological examples of this phenomenon are the
Oaxaca and Gulf Coast barrios at the site of Teotihuacan (Cowgill 2008). Bone chemistry
work at these neighborhoods shows that 29-40% of the occupants had either been born
elsewhere or lived elsewhere for a significant portion of their lives (Price, et al. 2010;
White, et al. 2004). However, the archaeology of the Pipil migrations into Central
America (Bove, et al. 2012) and of the Malpais sites of west Mexico also places them
firmly into this category of large-groups-of-people-moving.
Career migration can also occur over more moderate geographic and cultural
distances. At Early Postclassic Xaltocan in the northern Basin of Mexico, immigrants,
likely from the southern Basin of Mexico, settled around the periphery of the previously
established site center (De Lucia and Overholtzer 2014). Households in the periphery
have different house layouts, construction techniques, and burial practices from those
characteristic of households in the longer-occupied site core. At the archaeologically298
visible temporal resolutions, it is difficult to determine whether Xaltocan and similar
cases represent chain migration (an ongoing influx of people over time) or career
migration (a single, relatively cohesive movement of people). If Aztec immigrants are
present at Calixtlahuaca during a single phase, it will most likely be impossible to
determine whether they arrived via chain migration building on previous trade-based ties,
or as a result of imperial-sponsored resettlement.
An additional useful distinction in looking at migration is the difference in the
size of the migrant population relative to the preexisting local population (Rouse 1986).
In cases where migrants move in relatively small units and/or constitute a minority
relative to the local population, they will face a choice between assimilating into the local
culture or explicitly marking and maintaining their differences. In contrast, in cases
where an immigrant population either moves into an uninhabited area or is large enough
to swamp the preexisting population, tensions in local/immigrant identity are not as
significant of a factor and immigrants may innovate with their identity in ways tied to
neither preexisting local populations nor their homeland. All cases of local and chain
migration fall into the former category. However, cases of career migration straddle the
two categories, and should be considered on a case by case basis. At Calixtlahuaca, it is
unlikely that Aztec immigrants arrived in sufficient numbers to swamp the considerable
preexisting local population. This means that they could have muted their cultural
differences in the face being a disliked foreign minority, or they could have been
confident enough in their position as members of the conquering group to express a
Mexica identity openly.
299
This category of career migration also plays a large role in the ethnohistory of
Postclassic Mesoamerican peoples. In native histories, large groups, which could be
glossed as ethnic groups under a loose definition of ethnicity, are often recorded as
moving through dozens of cities over multiple centuries. While the arrival of the Mexica
to Tenochtitlan may be the best known of these stories, it is only one example of a
widespread genre. Due to this written evidence, migration played a large role in early
culture-historical reconstructions of the Postclassic Basin of Mexico and surrounding
areas, with changes in material culture linked to the arrival of particular groups at
particular sites. With the advent of chronometric dating showing that such correlations
were not likely to be valid, and an increased skepticism in the historical validity of the
mythic histories (Smith 1984), migration was set aside as a topic in inquiry in Central
Mexico. More recent work (Beekman and Christensen 2003; Cowgill 2013) has returned
to the subject, using multiple lines of linguistic, biological, and archaeological evidence
to argue for large-scale population movements in Mesoamerica during the Epiclassic and
Early Postclassic.
There have also been a limited number of artifact-based studies of Postclassic
migration in the context of the Aztec Empire. These address longer-distance migration,
usually in cases where there was a historically documented movement of people from the
Basin of Mexico to a peripheral portion of the empire. The resulting settlements show
varying degrees of distinction from the surrounding regions. Near the Tarascan border,
the Aztec fortress of Oztuma and surrounding Chontal sites can be clearly distinguished
on the basis of decorated ceramic types and obsidian from particular sources (Silverstein
2001). In contrast, efforts to identify the material presence of the Aztec garrison at
300
Coixtlahuaca, in Oaxaca, found only trace evidence for an Aztec presence at the site
(Kowalewski, et al. 2010), obsidian source frequencies and the majority of recovered
ceramics matched those of other sites in the Mixteca Alta.
Local Acculturation
In addition to the appearance of foreign styles due to the movement of people,
foreign styles or practices may also appear due to their adoption by local peoples. Local
adoption of foreign practices or styles of material culture may occur for a variety of
reasons, variously referred to as emulation, appropriation, and assimilation (Stark and
Chance 2012). These may generally be grouped into two axes of variation, based on the
status of the people adopting non-local styles, and the degree of fidelity to the original
practice.
Non-local objects and/or practices are often more widely adopted by elite
members of a society than by commoners, as a means of signaling status (Neff 2014). At
the most basic level, this appears as higher levels of imported goods in elite households
(Smith 1987b). Higher social classes seek to distinguish themselves via unique markers,
which are gradually adopted by lower classes, at which point they begin to fall out of use
by the original higher class users because they no longer reliably fulfill their original
function as exclusive status markers (Bourdieu 1984). (For the relationship between taste
and status from modern Mexico, see Bustamante and Garcia (2015).) Non-local items can
often fill this role, as elites often have more geographically extensive social networks due
to marriages and other alliances. This may take the form of widespread aspects of elite
culture that cross-cut ethnic or other regional identities, due to marriage ties or
301
participation in “high culture” used to mark elite status. Archaeological examples of this
phenomenon include the use of specialized drinking sets across much of the Greek
Mediterranean (Hodos 2009) and the widespread use of Harappan-style luxury items by
elites in areas outside the Harappan area proper (Chase, et al. 2014). In Mesoamerica, this
can be seen in the uniform appearance of palaces across much of Postclassic Central
Mexico (Smith 2008:115-119), a uniformity which cross-cuts variation in the size, layout,
and construction methods of commoner houses in the same regions. Elites may also adopt
non-local material culture due to its association with politically or ideologically powerful
locations, such as Maya rulers’ references to Teotihuacan (Spence 1996; Stuart 2000).
Provincial elites may adopt imperial or higher ranking elites’ practices as a means of
demonstrating their own status (Elson and Covey 2006)
Commoners may adopt non-local practices either in direct reference to the source
location, or indirectly, in imitation of their local nobility. For example, in Cuetlaxtlan,
there was an Aztec imperial presence in the regional capital, where local nobles
interacted with Aztec officials using a subset of the full range of Aztec material culture.
In more rural surrounding areas, commoners made and used locally-produced versions of
a subset of that subset of items (Ohnersorgen 2006). At Calixtlahuaca, the Aztec style
items that were recovered in the current excavations of commoner households are
categorically similar to those recovered during Garcia Payón’s excavations of elite and
temple contexts at the site.
There is also variation in the degree of fidelity to which a non-local item or
practice is adopted by a new area. On one end of the scale, non-local practices may be
adopted in forms and with meanings very close to those in the source region, while on the
302
other end all that may be desired is an ability to signal “foreignness” within local
frameworks of meaning. This range of variation may be seen in Aztec Black-on-Orange
pottery, where production outside the Basin of Mexico ranges from very close copies at
Tlacotepec (McVicker, et al. n.d.) to very general references in the Soconusco (Voorhies
and Gasco 2004). In yet other cases, the meanings of objects may be actively subverted
or changed to fit local frameworks. As a result, the meaning of an object in a provincial
area may not be the same as in the source area, though some degree of reference to that
source is generally intended.
Ethnohistoric Evidence for Migration in the Toluca Valley
Ethnohistoric evidence for an Aztec presence in the Toluca Valley is mixed.
Ethnohistoric information on the prehispanic Toluca Valley comes from two groups of
sources – secondary references in sources from the Basin of Mexico, and a number of
bureaucratic documents from the Toluca Valley itself. Each presents its own biases. The
Basin sources show a Triple Alliance view of Central Mexico, presenting Aztec
dominance and Aztec forms of social organization as the natural and proper state of
affairs. In this narrative, the lords of the Matlatzinca are descendents of the kings of Tula,
and the Matlatzinca first appear as one of the groups arriving from Aztlan. They appear in
eight of the twelve migration histories, though García Payón (1942:19-24) argues that
they were added in an effort to justify the Aztec conquest of the Toluca Valley, a position
also supported by Umberger (Umberger and Hernández Fahan submitted 2014). After
this time, there are suggestions that the Tepenac state extended into the eastern Toluca
303
Valley, though the precise geographic extent of this conquest is unknown. The estimated
limits of this zone parallel Sugiura’s proposed Postclassic “Grupo Mica” ceramic
complex identified during her regional survey (Sugiura Yamamoto 2011), though any
definitive association is dependent on a more refined dating of the complex. During this
time period, Calixtlahuaca (“Matlatzinco”) may have paid tribute to the Tepaneca state
(Carrasco 1999), though there is little archaeological evidence of interaction with the
Basin of Mexico during this time period. The written record becomes richer with the
Triple Alliance conquest of the region, providing the names of kings and suggestions of
alliances. Most importantly for my discussion in this chapter, Basalenque (1975 [1642])
describes many Matlatzinca as fleeing to the west after the Aztec conquest of the Toluca
Valley. The historical presence of speakers of Matlatzinca and similar Oto-Pamean
languages just inside the eastern border of the Tarascan Empire would support the
position that at least some of the occupants of the Toluca Valley did emigrate in the face
of Aztec conquest (Pollard 1993:101-103). However, archaeological work at Xaltocan
has demonstrated that Aztec descriptions of the complete abandonment (and later Aztec
reoccupation) of towns may be more political propaganda than truth (Overholtzer 2013).
The Toluca Valley also served as a later source of colonists for imperial colonies in more
distant parts of the empire (Umberger 1996:152-159). Given that such colonist
populations were almost always drawn from either the Basin of Mexico itself or very
heavily Nahuacized populations in immediately adjacent areas, this suggests either the
presence of a significant Nahua population in the Toluca Valley or a high degree of
regional acculturation toward Aztec cultural norms.
304
In contrast, many of the local colonial bureaucratic documents provide more
prosaic information about the Toluca Valley, though even this is often filtered through
particular agendas. There is little Spanish documentation on the Toluca Valley prior to
about AD 1550. Based on the Sumas de Visitas and ecclesiastical records, García Castro
(1999) has reconstructed overlapping zones of Matlatzinca, Mazahua, and Otomi
speakers within the Toluca Valley. The distributions of these languages were
accompanied by the widespread distribution of Nahuatl speakers, who decreased in
frequency with distance from the Basin of Mexico. While the timing of the records means
that there is no way to determine whether the presence of Nahuatl speakers in the Toluca
Valley dates primarily before or after the Spanish conquest, their pronounced east-to-west
distribution gradient does suggest that they were relative latecomers to the region.
Differentiating Migration and Emulation
I apply a framework based on the use of local and non-local style objects and
practices in low and high visibility contexts to distinguish between immigration and local
emulation at Calixtlahuaca (Figure 7.1). Based on archaeological research in the US
Southwest and an extensive review of ethnographic cases, Clark (2001) has presented a
model arguing that low-visibility traits are less likely to be consciously used to signal
group identities, and are therefore more likely to be maintained among immigrant groups.
(Burmeister (2000) makes a similar argument, based on historical and Anglo-Saxon data,
as do Carr and Neitzel (1995) in wide range of case studies.) In contrast, in cases where a
local household is emulating aspects of a foreign group, they are most likely to emulate
305
primarily high-visibility traits. This model is based on the intersection between
Bourdieu’s (1977) argument that many aspects of culture are due to habitus or learned
frameworks, and work showing that some types of material culture are more useful than
others for signaling social differences (Sackett 1977; Wiessner 1983).
As a result, low visibility traits, such as hearth construction, plainware pottery
production techniques, or cooking methods, are likely to be preserved though migration
or cultural change. In contrast, high visibility traits, such as decorated pottery, are likely
to be actively manipulated, often to minimize differences between the immigrant and host
populations or to forge new, common identities between the two. This method of
focusing on low visibility traits has been used to identify immigrant sites in various parts
of the Southwest, and has proven very useful in identifying immigrants in cases which
fail to make the cut-off as full scale site unit intrusions (Clark, et al. 2013; Clark, et al.
2009). Other researchers have found similar contrasts between evidence for migration in
low and high visibility contexts in historical cases (Burmeister 2000; McGuire 1982),
though the maintenance of low-visibility traits is tempered by their appropriateness to the
new environment. While these types of models have not been explicitly applied to
Mesoamerican contexts, there is a growing awareness of the distinction between low and
high visibility traits and their utility for distinguishing between potential causes of
cultural change (Beekman and Christensen 2003; Cowgill 2013).
The model requires some modification for use in more complex societies,
especially those with market-based economies. First, households in such cases are likely
to acquire a significant portion of their household goods from other producers, rather than
making them directly. This makes interpretations based on low visibility production
306
techniques of portable material culture poor candidates for identifying immigrants in
these societies. Instead, the analysis of how items were used in high- and low-visibility
settings is likely to be a more effective line of evidence for identifying non-local
households. At Calixtlahuaca, the relatively low levels of evidence for production above
the household level paired with the relatively even access to non-local goods, suggests an
intermediate situation, where households produced some goods for their own use, but
purchased others. In this case, practices of use in different visibility contexts remain the
best potential line of evidence for identifying immigrant households as people can use
either produced or purchased goods for the same activity.
A: Isolated
Low
High
Visibility Visibility
B: Local Emulation (or Local
Female/Immigrant Male)
Low
High
Visibility Visibility
Local
Style
Foreign
Style
Local
Style
Foreign
Style
X
X
C: Hidden Migration
Low
High
Visibility Visibility
Local
(X)
X
Style
Foreign
X
Style
X = Required for pattern
X
(X)
X
D: Site Unit Intrusion Migration
Low
High
Visibility Visibility
Local
(X)
(X)
Style
Foreign
X
X
Style
(X) = Not required for pattern, but may be present
Figure 7.1 Expected patterns of high- and low- visibility items in cases of migration and
local emulation
307
The second point requiring a modification of the models is that immigrants,
especially from dominant groups, may continue to use both high- and low-visibility
objects and/or practices from their homeland, rather than attempting to assimilate into the
host culture. While this makes immigrant households easier to see archaeologically, it
also makes them easier to confuse with cases of local emulation.
There are four possible pairings of high and low visibility traits and local and
foreign practices. These are presented in Figure 7.1 and referred to as the Isolated, Local
Emulation, Hidden Migration, and Site Unit Intrusion patterns. The patterns described
here are stereotyped extremes, and most archaeological cases will be messier, falling on a
continuum between two of the patterns described here.
In the Isolated pattern, both high- and low- visibility traits are primarily
associated with local style objects and practices. This pattern is probably the most
common archaeologically, associated with sites with little population movement and
either little external interaction or little desire to reference foreign groups. However, sites
in this category are rarely brought in to discussions of migration or emulation because
they are basically the “negative” case where the processes of interest were not occurring.
In the Local Emulation pattern, foreign objects or practices are primarily
associated with high-visibility traits. Low visibility traits continue to follow local
patterns, and there may be some use of high-visibility local traits in conjunction with
foreign ones. Archaeologically, this pattern can be seen in the Mantaro Valley of Peru,
where decorated Inka pottery occurs in conjunction with local plainwares (D'Altroy and
Hastorf 2001), or in the Cuexcatlan region of Mexico where most of the evidence for an
Aztec presence occurs in high-visibility objects (Ohnersorgen 2006). Due to the
308
association between female activities and low-visibility activities, this pattern does run
the risk of equifinality with the pattern produced by settlements made up of local females
and foreign males. This latter interpretation was originally pioneered by Deagan (Deagan
1973; Deagan and Koch 1983) for situations of Spanish/Native American interaction. It
has continued to be applied primarily in historical contexts, though there are some
applications to prehistoric contexts, such as the Egyptian/Nubian frontier (Smith 2003f),
and Uruk period Hacinebi (Stein 2012). The latter case has the advantage of being
sufficiently nuanced to avoid the issue of equifinality; both male (butchering) and female
(cooking) low visibility practices are examined, showing that the former were foreign,
but the latter were local.
The third pattern, referred to as Hidden Migration, is characterized by foreign
style objects or practices in low visibility contexts, but local style objects in high
visibility contexts. This is the pattern that originally drove Clark’s model of ways to
identify migrants. To date, cases matching this pattern are primarily limited to the US
Southwest (Clark 2001; Clark, et al. 2009). I suggest that this pattern is likely to occur
when immigrants have a relatively low status vis-à-vis the local population.
The final pattern, of Site Unit Intrusion, is what is most commonly thought of as
evidence for migration. Foreign traits occur in a wide range of both high and low
visibility contexts. Depending on the size of the immigrant group and the type of local
economy, immigrant households may also use substantial amounts of local items. In
Mesoamerica, this pattern can be seen in the Middle Postclassic Mixtequilla, where
foreign items appear in both high (serving vessels) and low visibility (storage and
cooking vessels) objects (Stark and Chance 2008). It may also appear at the Aztec
309
garrison site of Oztuma on the Tarascan frontier, where both food preparation and serving
assemblages include characteristically Aztec elements (Silverstein 2000). Based on
McGuire (1982), this pattern is likely to occur when either the dominant or the
subordinate group strongly benefits from the visible signaling of cultural differences.
Questions for Calixtlahuaca
The combination of historically specific information on the Toluca Valley, the
expected effects of different rulership strategies, and the general implications of different
types of cultural interaction allow for the formulation of three questions:
1. Pre-Aztec Cultural Diversity and Collectivity of Rulership: How culturally
diverse were households at Calixtlahuaca prior to the Aztec conquest of the site?
More collective forms of social organization should generally produce more
homogenous cultural identities, particularly in high-visibility practices and
material culture. If there is substantial cultural variation at the site, I expect to see
distinctive clusters of specific types of material culture, representing ethnically
specific cultural practices, dividing the household components into two or more
distinctive groups during multiple phases. I expect similar dimensions of variation
to continue through time, and that households that remain occupied from one
phase to the next continue to show affiliation with the same group. This line of
inquiry establishes the existing local baseline.
310
2. Overall changes resulting from Aztec rulership: What was the overall degree of
cultural change at Calixtlahuaca following the Aztec conquest of the site and does
it likely relate to imperial rulership strategies? The magnitude of change toward
activities more characteristic of the Basin of Mexico is a product of the directness
of rule. The evenness of the distribution of any such changes among households
measures the degree to which locally collective institutions persisted or formed
under Aztec rule.
3. Causes of changes under Aztec rule: Can the mechanisms of cultural change in
particular households be explained as either local emulation or foreign
immigration? Cases of emulation are expected to be visible as the appearance of
Aztec material culture primarily in high visibility contexts, conforming to the
“Local Emulation” pattern. In contrast, cases of migration will be visible as the
appearance of specific households featuring Basin of Mexico style household
practices (and associated material culture) in low visibility contexts. These
households may also use Aztec-style material in high visibility contexts, but this
is not required to demonstrate the presence of immigrants. As a result, households
in this category may conform to either the “Hidden Migration” or “Site Unit
Intrusion” patterns of low and high visibility material culture.
311
Public and Private Ritual at Calixtlahuaca
I evaluate the three preceding questions using ritual items from Calixtlahuaca and
surrounding sites. Carballo (2015) identifies ritual as an important component of social
dynamics, especially of rulership. Ritual can be either inclusive, such as community-wide
activities meant to promote group solidarity, or exclusive, such as events intended to
demonstrate ownership of esoteric knowledge. As such, evidence for household ritual
practices can both serve to identify the degree to which households were participating in
shared, community-wide practices, and to what extent they were attempting to distinguish
themselves from their neighbors. Higher visibility ritual practices will serve as better
vectors for communicating group identity than lower visibility practices.
More specially, in terms of assessing migration and local emulation, ritual has
both higher and lower visibility aspects. Clark (2001) does not include ritual practices in
his review of ethnographic practices and their utility for identifying migration. However,
Smith (2002) has previously suggested a quadripartite classification of Mesoamerican
ritual practices, dividing them into state and domestic rituals, and between public and
private rituals. In the context of the discussion of evidence for migration, the latter
distinction provides a rough approximation of high and low visibility activities. Ritual
practices are not generally subconscious or isochrestic variations in the strict sense of
low-visibility traits. However, some aspects of ritual do have a much lower visibility than
others, due to the context in which they took place, the size of the artifacts involved, and
the size of the group involved.
312
In order to contrast changes in public and private ritual practices at Calixtlahuaca,
it is first necessary to determine which artifacts are generally associated with each
category. Many objects were likely used in multiple contexts, with multiple degrees of
public visibility. As a result, a perfectly clean split between artifacts used in “public” and
“private” contexts is unlikely, and the assignment of artifacts to one category or the other
should be seen as indicative of a general trend rather than an absolute division.
Objects used primarily in public contexts were determined based on the results of
García Payón’s excavations at Calixtlahuaca. While specific objects cannot be linked to
particular structures in most cases, the 1930s excavations focused on monumental
architecture, including the royal palace, several temple groups, and cemetery burials in
the associated plazas (García Payón 1979). All three of these types of contexts are far
more likely to be associated with public, group oriented ritual activities than with private
rituals. As a result, the artifact classes more commonly recovered in the García Payón
excavations, or shared between the García Payón and domestic excavations, can
generally be taken to be indicative of more public aspects of ritual, while those artifact
classes recovered primarily in domestic contexts can be taken to be characteristic of more
private ritual practices.
Based on this comparison of contexts at Calixtlahuaca, figurines and scored
censers are primarily associated with household contexts, large braziers and Tlaloc
vessels are primarily associated with public, state contexts, and most other censer and
sahumador forms were shared between domestic and state ritual practices (Huster, et al.
2015). This suggests that of the ritual items used in household contexts, figurines are
lowest visibility, both due to their lack of use in contexts associated with state ritual, and
313
because of their small size. Censers and sahumadors are likely to have been mediumvisibility items, with some use in private settings and some in public settings.
Figurines
Figurines are the most commonly used type of artifact in discussion of Aztec
ritual. They are generally associated with private, household-level ritual practices,
particularly those performed by women (Brumfiel 1996; Brumfiel and Overholtzer 2009;
Faust and Halperin 2009; Klein and Victoria Lona 2009; Smith 2002). They are usually
thought to be associated with domestic concerns, such as pregnancy, healing, and
possibly agriculture. This is contrasted with the focus of public-and-temple ritual, which
concentrated on state concerns such as warfare. This contrast can be seen in the complete
absence of figurines in the Templo Mayor excavations (Klein and Victoria Lona 2009),
compared to the ubiquity of figurines in domestic excavations. In addition, the subjects of
figurines, which include animals, humans, and gods/god-impersonators, occur in
decidedly different frequencies than would be expected considering the importance of
various topics in state level Aztec religion (Brumfiel 1998; Klein and Victoria Lona
2009). As a result of both their relatively private contexts of use, and their lack of
association with state religion, figurines provide an appropriate artifact class for
evaluating lower-visibility ritual practices. In addition, figurines should not be considered
a strictly gendered artifact type. In modern Nahua communities, both male and female
curers use cut paper figurines (Sandstrom 2008), which are likely used similarly to
prehispanic ceramic figurines (Sandstrom 2009). Because of their low visibility, changes
314
in figurine use should be considered indicative of more fundamental cultural shifts. These
could either be due to a substantial integration of Aztec beliefs into local society, or due
to the immigration of people with different practices of figurine use.
At Calixtlahuaca, figurines cover a wide range of production methods, time
periods, subjects, and styles (Figure 7.2). The figurines include both molded and handformed pieces. Hand-formed pieces are decorated using a range of incising, punctate, and
applique techniques. The figurines also come from a range of time periods, with
stylistically Formative and Classic period pieces occurring in otherwise Postclassic
contexts.
The figurines depict animals, humans, a small number of deities/deity
impersonators, and an even smaller number of temple models. Stylistically, the figurines
include Aztec style pieces, local Matlatzinca style pieces, and pieces belonging to
broader, shared, Central Mexican traditions of figurine manufacture (“ghost” or “mud
men” figurines). The Aztec style pieces include examples most likely imported from the
Basin of Mexico (fine, hard, orange paste), and examples probably produced locally
(white paste with more inclusions). The figurine coding at Calixtlahuaca classified each
piece by “Type” and “Group”. Type codes the subject matter of the figurine. Group was a
composite category which first separated out non-Postclassic pieces into period based
groups, and then divided the remaining pieces by style and then by paste. As discussed in
Smith (2002), the various aspects of figurines described above can be cross referenced
against each other in assorted combinations, depending on the purpose of inquiry.
315
Figure 7.2 Selected figurines from Calixtlahuaca. A-Curated, Classic period; B-Curated,
Formative period; C-Woman, Aztec style and paste; D-Woman, Aztec style, local paste;
E-Man, Aztec style; F-Man, Matlatzinca style; G-“Mud Man/Ghost”; H-Human; I&JAnimals.
I analyze figurines in three stages. First, I present the general categories of
figurines present at the site, based on type coding, and evaluate the general trajectory of
change in figurine use and amount of variation among household figurine assemblages
during each period. This contributes toward identifying the amount of inter-household
variation present before and after the Aztec conquest of the site. Second, I compare the
use of figurines at Calixtlahuaca with those from other domestic contexts in Postclassic
316
Central Mexico, again based on the subject matter of the figurines. This identifies to what
degree figurine use at Calixtlahuaca was shifting toward a more characteristically Aztec
pattern over time. Third, I consider whether any groups of figurines should be considered
to be imported, based primarily on the group coding. This is accompanied by a
determination of whether non-local figurines are concentrated in particular household
components. This provides information on household-level variation in non-local object
ownership, which complements the prior analysis of non-local use practices. The results
of these analyses show that figurine use at Calixtlahuaca moves toward a more
characteristically Aztec pattern over time, based on total quantity of figurines, their
subject matter, and the frequency of imported pieces.
Interhousehold Variation in Figurine Use over Time
Differences in figurine content over time and between elite and commoner
contexts have provided useful information in other studies (Brumfiel 1998). While the
approach has generally been used in collections consisting primarily of molded, Aztectype figurines, it can be applied more broadly across multi-style figurine assemblages
(e.g. Olson 2007). In this case, contemporaneous differences in subject matter might
reflect differences in ritual practices resulting from differences in social group identity. In
order to allow comparisons both among households at Calixtlahuaca and among sites, I
coded grouped figurines into general subject-matter based categories.
Figurine Type Categories: In order to allow for both inter- and intra-site comparisons of
figurine subject matter, I categorized figurines into very broad subject categories: curated
317
pieces, women, men, unspecified human, ghost/mud-men, animal, and temple model. At
Calixtlahuaca, these data were coded under figurine “Type”, though some projectspecific categories had to be combined to match the intersite categories, and Formative/
Classic pieces were separated based on “Group” codes. Other items sometimes included
in figurines analyses, such as non-anthropomorphic rattles, ornaments from whistles,
flutes, censers, or sahumadors, and effigy vessels are excluded from this analysis because
they most likely were used in different contexts.
Figurines dating to earlier time periods are commonly found in Postclassic
contexts, and are referred to as curated figurines for the purposes of this analysis (Figure
7.2, A, B). They were likely found at older sites nearby, similarly to how farmers in the
modern village of Calixtlahuaca will often have a collection of figurines that they have
found in their fields. They are treated as a separate analytical category, rather than
grouped by subject matter, as they were likely already broken when found, and thus used
differently. At Calixtlahuaca, these “recycled” figurines include pieces in both Formative
and Classic period styles. Pastes are generally buff colored and not especially fine,
generally matching vessel clays from the Toluca Valley. All of the comparative sites
discussed in the next section are also located near earlier period sites, meaning that it
would not have been difficult for Postclassic residents to acquire these pieces.
Women are one of the most common subjects depicted in figurines, and these
depictions crosscut most of the major stylistic types at the studied sites (Figure 7.2, C, D).
The frequency of figurines showing women tends to increase over time at Postclassic
sites. At Calixtlahuaca, figurines were coded as women based on the presence of breasts,
skirts, quechquemitl (triangular blouses) or female-associated hairstyles/headdresses. If
318
hollow rattle figurines were separated in an original classification, these were considered
women for the purposes of this analysis, as most rattle figurines are women (The rare
exceptions are usually monkeys). At other sites, I followed the original analysts’
determinations of gender, though most use similar characteristics. This category includes
primarily flat-molded and hollow molded pieces at most of the comparative sites, and a
mix of these and hand-formed pieces at Calixtlahuaca.
Men are somewhat less common among figurines, but like women, cross-cut
stylistic categories. They tend to decrease in frequency over time at Postclassic sites
(Figure 7.2, E, F). At Calixtlahuaca, figurines were coded as men on the basis of
loincloths, headdresses, or attributes associated with male deities, and the analysts at the
comparative sites generally used similar criteria. This category includes both molded and
hand-formed pieces.
Figurines were coded as human if they were generally humanoid, but could not be
assigned to a particular gender (Figure 7.2, H). As a result, this category includes both
pieces that were probably not gendered, as well as pieces that probably were gendered,
but were too small to identify.
“Ghost” or “Mud Men” figurines are a stylistically consistent group of crude,
hand-molded humanoid figurines with punctate facial features and/or body decoration
(Figure 7.2, G). They occur widely across Postclassic Central Mexico, but usually match
local clays and there is no reason to think that they were traded from a single point of
origin. They generally decrease in frequency over time, indicating that they are
predominately a Middle Postclassic (or earlier) type in most regions. Brumfiel and
Overholtzer (2009) argue that they may be intended to represent non-human beings.
319
Animal figurines are a diverse group, including both molded and hand-formed
pieces (Figure 7.2, I, J). They feature a range of species, including dogs, opossums, frogs
and monkeys. (Most ceramic birds were attached to musical instruments, and are not
considered here.) There is some variation in which species occur at different sites.
Animal figurines are generally made from local pastes.
Temple models are small figurines of stepped pyramids, topped by a temple or
image of a deity. They are a rare form of figurine across the entire comparative area and
only three examples were found at Calixtlahuaca, none of which are in the DS-2 sample.
Variation at Calixtlahuaca. The use of figurines at Calixtlahuaca over time has two
dimensions. First, did average figurine use practices change over time? Second, did the
amount of variation among households dating to the same phase change over time?
The figurine assemblage at Calixtlahuaca shows a trend toward increasing
frequencies of figurines over time, especially between the Ninupi and Yata phases (Table
7.1). While the difference in figurine frequency between the Dongu and Ninupi phases is
not statistically significant, the increase between the Ninupi and Yata phases is at the .90
level, based on a t-test of means. The Yata phase households are also more variable, with
the coefficient of variation being much higher than those for either preceding phase.
Additionally, the only two components to fall outside of one standard deviation from the
mean for all phases, 307-Ph6 and 309-Ph6, are both from the Yata phase. These
components have figurine frequencies 2-4 times higher than any other component during
any other phase. Taken together, these patterns suggest an uneven adoption of changes to
figurine use during the Yata phase.
320
Figurine content also shifts over time (Table 7.1). When the components with
very low numbers of figurines (1-2 pieces) are excluded, the assemblage is characterized
by increasing proportions of female and male figurines, and decreasing proportions of
curated, ghost, and unspecified human figurines over time. The remaining two type
categories are either absent during all phases (temple models) or occur only in a few
pieces in each phase (animals). Due to the small number of cases, I cannot formally test
whether these patterns represent a significant change.
The level of variation in figurine content among components remains similar over
time. The average Euclidian distances among cases with more than three identifiable
figurines are similar across the three phases, ranging from 38-41. The ranges of withinphase variation for most figurine categories also cover similar ranges across the three
phases. In summary, there are larger changes and more variability in the quantity of
figurines used at Calixtlahuaca over time. These are accompanied by relatively minor
changes in the content of figurines.
321
F/1000 N.
% of Known Subjects
Unit N. Figs Sherds Known Curated Female Male Human Ghost Animal Temple
Dongu
307
4
0.69
2
0.0
0.0
0.0 100.0
0.0
0.0
0.0
315
9
0.54 11
27.3
9.1
0.0
54.5
9.1
0.0
0.0
316
0.00
320
6
0.49
4
50.0 25.0
0.0
25.0
0.0
0.0
0.0
323
12
0.45 10
30.0 10.0
0.0
0.0 60.0
0.0
0.0
324
1
1.09
1
0.0
0.0
0.0
0.0
0.0 100.0
0.0
Ninupi
303
7
0.77
9
66.7
0.0
0.0
22.2
0.0 11.1
0.0
307
12
0.54 11
18.2 27.3 18.2
0.0 36.4
0.0
0.0
308
3
0.69
3
33.3 33.3
0.0
33.3
0.0
0.0
0.0
311
8
1.02
6
83.3
0.0
0.0
16.7
0.0
0.0
0.0
316
12
0.53 11
18.2 18.2
9.1
18.2 27.3
9.1
0.0
322
1
0.54
1
0.0
0.0
0.0 100.0
0.0
0.0
0.0
Yata
307
40
4.29 40
22.5 30.0 17.5
22.5
5.0
2.5
0.0
309
9
2.85
9
11.1 33.3 33.3
11.1
0.0 11.1
0.0
316
6
0.59
6
16.7 66.7
0.0
16.7
0.0
0.0
0.0
317
8
1.01
8
25.0
0.0
0.0
25.0 37.5 12.5
0.0
324
0.00
327
0.00
Dongu Mean, All
Ninupi Mean, All
Yata Mean, All
0.54
0.68
1.46
Dongu Mean, Large
Ninupi Mean, Large
Yata Mean, Large
Dongu SD
Ninupi SD
Yata SD
21.5
36.6
18.8
8.8
13.1
32.5
0.0
4.5
12.7
35.9
31.7
18.8
13.8
10.6
10.6
20.0
3.4
6.5
0.0
0.0
0.0
35.8
43.9
18.8
14.7
15.8
32.5
0.0
5.5
12.7
26.5
18.1
18.8
23.0
12.7
10.6
0.0
4.0
6.5
0.0
0.0
0.0
0.36
0.19
1.74
Dongu CoVar
0.65
Ninupi CoVar
0.28
Yata CoVar
1.19
Note: Summary statistics for subject categories do not include components without
figurines
Table 7.1 Total figurine frequencies per 1000 sherds and subject matter percentages of
identifiable pieces, by household component and phase at Calixtlahuaca
322
Intersite Comparisons of Figurines
The analyses in the prior section demonstrated that figurine use at Calixtlahuaca
changed over time, in quantity and possibly also in content. However, in order to
determine whether this change was influenced by cultural practices originating in other
regions, I now compare the quantity and subject matter of the figurine assemblages at
Calixtlahuaca to those from other sites in Central Mexico. On a general level, I compare
the figurine frequencies at Calixtlahuaca to a wide range of contemporaneous sites in
Central Mexico (Table 7.2). I then compare the specific subject matter of figurines at
Calixtlahuaca, using the previously described “Type” categories, to those from the sites
of Cuexcomate, Capilco, and Yautepec in Morelos (M. Smith, personal com.), and
Xaltocan (Brumfiel 2005b; Brumfiel and Overholtzer 2009), Chalco (Hodge 2008), Xico
and Huexotla (Brumfiel 1996) in the Basin of Mexico (Table 7.3).
Calixtlahuaca and the two comparative regions show differences in both the total
quantity and the relative frequencies of different types of figurines (Table 7.2, Table 7.3).
Prior to the Yata phase (LPC-B), Calixtlahuaca and other Toluca Valley sites have
frequencies substantially lower than those seen in Morelos and at the lower end of the
range for sites in the Basin of Mexico. During the Yata phase, figurine frequencies jump
into the range seen in Morelos and above the frequencies seen at most sites in the Basin
of Mexico. Sites in the Basin of Mexico and Morelos show inconsistent temporal
patterning in figurine frequencies; at some sites figurines increase over time, while at
others they decrease. As a result, the increase in figurine use at Calixtlahuaca during the
Yata phase is consistent with a shift toward more pan-Central Mexican patterns of
323
figurine use, rather than being part of a broader regional trend toward increased figurine
usage across the Postclassic period.
The most pronounced differences in figurine content among regions occur in the
Curated, Male, and Ghost figurine categories. As a general pattern, sites in Morelos have
very low frequencies of curated figurines, and sites in the Basin have moderate levels.
During the Dongu and Ninupi phases, Calixtlahuaca has higher frequencies of curated
figurines than either of the other regions, but during the Yata phase the frequency drops
to midway in the Basin of Mexico value range.
Morelos generally has lower frequencies of male figurines than the Basin of
Mexico. During the Dongu and Ninupi phases, Calixtlahuaca generally has lower
frequencies than either of these regions. During the Yata phases, the frequency of male
figurines at Calixtlahuaca is comparable to those observed in Morelos, and within the
lower end of the range for the Basin of Mexico.
Ghost figurines are absent at all Basin of Mexico sites except Xaltocan, and
primarily present in Middle Postclassic components in Morelos. At Calixtlahuaca, they
occur in the highest levels of any region during all phases, and the primary decrease in
quantity occurs between the LPC-A and B, later than in Morelos.
Animal figurine frequencies are similar for the Basin of Mexico and Morelos,
while those for Calixtlahuaca are lower. There are also site-specific differences in the
particular types of animals depicted in figurines. The three regions under comparison all
have highly overlapping ranges for the other categories of figurines: Female, Human, and
Temple Models.
324
Site
Phase
N. F/1000 F/100
Figs Sherds Rims
Total
sherds
Total
Rims Source
Toluca Valley
Calixtlahuaca MPC
32
0.49
0.50 65,709
6,382
Calixtlahuaca LPC-A
43
0.63
Calixtlahuaca LPC-B
73
1.81
0.59 67,946
1.60 40,321
4,568
4
0.02
16,689
PiñaChan 1981, p. 92, 119
71
1.10
64,424
PiñaChan 1975, p.194,322-4
4
0.64
6,204
Huamango
EPI-EPC
Teotenango
PC
Cerro Toloche LPC
7,317
Jaramillo & De la Peña
2012, Tbl 2
Basin of Mexico
Xaltocan
EPC
8
0.83
Xaltocan
MPC
33
0.88
Xaltocan
LPC
23
1.00
Chalco
PC
15
1.04
Chalco
MPC
10
0.48
965 Brumfiel 2005a, Tables 4.4,
3,740 4.5. Phases revised per
Overholtzer 2014
2,298
1,442 Hodge 2008, Tables 9.2-9.7,
2,101 17.2, 17.3
Chalco
LPC
18
0.79
2,288
Huexotla
a
MPC
24
0.74
3,225 Brumfiel 1996, Table 5.1
Huexotla
a
LPC
76
0.64
11,874
Huex. Pied.a
LPC
59
0.77
7,659
Xicoa
MPC
25
0.49
5,062 Brumfiel 1996, Table 5.2
LPC
12
0.37
3,248
Cihuatecpan
LPC
409
4.21
97,252
Evans 1988, Table 1.2
Tlatelolco
LPC
62
5.05
12,284
Gonzalez Rul 1988a, p. 186
Zacatenco
LPC
32
3.68
8,681
Gonzalez Rul 1988b, Tbl. 1
Capilco
MPC
10
2.06
4,848
Capilco
LPC-A
56
1.23
45,632
Smith 2006a, Table B1-9, B110 and personal com.
Capilco
LPC-B
79
1.33
59,398
Cuexcomate
LPC-A
67
2.88
23,238
Cuexcomate
LPC-B
262
2.72
96,289
Yautepec
MPC
64
1.17
54,682
Yautepec
LPC-A
262
1.48
176,602
Yautepec
LCP-B
839
1.50
559,339
Yautepec
COL
24
2.77
8,677
Xico
a
Morelos
Smith 2006c, Table C1-5,
personal com.
Note: Postclassic Central Mexican sites have approximately 1/10 rim-to-total sherd ratios, so
the figurines per total sherds and figurines per rims are offset by a factor of ten to produce
equivalent values between the two data types
a
Surface Survey
Table 7.2 Figurine frequencies per 1000 sherds and/or 100 rims at Calixtlahuaca and
comparative Postclassic Central Mexican sites, arranged by region
325
Table 7.3 Percentages of figurine subject categories at Calixtlahuaca and comparative
sites, with K-means clustering results for 2-6 groups (Sources same as Table 7.2)
326
Site
Toluca Valley
Calixtlahuaca
Calixtlahuaca
Calixtlahuaca
Basin of Mexico
Xaltocan
Xaltocan
Huexotla
Huexotla
Huex. Piedmont
Xico
Xico
Chalco
Morelos
Yautepec
Yautepec
Yautepec
Capilco
Capilco
Capilco
Cuexcomate
Cuexcomate
Phase
MPC
LPC-A
LPC-B
EPC-MPC
MPC-LPC
MPC
LPC
LPC
MPC
LPC
PC
MPC
LPC-A
LPC-B
MPC
LPC-A
LPC-B
LPC-A
LPC-B
N.
% of Known Subjects
N. Figs Known Curated Female Male Human Ghost Animal Pyramid
K-Means Clusters
2 3 4 5 6
32
43
73
28
41
63
28.6
39.0
20.6
10.7
14.6
30.2
0.0
7.3
15.9
32.1
17.1
20.6
25.0
17.1
7.9
3.6
4.9
4.8
0.0
0.0
0.0
1
1
2
1
2
1
1
4
1
1
1
5
1
1
3
24
76
59
25
12
43
107
177
24
76
59
25
11
33
18.7
11.3
4.2
6.6
1.7
56.0
27.3
15.2
10.3
33.9
45.8
57.9
62.7
20.0
0.0
18.2
13.1
10.7
33.3
19.7
13.6
20.0
27.3
12.1
31.8
28.2
0.0
2.6
8.5
4.0
0.0
30.3
14.0
4.5
0.0
0.0
0.0
0.0
0.0
0.0
12.1
11.3
16.7
11.8
8.5
0.0
36.4
24.2
0.0
0.0
0.0
1.3
5.1
0.0
9.1
0.0
1
2
2
2
2
1
1
2
1
1
3
3
3
2
2
1
1
1
2
2
2
4
3
1
1
5
4
4
4
2
3
5
3
3
6
6
6
4
2
3
64
262
839
10
56
79
67
262
58
226
744
6
48
69
59
212
1.7
2.2
1.9
0.0
0.0
1.4
0.0
0.5
25.9
36.3
33.9
16.7
20.8
21.7
22.0
13.7
3.4
10.2
13.4
16.7
18.8
15.9
15.3
17.9
51.7
37.6
32.4
50.0
29.2
37.7
25.4
33.0
6.9
2.2
1.6
16.7
4.2
1.4
0.0
0.9
10.3
11.5
15.2
0.0
8.3
8.7
11.9
10.8
0.0
0.0
1.6
0.0
0.0
0.0
0.0
0.0
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
5
5
5
5
5
5
5
5
5
3
3
5
3
3
3
3
In order to identify patterns of variation that might cross-cut regional divisions, I
performed both hierarchical clustering and k-means clustering for 2-6 groups (Table 7.3,
Figure 7.3). These two methods produced relatively similar clustering patterns. Both
separate out the cases from Xico and Huexotla relatively early, suggesting that the
survey, rather than excavation, nature of these collections produced skewed results.
Among the remaining cases, temporal patterns play a significant role, with Late
Postclassic cases generally clustering together and Middle Postclassic cases generally
clustering together. The first two phases at Calixtlahuaca are generally relatively
distinctive. The cases from the Basin of Mexico and Morelos divide primarily along
temporal lines, with the Middle Postclassic cases separating from the Late Postclassic
cases at higher numbers of clusters. Those cases from Morelos tend to be more uniform
than those from the Basin of Mexico, possibly because they were all coded by a single
research team, which would have reduced variability in how ambivalent pieces were
classified. Most importantly the Dongu and Ninupi-phase Calixtlahuaca cases group
together while the Yata phase is usually grouped with sites from other regions, suggesting
a shift in figurine content toward a more Aztec pattern during the last phase of the site’s
occupation.
327
Figure 7.3 Hierarchical clustering results for figurines for Calixtlahuaca by phase and
comparative sites
Local and Imported Figurines at Calixtlahuaca
In addition to coding by subject matter (“Type”), figurines at Calixtlahuaca were
also sorted in to groups. Groups first sorted out stylistically non-Postclassic figurines by
period, and then sorted stylistically Postclassic pieces by paste. This classification allows
for the evaluation of temporal trends in the total quantity and interhousehold variability in
access to non-local figurines. In conjunction with the analyses in the previous section,
this provides a measure of the degree to which figurines as physical objects and figurine328
related ritual processes were being adopted together. The use of imported Aztec figurines
without the adoption of associated practices suggests a relatively low level of cultural
diffusion, while the use of both suggests a higher level of information transfer.
Based on the Group codes, the figurines at Calixtlahuaca can be divided into three
source groups: Aztec (consisting of figurines assigned to the Aztec Orange, Orange
slipped, and Fine White paste groups), Local (consisting of local pastes, miniatures, and
stylistically Formative and Classic pieces9), and unknown (a small number of truly
unknown pieces, and all of the “Colonial” group figurines depicting Spaniards, as these
include both Aztec and local paste pieces.) (Table 7.4). Due to INAH policy prohibiting
the export of figurines for technical analysis, no petrographic or INAA sourcing data are
available. As a result, paste is discussed primarily in terms of color and fineness.
The groups assigned to the Aztec category are visually consistent with figurines
from the Basin of Mexico, consisting of very fine white and/or orange clays. Where these
pastes occur in fragments large enough to identify stylistically, they are also consistent
with pieces from the Basin of Mexico.
The primary local figurine category is consistent with those observed in the
majority of locally produced ceramic vessels. These pieces incorporate a wide range of
styles, including local-molded, local-hand shaped, mud-men, and a few Aztec-style
figurines. These pastes account for about two thirds of the figurines in the DS-2 sample
used for this analysis.
Formative and Classic period pieces are included in the “local” category on the premise that the
Postclassic residents of Calixtlahuaca found them at nearby sites.
9
329
Unit N. Sherds
Dongu
307
5,810
315
16,775
320
12,189
323
26,947
324
914
Ninupi
303
9,043
307
22,330
308
4,359
311
7,838
316
22,563
322
1,855
Yata
307
10,257
309
4,217
316
10,091
317
10,860
Local &
PrePC
Aztec
Col &
Unk
Total
Az/Local
Figurines
Ratio
3
8
5
11
1
1
1
1
1
0
0
0
0
0
0
4
9
6
12
1
0.33
0.13
0.20
0.09
0.00
7
8
2
8
8
1
0
1
1
0
3
0
0
3
0
0
1
0
7
12
3
8
12
1
0.00
0.13
0.50
0.00
0.38
0.00
16
7
4
7
9
3
2
2
19
2
0
2
44
12
6
11
0.56
0.43
0.50
0.29
Dongu Mean
Ninupi Mean
Yata Mean
0.15
0.17
0.44
Dongu SD
Ninupi SD
Yata SD
0.13
0.22
0.12
Table 7.4 Figurine paste groups (Local/Pre-Postclassic, Aztec, Colonial/Unknown) by
household component. Excludes components without figurines.
Based on these source group assignments, the distribution of figurines among
households can be assessed two ways: the frequency of figurines in each paste group per
1000 sherds, and the proportion of Aztec to local paste pieces (Figure 7.4). There are two
components with a percentage of Aztec figurines more than one standard deviation from
330
the all-component averages, 308-Ph4 and 316-Ph6. However, both of these components
have low numbers of figurines (3 and 6, respectively), making it likely that their
somewhat elevated frequencies of Aztec figurines are due to sample size. Unfortunately,
two of the Yata phase components (324 and 327) did not recover any figurines, most
likely due to their small sample size, making it difficult to determine whether the
anomalously high values for total and/or Aztec figurines during the Yata phase are
specific to particular households or are part of a broader chronological pattern.
Figure 7.4 Figurine frequencies per 1000 sherds by household component, subdivided by
figurine source group
The proportion of Aztec-paste figurines approximately triples during the Yata
phase. According a t-test of means, this increase is significant at the .90 level, relative to
either of the preceeding phases. However, the Yata phase does not have a higher standard
deviation for the proportion of Aztec-paste figurines, indicating that this increase was
relatively evenly distributed among all households with figurines during this phase.
331
Figurine Analyses: Results
The analyses of figurines pertain to all three of the questions presented for this
chapter: overall trends over time, the adoption of potentially non-local practices under
Aztec rule, and the possible explanations for any changes.
First, there are relatively few changes between the Dongu and Ninupi phases, and
households are relatively homogenous during these periods. The overall household use of
figurines is similar between these two phases, based on the total quantity of figurines and
the proportion of Aztec-paste figurines. The figurine subject matter of these two phases is
also similar, as they usually cluster as each other’s closest case when compared with
figurine assemblages from sites on a regional scale. The amount of interphase variation is
also relatively low, based on multiple measures. While many households at Calixtlahuaca
do not have large enough assemblages of figurines to permit formal analyses of
similarity, most of the households have assemblages with high frequencies of curated
figurines. Most households also have female, human or ghost figurines, with few to no
male, animal or temple model figurines. All household components with more than two
figurines have pieces in at least two of the primary categories, and these do not form
mutually exclusive groups (i.e., women-and-ghosts vs. curated-and-human). This
suggests a lack of multiple, distinctive groups and a relatively collective orientation to
figurine-based ritual at Calixtlahuaca prior to the Aztec conquest of the site.
Second, there are significant changes in figurine usage at the site during the Yata
phase. The total quantity and the proportion of Aztec-paste figurines both show
statistically significant increases relative to the two preceding phases. In most clustering
solutions, this phase at Calixtlahuaca is grouped with comparative cases from other parts
332
of Central Mexico, indicating that the content of the figurines at the site is also shifting
toward a more characteristically Aztec pattern. These overall changes are accompanied
by an increase in interhousehold variation, indicating that the changes were not evenly
adopted across households at the site. This uneven adoption is consistent with relatively
network-oriented rulership by the Aztec Empire, where cultural change was not targeted
toward commoners.
As a relative low visibility item demonstrating the adoption of both non-local
items and non-local practices, the Yata-phase figurine assemblage at Calixtlahuaca
demonstrates a relatively high level of “Aztec-ization”. Due to the low visibility of
figurines, they may serve as a marker for immigrant households from the Basin of
Mexico, though this will require further comparisons with other lines of evidence.
Censers
Postclassic Central Mexican censers are a catch-all category, encompassing a
wide range of sizes, shapes, and styles of artifacts (Figure 7.5). They include small
censers that could have been held in a single hand to temple braziers that would have
required several people to move. They also include items that were intended to be used in
motion, such as sahumadors (frying pan censers), and items that were most likely
intended to be stationary, such as the biconical censers found at Calixtlahuaca. Censers
are presumed to have been used for burning incense, or offerings of blood on grass or
paper. This is rarely demonstrated though residue analysis and some plainer examples
may have been heating/cooking braziers rather than ritual items.
333
As with the figurines, I analyzed censers in three stages. The first stage was to
measure trends of overall censer use and interhousehold variation over time at
Calixtlahuaca. The second stage was to compare the uses of censers at Calixtlahuaca with
those found at other Postclassic Aztec sites. The third stage was to determine whether
imported/foreign style censers occur at Calixtlahuaca, and whether they were
concentrated in particular households.
Figure 7.5 Censer and sahumador forms. A-D are freestanding forms found in the Toluca
Valley (A), the Basin of Mexico (B-D), and/or Morelos (C-D). E is a scored censer, found
in all three regions. F-H are sahumadors, with F characteristic of the Basin of Mexico
and G-H found largely in the Toluca Valley. (Relative sizes of censers approximate)
334
These analyses show that household censer use at Calixtlahuaca is relatively
homogenous within phases, and highly stable over all three phases of the site’s history.
This stability occurs despite the presence of potential alternative patterns of censer use in
other parts of Central Mexico, including in the southern Basin of Mexico core of the
Aztec Empire.
Interhousehold Variation in Censer Use over Time
In order to allow for both intra- and inter-site comparisons of censers, the
individual types at Calixtlahuaca were grouped into three broader functional categories
which could also be applied to other regions: top-loaded censers generally used in
stationary manner (large freestanding censers), bottom-loaded censers which would not
have produced a smoke plume (scored censers), and censers used in motion (sahumadors)
(Figure 7.5).
The first category, stationary censers, includes a range of vessel forms and
decorative surface finishes, including both bowl and hourglass shapes (Figure 7.5, A-D).
The types placed in this category are characterized as being able to be moved by a single
individual without too much trouble, but with the ability to be freestanding when placed
on a surface. Most pieces in this category are decorated with appliqué points, spikes,
flanges, or pie-crust rims. They include pieces with a range of plain, slipped, and
plastered finishes. At Calixtlahuaca, the only type placed in this category are biconical
hourglass censers (Type 267).
The second category, scored censers, is a distinctive category of braziers or
censers shaped in the form of a small round base, topped by three legs, supporting a dome
335
(Figure 7.5, E). In the sense that they are both reasonably portable and freestanding, they
are similar to the previous category. However, they are very rough pieces with no
decoration other than very heavy surface scoring, which is likely functional rather than
decorative due to its irregular depth, angle, and spacing. Their form also requires
different mechanics of use; coals and copal can be dropped directly into the open mouth
of vessels in the previous category, while scored censers require loading from the bottom.
These censers are variably referred to as “Saturnos”, “Scored Censers”, “Lantern
Censers” and “Cross-Hatched Ware” by different researchers. At Calixtlahuaca, this
category is listed under two type codes, 101 and 180, because the upper and lower
portions of the type were originally considered two separate vessel forms.
The third category, sahumadors, consists of a small bowl at the end of a single
long handle (Figure 4.5, F-H). Some examples also include two supports. They are lighter
than any of the previous types and intended for use in motion. Depictions of Aztec priests
regularly show them holding and using this form of censer. This category includes both
the Texcoco molded and filleted sahumadors characteristic of the Middle and Late
Postclassic Basin of Mexico, as well as other types with similar forms. At Calixtlahuaca
this category includes local Red-on-Buff sahumadors (Type 262), Basin of Mexico stype
Texcoco molded/filleted sahumadors (Type 64), and the miscellaneous Other Censers
(Type 170) category, because after the initial ceramic classification we realized that most
of the pieces in this category were low-frequency local sahumador variants.
A fourth potential category, massive temple braziers (freestanding, not easily
portable), was excluded from the analysis because it is primarily associated with state
ritual and very rarely occurs in domestic contexts. The few fragments of this type
336
recovered at Calixtlahuaca, which in at least one case were likely buried as offerings,
were coded as Type 65.
Within-site variation in censer use of these three functional types at Calixtlahuaca
over time includes two components: average trends over time and the degree of intraphase variation. Average trends over time provide information about general shifts in
ritual practices, while intra-phase variation measures how uniformly any changes were
adopted by the residents of Calixtlahuaca. The rim- and total-sherd based frequencies of
each functional censer category are given in Table 7.5.
Based on these frequencies, there are no changes in overall censer use at
Calixtlahuaca over time. On a simple presence/absence basis all but one household
component during all three phases have examples of all three functional types. On a more
nuanced level, based on t-tests of means, none of the phase means for the total quantity of
censers are significantly different at the .90 level. In addition, the patterning of change is
different between the total sherd and rim sherd based values, suggesting a relatively high
degree of random noise. The temporal patterning for individual functional classes of
censers in more internally consistent, with sahumador frequencies dropping and scored
censer frequencies rising in both data sets between the Ninupi and Yata phases. However,
the only pair of means for these types which are significantly different at the .90 level are
scored censer rims between the Ninupi and Yata phases.
337
N.
Unit Rims
Dongu
307
411
315 1,374
316
290
320
395
323
834
324
77
Ninupi
303
824
307 2,127
308
329
311
570
316 1,851
322
211
Yata
307 1,130
309
332
316
921
317 1,114
324
373
327
97
% of Total Rims
Free Scored Sahum Total
N.
Sherds
% of Total Sherds
Free Scored Sahum Total
2.92
3.64
6.21
2.53
3.24
2.60
0.24
0.15
0.69
0.25
0.48
0.00
0.49
1.02
1.72
1.01
4.56
1.30
3.65
4.80
8.62
3.80
8.27
3.90
4,756
13,772
2,684
3,951
8,827
892
0.53
1.29
1.97
1.47
1.01
0.67
0.11
0.11
0.22
0.13
0.14
0.11
0.17
0.25
0.34
0.10
0.79
0.22
0.80
1.65
2.53
1.70
1.94
1.01
2.18
4.28
3.34
1.75
4.27
2.84
0.24
0.09
0.30
0.18
0.27
0.00
0.24
1.65
1.22
1.05
2.76
0.47
2.67
6.02
4.86
2.98
7.29
3.32
8,991
20,259
3,724
4,986
15,794
1,668
1.18
2.18
1.13
0.88
2.00
2.04
0.09
0.25
0.19
0.06
0.25
0.12
0.09
0.45
0.27
0.26
0.68
0.18
1.36
2.88
1.58
1.20
2.93
2.34
2.65
1.81
4.02
2.06
1.34
6.19
0.27
0.30
0.54
0.54
0.27
1.03
0.35
0.60
0.76
0.72
1.61
0.00
3.27
2.71
5.32
3.32
3.22
7.22
10,225
3,096
7,492
9,616
3,754
945
1.18
0.94
1.49
1.22
1.07
1.80
0.20
0.16
0.21
0.26
0.13
0.32
0.15
0.19
0.31
0.22
0.24
0.00
1.53
1.29
2.02
1.70
1.44
2.12
Dongu Mean
Ninupi Mean
Yata Mean
3.52
3.11
3.01
0.30
0.18
0.49
1.68 5.51
1.23 4.52
0.67 4.18
1.16
1.57
1.28
0.14
0.16
0.21
0.31 1.60
0.32 2.05
0.18 1.68
Dongu SD
Ninupi SD
Yata SD
1.38
1.05
1.81
0.25
0.12
0.29
1.46 2.32
0.90 1.86
0.54 1.74
0.54
0.56
0.31
0.04
0.08
0.07
0.25 0.63
0.21 0.77
0.10 0.33
Dongu CoVar
Ninupi CoVar
Yata CoVar
0.39
0.34
0.60
0.82
0.64
0.60
0.87 0.42
0.73 0.41
0.80 0.42
0.46
0.36
0.24
0.33
0.51
0.31
0.80 0.39
0.66 0.38
0.57 0.19
Dongu Range
Ninupi Range
Yata Range
3.68
2.52
4.85
0.69
0.30
0.77
4.07 4.97
2.51 4.62
1.61 4.51
1.45
1.29
0.86
0.12
0.19
0.18
0.69 1.73
0.59 1.73
0.31 0.82
Table 7.5 Rim and total sherd based frequencies of functional censer classes
(Freestanding, Scored, Sahumador) by household component, with summary statistics by
phase
338
The overall degree of inter-household variation is also similar over time. The
coefficients of variation for both the total quantities of censers and the individual
functional categories are either inconsistently patterned between the rim and total sherd
based data sets or show little change. The ranges of values present during each phase are
also either similar over time or inconsistent between the two data sets.
Intersite Comparisons of Censers
As a result of this lack of variation over time or among phases at Calixtlahuaca, I
frame the macroregional comparison slightly differently than in the previous discussion
of figurines. Primarily, are there differences in censer use between Calixtlahuaca and
contemporaneous sites in other parts of Central Mexico? If there are, then the lack of
change at Calixtlahuaca may reflect a lack of adoption of other possible censer-based
ritual practices. However, if there are few interregional differences, then the lack of
change at Calixtlahuaca does not provide information for or against the adoption of nonlocal practices.
In order to evaluate regional variation, I compared censers from Calixtlahuaca
with those recovered in domestic excavations in the Basin of Mexico, Morelos, and
Puebla using the same three previously-described functional categories. As noted earlier
in this chapter, in a market-based economy, immigrants may choose to purchase
stylistically local items, but use them in non-local manners. They may purchase local
items because comparable items from the homeland are not commercially available and
because the members of the household either lack the skills, raw materials, and/or time to
reproduce such items for themselves. As a result, functional equivalency may be an
339
acceptable substitution, when goods from the homeland or in the homeland style are not
available. Conversely, local people may acquire imported items, but not the associated
knowledge concerning their use, especially in situations of relatively low-intensity
culture contact.
At a general level, there are broad regional trends in censer use. Censer
assemblages at sites in Puebla, Morelos and the southern Basin of Mexico are dominated
by sahumadors. In contrast, sites in the northern Basin of Mexico and the Toluca Valley
tend to be dominated by either scored or freestanding censers. As with figurines,
Xaltocan flips from a northern Basin pattern during the Early and Middle Postclassic to a
more characteristically southern Basin pattern during the Late Postclassic. The total
percentage of censers is not geographically sensitive; there is more variation within
regions than between them and the regional ranges show a high degree of overlap.
In a more formal analysis, I compared the frequency of functional censer classes
among regional sites via k-means and hierarchical clustering. Due to differences in
whether ceramic counts were recorded as rim sherd or total sherd counts, I ran two sets of
analyses. Using rim sherd counts resulted in a comparison between Calixtlahuaca and
sites in both the northern and southern Basin of Mexico. Using total sherd counts resulted
in a broader regional comparison among sites in the Toluca Valley, Morelos, Puebla, and
a smaller set of sites from the Basin of Mexico. All of the comparative datasets are from
either broad sitewide testing or primarily residential deposits, and probably represent
material primarily from commoner households. Due to the way the comparative data
were generally reported, in most cases, the unit of comparison is either all material from a
site or all material from a phase. Due to the small number of variables, K-means and
340
hierarchical clustering methods produced very similar results for both data sets. The Kmeans groups were also very stable, with each additional group (from 2-4) simply
splitting a single previous group.
In the rim sherd based comparative analysis (Table 7.6, Figure 7.6), the total
percentages of censer rim sherds in the Basin of Mexico divide strongly by sub-region,
with the lone site in the southern Basin of Mexico, Chalco, standing out relative to the
remaining northern Basin sites. The phase averages for total censers at Calixtlahuaca fall
near the midpoint of the range seen in the Basin of Mexico, and all individual households
at Calixtlahuaca also fall within the range seen among Basin sites. Censers range from
about 1.5-11% of rim sherds in the Basin sample, and about 2.5-9% in the Calixtlahuaca
sample.
Site
Toluca Valley
Calixtlahuaca
Calixtlahuaca
Calixtlahuaca
Basin of Mexico
Tepexpan
Maquixo
Teacalco
Xometla
Xaltocan
Xaltocan
Xaltocan
Chalco, Unit A
Chalco, Unit B
Chalco, Unit B
Phase
N.
Rims
MPC
LPC-A
LPC-B
LPC
LPC
LPC
LPC
EPC
MPC
LPC
PC
MPC
LPC
1,961
4,000
1,856
562
867
3,731
2,335
1,442
2,101
2,288
% of Total Rims
K-Means Clusters
Free Scored Sahum Total 2 3 4 5 Source
3.52
3.11
3.01
0.30
0.18
0.49
1.68 5.51
1.23 4.52
0.67 4.18
1
1
1
1
1
1
1
1
1
1
1
1
0.25
0.05
0.00
0.00
1.15
0.72
0.60
2.50
2.81
2.14
0.41
0.63
0.54
0.71
0.12
0.21
0.17
0.76
1.90
1.62
1.63
1.00
0.75
2.67
0.46
0.48
2.61
5.69
5.28
7.04
1
1
1
1
1
1
1
2
2
2
3
3
3
3
3
3
3
2
2
2
3
3
3
4
3
3
4
2
2
2
3
3
3
3
5
5
3
4
4
2
2.29
1.68
1.29
3.38
1.73
1.42
3.38
8.95
10.00
10.80
Parsons 1966, Tbls
28-40
Brumfiel 2005a, Tbls
4.2-3. Phases per
Overholtzer 2014
Hodge 2008, Tables
9.2-9.7
Table 7.6 Functional censer form (Freestanding, Scored, Sahumador) frequencies out of
total rim sherds at Calixtlahuaca and comparative sites, with K-means cluster results
341
Both clustering methods divide Calixtlahuaca and the rim-sherd based
comparative cases primarily by site (or site cluster for the rural Teotihuacan Valley sites)
at the higher cluster levels. In addition, both methods separate the Chalco components as
being the most different from all other sites, and group the rural Teotihuacan Valley sites
and Xaltocan together during lower cluster solutions.
Figure 7.6 Hierarchical clustering results for functional censer forms at Calixtlahuaca
and comparative sites, based on percentages of rim sherds..
342
The second set of cluster analyses compares Calixtlahuaca with a broader range of
other regions of Central Mexico, based on total sherd counts (Table 7.7, Figure 7.7). In
these analyses, the cases from Calixtlahuaca generally have a higher total percentage of
censers than the comparative cases from Puebla and Morelos, but similar values to those
seen in the Basin of Mexico.
Site
Toluca Valley
Calixtlahuaca
Calixtlahuaca
Calixtlahuaca
Cerro Toloche
Basin of Mexico
Tlateloco
Zacatenco
Morelos
Capilco
Capilco
Capilco
Cuexcomate
Cuexcomate
Yautepec
Yautepec
Yautepec
Yautepec
Puebla
Cholula
Cholula
Cholula
Phase
MPC
LPC-A
LPC-B
M-LPC
N.
% of Total Sherds
Sherds Free Scored Sahum Total
K-Means Clust.
2 3 4 5 Source
1.16
1.57
1.28
6,204 1.21
0.14
0.16
0.21
0.02
0.31
0.32
0.18
0.90
1.60
2.05
1.68
2.13
2
2
2
2
2
2
2
2
2
2
2
2
12,303 0.30
8,681 0.07
0.01
0.01
2.15 2.46
1.26 1.34
1
1
1
1
4
1
2
2
2
5 Jaramillo & De la Peña
2012, Table 2
4 Gonzalez Rul 1988a, 186-7
1 Gonzalez Rul 1988b, Tbl 1
MPC
3,496 0.12
LPC-A 23,115 0.07
LPC-B 29,444 0.09
LPC-A 20,286 0.43
LPC-B 99,828 0.16
MPC
30,795 0.14
LPC-A 130,206 0.10
LPC-B 204,243 0.15
COL
8,866 0.09
0.00
0.00
0.00
0.00
0.00
0.29
0.18
0.09
0.06
1.19
1.42
1.18
1.97
1.08
0.36
0.52
0.45
0.40
1.30
1.49
1.27
2.39
1.23
0.79
0.80
0.69
0.54
1
1
1
1
1
2
2
2
2
1
1
1
1
1
3
3
3
3
1
1
1
4
1
3
3
3
3
1 Smith 2006a, Table B2-2
1
1
4
1
3 Smith 2006c, Table C2-3
3
3
3
LPC
LPC
MPC
0.28
0.53
0.00
0.28 0.85
0.27 0.80
0.59 0.59
2
2
2
3
3
3
3
3
3
3 McCafferty 2001, Table
3 5.2
3
LPC
LPC
355 0.28
377 0.00
170 0.00
Table 7.7 Functional censer form (Freestanding, Scored, Sahumador) frequencies out of
total sherds at Calixtlahuaca and comparative sites, with K-means cluster results
Once again, the two clustering methods produced similar results, with a strong
tendency to group cases by site. Both methods identify two outlier components –
Tlateloco in the Basin of Mexico and Late Postclassic-A Cuexcomate in Morelos. Both
343
cases separate due to their higher overall frequencies of censers, though their relative
frequencies of the three functional categories otherwise correspond to their respective
regions and at lower k-means cluster values both do group geographically. Otherwise, the
division between the Toluca Valley sites of Calixtlahuaca and Cerro Toloche and all of
the comparative cases forms the first division in the data set. The second division divides
Yautepec in Morelos and Cholula in Puebla from the Basin of Mexico and rural Morelos
sites of Cuexcomate and Capilco.
Figure 7.7 Hierarchical clustering results for functional censer forms at Calixtlahuaca
and comparative sites, based on percentages of total sherd counts.
344
Taken together these analyses demonstrate a reasonably high level of variation in
the use of various censer forms in Middle and Late Postclassic Central Mexico. While
much of this variation is site or sub-region specific, there are general trends toward
higher levels of sahumador use in the southern Basin of Mexico and Morelos, and higher
levels of freestanding and/or scored censer use in the northern Basin and Toluca Valley.
None of the individual Yata-phase household components at Calixtlahuaca have the
higher sahumador frequencies characteristic of the southern Basin heartland of the Aztec
Empire. Given the presence of regional variation in censer use, the consistency of censer
use over time at Calixtlahuaca demonstrates an overall lack of adoption of Aztec
practices, both at the sitewide and individual household levels.
Local and Imported Censers at Calixtlahuaca
As with the analysis of local and non-local figurines, the distribution of non-local
censers at Calixtlahuaca provides an independent means of assessing the relationship
between the adoption (or rejection) of non-local practices and non-local items. I assigned
censer types to local and imported categories, based on two lines of evidence. First, I
considered the INAA and petrographic sample results. I also considered the relative
frequency and antiquity of each type within the Toluca Valley and surrounding areas,
based on published data, on the premise that a type will likely be the oldest and probably
the most common in its source region.
The specialized analysis samples (petrography and INAA) were not selected with
ritual vessels as a primary focus. As a result, the frequency of samples roughly parallels
their frequency within the assemblage as a whole, and some rarer types are not included
345
at all (Table 7.8, Table 7.9). Both types of analysis demonstrate that the majority of all
censers recovered at Calixtlahuaca were locally produced. Interestingly, most censer
types occur in a range of local paste variants, demonstrating that censers were not
produced by a single subgroup of specialists. For three of these types, biconical censers,
scored censers and local style sahumadors, there is no reason to think that they do not
have a long history in the Toluca Valley. In addition to local sourcing results, all three
types occur in all three phases at the Calixtlahuaca excavations and are commonly found
in other Postclassic excavations in the Toluca Valley (Piña Chán 1981:101, 111; Romero
Padilla 2015; Tommasi de Magrelli 1978; Vargas Pacheco 1975:235).
Type No.
267
101/180
262
64
170
Total
Type
Censer, Biconical
Censer, Scored
Sahumador, Toluca
Sahumador, Texcoco
Censer, Other
Local
1
2
3
14
2
1
1
8
1
1
2
1
27 1
3
N.
20
4
11
2
4
41
Aztec
4
5
0
0
S/SW
7
8
1
1
Unknown
5
2
1
1
9
0
Table 7.8 INAA group results (1-9, Unknown) with regional affiliations (Local, Basin of
Mexico, South-Southwest State of Mexico, Unknown) for censers and sahumadors, by
project ceramic classification type
Type No.
267
180/101
262
170
Total
Type
Biconical Censer
Scored Censer
Sahumador, Toluca
Censer, Other
N.
12
5
2
2
21
I-A
5
1
6
I-B
2
1
1
4
Local
II
I-C
1
2
2
III
2
1
Aztec
XIII
1
1
1
5
1
4
1
Table 7.9 Petrography group results with regional affiliations for censers and
sahumadors, by project ceramic classification type
346
The assemblage includes two censer types with reasonable evidence of a possible
non-local origin: Texcoco molded/filleted sahumadors and Other Censers. Texcoco
sahumadors were represented by two sherds in the INAA analysis, but were not included
in the petrography sample. Of the two INAA samples, one was actually assigned to the
primarily local group and Calixtlahuaca and the other remained unassigned. This suggests
that some examples of the type may have been locally produced, possibly by the same
group which produced the Aztec Black-on-Orange pieces that were assigned to the local
source groups. However, the broader temporal and spatial distribution of the ceramic type
in Postclassic Central Mexico makes it clear that it is not local to the Toluca Valley. It
first appears at Calixtlahuaca during the Ninupi phase (AD 1380-1450), and actually
decreases slightly in frequency during the subsequent Yata phase. In the Basin of
Mexico, the type is often attributed to the Late Aztec period (Parsons 1966:255-256), and
there is no doubt that it reaches its highest frequency and extent during this period.
However, an examination of well dated contexts at Chalco (Hodge 2008), Xaltocan
(Brumfiel 2005b, chronology revised based on Overholtzer 2014), and in Morelos (Smith
2006a, 2006c) consistently show that the type first appears during the Middle
Postclassic/Early Aztec period, clearly making it older in both of these regions than in the
Toluca Valley. Once it appears in the Basin of Mexico, it remains in use until after the
Spanish conquest. After Aztec Black-on-Orange, Texcoco-style Sahumadors are one of
the more widespread Late Postclassic ceramic types, appearing from Morelos (Smith
2002), to the Gulf Coast (Garraty and Stark 2002; Ohnersorgen 2006; Venter 2012). The
type also occurs in higher frequencies in the Basin of Mexico, accounting for about 22.5% of rim sherds at Basin of Mexico sites, compared to 0.2% of rim sherds at
347
Calixtlahuaca. As a result, it can clearly be considered a non-local type at Calixtlahuaca,
though one whose original appearance predates the Aztec conquest of the site.
The second case, Other Censers, includes both a fairly uniform group of
sahumador style censers with triangular cut-outs, red or white-on-red decoration, and
strongly outflaring rims on the bowl (Figure 7.5 H), as well as an assortment of
miscellaneous censer pieces that did not match any of the other defined categories of
censers. As a result, I expected that any sourcing of the category was likely to include a
fair amount of noise. However, with the exception of one unassigned INAA sample, all
of the analyzed sherds for this type (4 INAA and 2 petrography) were placed into a local
source or paste group. Not surprisingly, they include a wide range of local assignments.
The dominant subgroup within the type does occur relatively widely in the Toluca
Valley, though examples have also been found in other parts of Central Mexico.
Based on this evaluation of the stylistic origins and likely perceived cultural
affiliation of each type of censer found at Calixtlahuaca, only Texcoco molded/filleted
sahumadors can be considered a type with non-local affiliations. On a sitewide level, the
average household frequency of Texcoco sahumadors rises from completely absent
during the Dongu phase to low, but similar levels during the Ninupi and Yata phases.
This increase parallels the generally increasing levels of imported goods from the Basin
of Mexico between the Middle and Late Postclassic periods.
Given that this is a very rare type (<1%), there is also little change in the level of
within-phase inter-household variation between the Ninupi and Yata phases. The type is
present in four of six Ninupi components and five of six Yata components (Figure 7.8).
The coefficient of variation shows a small decrease between phases, from .6 to .4,
348
indicating that insomuch as there is any change in inter-household variability, it is toward
reduced variation over time. This reduction in variability is also supported by the fact that
the only component with a value more than two standard deviations from the mean for
these two phases is 316-Ph4, a Ninupi phase household.
In summary, the initial appearance and subsequent use of Texcoco molded
censers at Calixtlahuaca is not correlated to the Aztec rule of the site. The censers appear
prior the Aztec conquest of the site, and do not differ dramatically in frequency or interhousehold distribution between the Ninupi and Yata phases.
Figure 7.8 Frequency of Texcoco Molded-Filleted Sahumadors at Calixtlahuaca as a
percentage of total sherds, by household component
349
Censer Analyses: Results
The analyses of censers pertain primarily to the first two of the three questions
presented for this chapter: the degree of local variation and its relation to rulership
strategies, and the degree to which prior patterns of use and variation changed under
Aztec rule. As the answers to these two questions demonstrate that there was little change
over time, the third question for the chapter, concerning the specific causes of variation,
does not apply.
First, in terms of local variation, censer use at Calixtlahuaca was relatively
homogenous during the periods when the site was under local rule, indicating that censerbased ritual was likely integrative at a community level. Based on censer form categories,
there are no strong, ongoing divisions in censer use at the site which might correspond to
multiple social identities. The level of intra-phase variation remains relatively steady
between the Dongu and Ninupi phases. When a new, imported censer type, the Texcoco
sahumador, was introduced at the site during the Ninupi phase, it appeared in the majority
of the households for the phase, indicating widespread access to and interest in the type.
However, total sahumador frequencies for the Ninupi frequencies are comparable to those
from the preceding Dongu phase, indicating the imported Texcoco sahumadors were
likely simply used to replace some local versions of the same functional censer form. The
high degree of interhousehold similarity over time, including in the adoption of a new
type, supports the argument for a relatively collective form of local governance at
Calixtlahuaca prior to Aztec rule.
Second, in terms of changes under Aztec rule, incorporation into the empire also
produced little change in censer use at Calixtlahuaca. Despite the presence of regional
350
variation in censer use in Postclassic Central Mexico, the Yata phase overall sitewide
frequencies of functional censer categories at Calixtlahuaca remain comparable to those
from both preceding phases at the site. Additionally, the levels of interhousehold
variation in both censer form and use of imported censers remain comparable during the
Ninupi and Yata phases. The appearance of Texcoco sahumadors prior to the Aztec
conquest and their integration into local patterns of censer use strongly suggests that their
adoption was on local terms, for local practices, rather than evidence for the spread of
any sort of state cult. The overall lack of change in censer use under Aztec rule is
consistent with both relatively indirect rule by the Aztec Empire, and the continuation of
relatively collective local practices.
Conclusions
In this chapter, I looked at the evidence for the social organization of local and
Aztec rule at Calixtlahuaca though changes in ritual practices over time. My analyses
focused on three questions introduced at the beginning of this chapter: the pre-Aztec
organization of rule, based on the trajectory of change and degree of inter-household
variability, the Aztec organization of rule, also based on the trajectory of change and
variability, and whether the observed Aztec-rule patterns better correspond to models of
local emulation or the arrival immigrants from the Aztec core.
351
Total
Quantity
Figurines
Imported
pieces
Subject Matter
Dongu-Ninupi
Shift toward
No shift
No shift
No shift
Basin-pattern
Evenness of low varation Low variation Low variation
both phases both phases
both phases
variation
Ninupi-Yata
Shift toward
Moderate
Moderate Moderate-high
Basin-pattern
increase
shift
shift
Evenness of
Increased
No increase No increase in
variability in variability
variation
variability
Censers
Imported
Total Quantity
pieces
Form
category
Modest
No shift
increase
Low variation Moderate Low variation
both phases unevenness both phases
No shift
No further
increase
Low variation No further
both phases
increase
No shift
No shift
No shift
Table 7.10 Summary of results of analyses of ritual practices showing shifts in
interhousehold variation and local vs. Basin of Mexico cultural patterns under local and
Aztec imperial rule
I analyze two particular forms of ritual objects, figurines, which are relatively low
visibility and used primarily within households, and censers, which are higher visibility
and used in both household and community level ritual settings. The results of these
analyses are summarized in Table 7.10. Due to the lower visibility of figurines, I
expected them to reflect more pronounced cultural changes than those that would be
reflected in censer use. I also employed a dual consideration of both use of non-local
goods, and the use of non-local practices. People may adopt foreign items but integrate
them into local practices. An adoption of both objects and practices suggests a higher
degree of interaction between cultural groups and contexts of display where foreign useknowledge matters.
First, in terms of local variations prior to Aztec rule, ritual practices at
Calixtlahuaca were both relatively similar among households and relatively similar over
time. They were also distinctive relative to ritual practices from contemporaneous sites in
352
adjacent regions of Central Mexico, for both of the artifact types studied. The total
quantities of both figurines and censers used by households are statistically equivalent
between the Dongu and Ninupi phases. There are minor changes in figurine content
between these two phases, but no significant changes in the relative proportions of major
functional classes of censers. One non-local censer type, the Texcoco molded/filleted
sahumador, first appears at Calixtlahuaca during the Ninupi phase, but the overall lack of
change in total sahumador frequencies suggests that it was simply replacing local
sahumadors in preexisting local contexts of use. Interhousehold variability is also
relatively low for both households during both time periods, both in terms of practices
and access to foreign goods. The same functional subtypes (figurine subjects, censer
forms) dominate almost all household assemblages during these two periods. Similarly,
overall access to non-local goods (Aztec-paste figurines, Texcoco censers) shows a small
increase between the Dongu and Ninupi phases, but the increases are relatively evenly
distributed across households within the Ninupi phase. This high degree of interhouseholds similarly within and between phases is consistent with a relatively collective
form of local governance prior to the Aztec conquest of the site, particularly one which
used collective participation in public ritual as a socially integrative force.
Second, in terms of changes under Aztec rule, the Aztec conquest and rule of
Calixtlahuaca produced changes in figurine use at the site, but few changes to censer use.
During the Yata phase, there is a significant increase in the average household frequency
of figurines at the site, bringing them into line with the frequencies seen in other regions
of Central Mexico, and an overall shift toward figurine subject matter more similar to that
seen in other regions of Central Mexico. However, these shifts toward a more Aztec
353
pattern of figurine use are unevenly distributed among households; there is also an
increase in the variation among households in the total quantity of figurines, and the
proportion of Aztec paste figurines. This suggests that some (particularly 307-Ph6 and
309-Ph6), but not all, households at the site were adopting more characteristically Aztec
practices of figurine use, accompanied by an increased use of imported figurines. In
contrast, the total quantities of censers, the relative importance of different functional
censer classes, the frequency of imported Texcoco sahumadors, and the amount of
within-phase variation among households all remain consistent with those seen during
preceding phases at the site. This contrast between the changes in figurine use and the
lack of changes in censer use provides an interesting interpretive issue. The continuation
of local practices of censer use, with their high internal homogeneity, suggests the
continuation of relatively collective local practices of public ritual. The continuity of
censer use is also consistent with relatively indirect rule by the Aztec Empire, with no
effort to enforce widespread public participation in new imperial rituals. In contrast, the
uneven changes in figurine use, which has been previously argued to have few ties to
state religion, suggested more bottom-up reasons for the adoption of non-local practices
in this area.
Third, in terms of specific causes for the cultural changes observed under Aztec
rule, the changes in ritual practices observed under Aztec rule are most consistent with
limited immigration from the Aztec core to Calixtlahuaca. The presence of non-local
cultural practices in lower visibility domains (figurines), paired with the presence of
apparently local practices in higher visibility domains (censers) is consistent with a
“hidden immigration” pattern, where immigrants downplay their foreignness in public
354
settings. This pattern is characteristic of cases where immigrants are a late-coming
minority arriving into an area and are at a power disadvantage relative to the larger host
population. If this is the case at Calixtlahuaca, it remains consistent with an argument of
strongly collective local institutions persisting under relatively indirect Aztec rule.
355
CHAPTER 8
CHAPTER 8. “THE PRINCIPAL FOODS OF THESE WERE TAMALES,
BEANS10”: CULTURAL CHANGES IN FOODWAYS
In this chapter I continue the discussion of change or continuity in cultural
practices at Calixtlahuaca, using evidence from a variety of food preparation and serving
practices. Particular rulership strategies have implications for both the diversity and
potential for change in cultural practices. The directness of rule primarily relates to the
magnitude of cultural change. More direct rule results in more interaction between the
core and provincial areas, as well as increasing the probability that the use of imperialstyle objects or practices will confer benefits in local interactions with imperial agents.
The collectiveness of local and imperial rule both have implications for the degree of
interhousehold variation in cultural practices, with more collective organizations of
power being more likely to produce widely shared practices among households,
especially public practices.
The biological necessity of eating, combined with the high diversity of potential
foodstuffs and food preparation techniques, means that food can be an important
component of creating and maintaining identities. The food preparation and serving
practices of the Toluca Valley diverge noticeably from those associated with the Basin of
Mexico during the Postclassic. As with the ritual activities discussed in the previous
chapter, foodways include both private/low visibility aspects and public/high visibility
10
Florentine Codex: A General History of the Things of New Spain. Book 10: The People. Chapter 29,
Section “The Quaquata, The Matlatzinca, The Toloque”.
356
aspects. The former are more likely to be a result of enculturation, while the latter are
more likely to be subject to active manipulation. More specifically, I compare low
visibility aspects of food preparation with high visibility aspects of food serving. I first
examine food preparation techniques, with a focus on various ways of consuming maize,
using ground stone and ceramic cooking vessel assemblages. These data are then
compared to household variation in serving vessel form and decoration, based on the
decorated portion of the ceramic assemblage from the households under study. I examine
whether households with higher frequencies of imported ceramics are also choosing to
use these ceramics in functionally different ways, as these are different dimensions of
identity formation (See Mills 2016 for an example of different scales of interaction in
different visibility levels of food-related practices)
The results of this chapter generally follow the same patterns seen in the
preceding chapter on ritual practices. During the Dongu and Ninpui phases, households at
Calixtlahuaca have relatively homogenous food preparation and serving practices, which
can be distinguished from those characteristic of other parts of Central Mexico. During
the Yata phases, both food preparation and serving practices shift toward patterns more
characteristic of the Basin of Mexico, but these changes are highly unevenly distributed
among households. Based on the distinction between low- and high-visibility practices,
Yata phase households at Calixtlahuaca include cases of local identity maintenance (324Ph6 and 327-P6), local emulation of Aztec practices (316-Ph6 and 317-Ph6), and either
very strong cases of local emulation or ethnically mixed households (307-Ph6 and 309Ph6). These results are most consistent with relatively collective local rule, paired with
relatively indirect and non-collective Aztec imperial rule.
357
Food and Identity
The saying that “you are what you eat” is true on multiple levels. At its simplest,
the phrase recognizes the basic biological conversion of food into human tissue and
energy. However, the saying is true on an additional level, because eating is also a
cultural practice. Anthropology has long recognized this, with several classic works
dealing with foodways (e.g. Bourdieu 1984; Lévi-Strauss 1969). What we eat, how we
prepare it, when we eat and who we eat with are all powerful markers of identity (Mintz
and Du Bois 2002). Food can be used to mark a wide range of facets of identity,
including ethnic or religious affiliation, gender, age, social status, and wealth (Johnson, et
al. 2011; Jones 2007; Twiss 2012).
People use foodways – either what is eaten or how it is eaten – as identifiers for a
variety of levels of group identity. Even at a band level of organization, many groups
identify subgroups or out-groups by what they eat – piñon eaters, rabbit people, seed
eaters, and so on (Steward 1938). This trend is also present in more complex societies,
where dietary practices may continue to form an important component of group
affiliation, such as kosher food laws in Judaism (Rosenblum 2010). In 20th century
California, the archaeological remains of Chinese and European households can be
distinguished by differences in the ways that they cooked the same raw foods
(Langenwalter 1980). In pre-colonial India, regional cuisines remained highly distinctive,
despite wide similarities in other cultural practices (Appadurai 1988). Similarly,
foodways at the Peruvian site of Pedregal remained similar over time, despite economic
shifts resulting from rule by the Chimú Empire (Cutright 2015). On the other side of the
358
food-and-empire relationship, during the Roman period, the distribution of a number of
non-native plants across northern Europe closely matches the boundaries of the Empire,
and they are rarely found outside of military or urban contexts (Livarda 2011). In
Colonial Central Mexico, Sahagun’s Nahua informants used food practices as an
identifier of ten of the fifteen other cultural groups that they described for him (Berdan
2008). While these answers may have been solicited by a specific question about culinary
practices, they still demonstrate a body of commonly held beliefs about the foodways of
the “other” in the Aztec world.
Food practices may also mark variation within an ethnic group, such as age or
gender. Many groups prohibit certain foods or involvement in certain food preparation
practices to individuals of certain genders or life stages. For example, there are
widespread taboos against what a new mother should or should not eat (Piperata 2008).
Even in cases where such norms are not formalized as taboos, they often still exist as
customary practice. In several Mesoamerican cases, bone chemistry shows differences in
diet between men and women in the same population (White 2005). In Aztec society,
excessive pulque (alcohol) consumption was limited to the elderly (Mendoza 1992
[1541]f. 71r).
Foodways may also mark wealth or status. Wealthy individuals are able to afford
a wider diversity of food items and a higher frequency of foods of high nutritional or
cultural value. In addition, wealthier individuals may also be able to invest more in
elaborately prepared foods. In cases where wealth and high status overlap, the
consumption of such highly valued or elaborately prepared foods may form an important
part of demonstrating and maintaining one’s status. In Mesoamerica, various lines of
359
evidence comparing elites and non-elites often show that elites consumed more meat
and/or maize (Lentz 1991; Somerville, et al. 2013).
In addition to marking identities, food can serve as a means of manipulating or
creating aspects of identities. The most well studied example of this is feasting, where the
competitive feeding of others serves as a means of gaining status (Perodie 2001;
Wiessner 2001). Feasts cover a spectrum from inclusive communal events characterized
primarily by the quantity of food, to exclusive events distinguished by the elaborateness
of the food (van der Veen 2003). Both extremes are means of creating of identities,
though the type of identity being marked is quite different. The degree of inclusivity in
feasting events corresponds well to the degree of collectiveness of rulership, with socially
inclusive rulership strategies requiring widespread food provisioning during feasting
events (Carballo 2015). In other scenarios, the relationship between food and identity can
cause immigrants to adopt aspects of the cuisine of their host culture (Bradby 1997), or
lower social classes to adopt dishes or cooking techniques from higher classes (Miller
1985).
Mesoamerican Foodways
In Mesoamerica, archaeologists are beginning to explore food as a means of
expressing identity and most extant work focuses on food as a status marker in the Maya
region or on feasting as a means of social advancement (e.g. Green 2010; Hendon 2003;
Turkon 2004; White 2005). The following section describes previous work in
360
Mesoamerica on variations in general food consumption practices and dietary use of
maize more specifically.
There is widespread within-site variation in food choices in highland
Mesoamerica by the Classic Period. Most research to date has investigated dietary
differences based on status, usually between commoners and elites. While status is
undoubtedly an important factor in foodways, the lack of investigation of other variables
means that the relative weight of status in food choice cannot be assessed at this time. At
Formative-period Cahal Pech, and Classic-period Lamanai and Altun Ha, isotope studies
of human remains from commoner and elite burials show status-based differences in the
amounts of maize and types of protein consumed (White, et al. 2010). At Classic-period
El Palmillo, faunal remains vary between households and some, but not all, of the
variation correlates to other indicators of household wealth (Haller, et al. 2006). At
Classic-period Copan, paleoethnobotanical remains show that higher-status households
had access to a greater diversity of plant resources (Lentz 1991).
Despite this variation, maize in many forms serves as the backbone of the
Mesoamerican diet. The first domesticated maize appears in Mesoamerica around 5500
BP (radiocarbon years), and was widely dispersed across the entire cultural region by
2500 BP (Blake 2010). While there are no existing isotope studies of the importance of
maize in Postclassic Central Mexico directly, studies from the Maya region for the
Postclassic (White 2005), and earlier time periods in Central Mexico (White, et al.
2004:184), demonstrate that maize was a major component of the diet across
Mesoamerica. While maize might be heavily supplemented by manioc in lowland
environments (Pope, et al. 2001; Sheets, et al. 2012) or by maguey products in arid,
361
highland environments (Evans 1992; Parsons and Darling 2000), such cases occur only
where maize cultivation is less than environmentally ideal. This overall emphasis on
maize corresponds well with an early colonial quantification of native children’s food
intake as increasing numbers of tortillas a day (Mendoza 1992 [1541]:f. 58r-60r),
equating food with maize. This high dependence on maize produced a proliferation of
ways of consuming the grain, including a wide variety of gruels, stews, tamales, and
tortillas, at least some of which were described as characteristic of particular cultural
groups. Changes in maize products are particularly archaeologically useful, because the
production of several different end product food types may be visible archaeologically.
The Basin of Mexico and the Toluca Valley show only minor differences in the
types of food crops that were consumed, but a wider divergence in how these crops were
prepared. This is visible in both ethnohistoric documents, and the archaeology of the two
regions. The Florentine Codex includes two passages on foodways specific to the Toluca
Valley. The first is a general mention that strong pulque was considered characteristic of
the area. The second is a longer quote about food and crops in general. “Nothing grew in
the land of these Quaquata; only maize, beans, amaranth; no chili, no salt. The principal
foods of these were tamales, beans; also their principal drink was fruit atole. Popcorn was
produced right there in their land” (Sahagún 1950-82:Book 10: The People. Pp 182-183).
Archaeologically, the Toluca Valley is characterized by a general lack of comals,
supporting the written assertion that the region consumed maize as tamales rather than
tortillas. This is in direct contrast to most other parts of the Central Mexican highlands,
where comals make up at least 10% of domestic ceramic assemblages. In addition, the
Toluca Valley has a longstanding tradition of decorated ceramic jars, which are quite rare
362
in the Basin of Mexico, suggesting differences in food serving practices. During the Yata
phase at Calixtlahuaca, there are some shifts in food-related practices toward a more
Basin of Mexico pattern. These changes are unequally distributed across households and
investigation of their patterning makes up much of the analyses in this chapter.
Questions for Calixtlahuaca
As presented in the previous chapter, I frame my interpretations of the changes in
these forms of material culture around three questions about local rulership strategies,
Aztec rulership strategies, and possible explanations for the particular patterns of cultural
change seen under Aztec rule.
1. Pre-Aztec Cultural Diversity and Collectivity of Rulership: How culturally
diverse were households at Calixtlahuaca prior to the Aztec conquest of the site?
More collective forms of social organization should generally produce more
homogenous cultural identities, particularly in high-visibility practices and
material culture. If there is substantial cultural variation at the site, I expect to see
distinctive clusters of specific types of material culture, representing ethnically
specific cultural practices, dividing the household components into two or more
distinctive groups during multiple phases.
2. Overall changes resulting from Aztec rulership: What was the overall degree of
cultural change at Calixtlahuaca following the Aztec conquest of the site and does
363
it likely relate to imperial rulership strategies? The magnitude of change toward
activities more characteristic of the Basin of Mexico is a product of the directness
of rule. The evenness of the distribution of any such changes among households
measures the degree to which locally collective institutions persisted or formed
under Aztec rule.
3. Causes of changes under Aztec rule: Can the mechanisms of cultural change in
particular households be explained as either local emulation or foreign
immigration? Cases of emulation are expected to be visible as the appearance of
Aztec material culture primarily in high visibility contexts, conforming to the
“Local Emulation” pattern. In contrast, cases of migration will be visible as the
appearance of specific households featuring Basin of Mexico style household
practices (and associated material culture) in low visibility contexts. These
households may also use Aztec-style material in high visibility contexts, but this
is not required to demonstrate the presence of immigrants. As a result, households
in this category may conform to either the “Hidden Migration” or “Site Unit
Intrusion” patterns of low and high visibility material culture.
Maize Preparation Practices
Maize processing is a low visibility activity. It occurs primarily within the
household, for use within the household. It is a multi-step process, but the effects of prior
364
stages are not usually directly visible in the finished product. I examined maize grinding
practices, which are likely to be almost completely enculturative, and cooking methods,
which are a product of the desired form of maize consumption.
Because maize processing techniques have remained remarkably consistent since
the Spanish conquest, a combination of historic (Landa 1941; Sahagún 1950-82) and
ethnographic documents (González 2001; Pennington 1963; Tozzer 1907; Vogt 1969)
can be used to create operational sequences and potential archaeological correlates for
five maize products: pinole, posole, atole, tamales, and tortillas. These products do not
represent the entire range of ways in which maize is consumed; the Tenejapa Tzeltal
recognize thirty-two named maize foods (Stross 2010:205-206), and Fournier identifies
seventeen possible forms of maize consumption (1998:20-21). They do, however, cover
the most widespread, frequently consumed, and non-seasonally dependent choices of
maize preparation. It should be noted that regional usage of the Spanish labels I am using
is rather variable and that some terms (particularly atole) may be used in the literature to
refer to a much wider range of products than I use them to describe here.
Like any multistep process involving learned skills, maize (or any other food)
preparation practices can be viewed as a chaîne opèratoire. A chaîne opèratoire, or
operational sequence, is the series of operations which transforms a substance from a raw
material into a manufactured product (Cresswell 1976). It incorporates both physical and
cultural constraints on manufacture and sees the resulting item as a combination of the
interplay between the two (Lemonnier 1993). The reconstruction of an operational
sequence allows the investigator to determine the allowable degree of variation within the
production process, on a step by step basis. While in archaeology most commonly
365
applied to studies of lithic reduction (e.g., Sellet 1993), the concept is equally applicable
to any process of manufacture involving decisions among possible alternatives. Jennings
et al. (2005) provide a useful application of the chaîne opèratoire concept to the study of
food production in their comparison of the brewing methods of various ancient forms of
alcohol. Because it incorporates both physical and social constraints, this method is
highly suitable to address the linked social and technological sides of food production.
Each of the five maize products presented above can be seen as the result of a
combination of three processing decisions: whether to soak the corn in an alkaline
solution (nixtamalizacion), how finely to grind the corn, and what cooking methods to
use (boiling, steaming, or toasting) (Figure 8.1). Each decision stage is indicated by
potential archaeological markers.
Figure 8.1 Stepwise processing decisions for five common maize foods
The first decision is whether to soak the corn in an alkaline solution. The practice
of nixtamalizacion involves soaking dry corn kernels in an alkaline (usually lime or ash
366
based) solution to soften them and increase the availability of certain amino acids. This
process may be archaeologically identifiable as a shift from basin metates, which are
more efficient for grinding hard seeds, to legged metates, which are more useful for
grinding soft or oily items (Adams 1999; Biskowski 2008). Legged metates are nearly
ubiquitous in ethnographic studies of Mesoamerican households, and the only type still
documented to be in production today. A single metate may last for more than one
generation, even if in daily use, so stylistic change may be slow.
The second decision is how fine to grind the corn grains. It has been proposed that
an increased fineness of grinding may be identifiable as a reduction in mano weight
(either as thinner simple manos, or as thin manos with bulbous, overhanging handles) as
means of reducing fatigue over extended periods of grinding (Biskowski 2000) , but this
association has not been well established to date. Grinding choices range from no
grinding for posole, to coarse grinding for atole or pinole, to medium grinding for
tamales, to very fine grinding for tortillas. It should be noted that increasing fineness is
primarily a result of increased grinding, rather than grinding differently.
The third decision is how to cook the corn. Boiling and steaming both involve the
use of large cooking pots. In contrast, toasting requires a relatively unique vessel form,
the comal or tortilla griddle. The distribution of comals is highly variable across time and
space in Mesoamerica. While first appearing in the Oaxaca Valley during the Formative
period, there were still large portions of West Mexico and the Maya lowlands that had not
adopted the use of the comal prior to the Spanish conquest (Fournier Garcia 1998). There
is no archaeological evidence for the use of stone baking surfaces, such as the piki stones
of the southwestern United States.
367
Based on the decision-making process for maize foods described above, I use
three indicators relating to maize processing decisions – overall quantities of ground
stone, variation in mano form, and specific cooking vessel frequencies - to investigate
changes in food processing and preparation at Calixtlahuaca. I use the overall ground
stone frequencies and variation in mano form to address the second decision in maize
processing – the fineness of grinding. I then use ceramic data on the frequency of comals
and jars to address the third decision in maize processing – cooking method. For each of
these lines of evidence, I consider the general direction of artifact frequencies at
Calixtlahuaca, the interhousehold variation over time, and the similarity or difference of
the assemblage at Calixtlahuaca compared to other sites in Central Mexico. Due to the
small number of metates recovered at Calixtlahuaca, and the near ubiquity of wetgrinding in Postclassic Central Mexico, I do not investigate variation or temporal
patterning related to the first decision making step.
Ground Stone
The dataset for ground stone analysis consists of mano and metate type counts
and descriptions (Mortensen n.d.). While the project recovered a wide range of ground
stone artifacts, only those relating to the grinding of food are discussed here. Because
ground stone is a rare category, this analysis uses the DS-2 sample. I evaluate both the
total frequencies of grinding tools, and more specific aspects of mano forms.
This expanded household sample includes 47 mano and 21 metate fragments.
Grinding tools were recovered for 13 of the 18 contexts under consideration, and the
dataset is summarized in Table 8.1. Stone grinding tools (manos and metates) occur at an
368
average frequency of about 1 per 2,580 sherds. Two of the four components where no
grinding stones were excavated have far fewer than this number of sherds and the
remaining three are in the bottom half of component sizes, suggesting that the lack of
grinding tools in these units may be due to sample size rather than true absence.
Unit
307
315
316
320
323
324
Phase N. Sherds Metates
Dongu
5,810
Dongu
16,775
Dongu
4,710
1
Dongu
12,189
3
Dongu
26,947
1
Dongu
914
2
303
307
308
311
316
322
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
Ninupi
9,043
22,330
4,359
7,838
22,563
1,855
2
3
3
2
1
1
5
307
309
316
317
324
327
Yata
Yata
Yata
Yata
Yata
Yata
10,257
4,217
10,091
10,860
3,438
1,266
3
3
2
Manos
5
1
1
3
5
6
2
4
9
a
0.55
0.22
0.00
0.13
0.27
0.00
0.88
0.47
0.69
1.01
0.00
0.00
1.48 (.24)a
0.20
0.51
Dongu Average
Ninupi Average
Yata Average
DS-2, All Phases
Grinding
Stones/
1000 sherds
0.00
0.30
0.42
0.33
0.15
7.66
21
47
0.73 (.32)a
Including (Excluding)component 324-Ph2
Table 8.1 Mano and metate counts and frequencies per 1000 sherds by household
component at Calixtlahuaca, DS-2 Sample
369
The average frequency of grinding tools drops dramatically from the Dongu phase
to the Ninupi and Yata phases, but this is caused by 324-Ph2, a highly anomalous context
(Table 8.1). (The reasons for high groundstone levels in this component have been
previously discussed in Chapter 6.) If this context is excluded, the average grinding tool
frequency is not significantly different (t-test of means, .90 level) between the Dongu and
Ninupi phases. The differences between either of the first two phases and the higher Yata
phase values are statistically significant at the .90 level, if small components (<2,500
sherds) are excluded. This latter pattern of comparable values during the Dongu and
Ninupi phases, followed by an increase during the Yata phase is consistent with the
results of ground stone frequencies in the broader, all phaseable contexts DS-3 sample,
indicating that it is a more accurate representation of general temporal trends at the site.
The range of variation in grinding tool frequencies among households (excluding small
components) is relatively similar between the Dongu and Ninpui phases, before
increasing during the Yata phase.
The overall frequencies of grinding stones at Calixtlahuaca and their increasing
frequencies over time both conform to broader Central Mexican patterns (Table 8.2). The
total range of variation in ground stone values seen over the course of Calixtlahuaca’s
occupation falls within the range seen in other regions of central Mexico. There do not
appear to be regional trends in grinding stone frequencies; instead the primary axis of
variation is urban/rural, with urban sites (Yautepec, Chalco) having lower frequencies of
grinding stones. While the relatively small number of comparative sites makes any
broader patterns of ground stone use over time somewhat tentative, it appears that there is
a general trend toward increasing frequencies of grinding stones over the course of the
370
Postclassic. Given the broad intra-regional variation in grinding stone frequencies, as
well as the macroregional trend toward increasing grinding stone use over time, the
increases at Calixtlahuaca are more likely part of broader regional shifts toward more
intensive maize-grinding than a conscious adoption of a specific non-local pattern of
maize processing.
Site
Toluca Valley
Calixtlahuaca
Calixtlahuaca
Calixtlahuaca
Cerro Toloche
Basin of Mexico
Cihuatecpan
Chalco
Morelos
Yautepec
Yautepec
Yautepec
Yautepec
Capilco
Capilco
Capilco
Cuexcomate
Cuexcomate
N.
N.
Phase Manos Metates
Total GS/100 GS/1000
GS Rims Sherds Source
MPC
LPC-A
LPC-B
LPC
15
11
21
13
7
6
8
3
22
17
29
16
LPC
PC
31
3
11
42
3
MPC
LPC-A
LPC-B
COL
MPC
LPC-A
LPC-B
LPC-A
LPC-B
4
17
108
1
5
23
13
12
30
59
2
5
17
25
4
22
131
1
0
15
17
47
84
0.34
0.23
0.63
0.05
0.33
0.25
0.72
1.20 Jaramillo & De la
Peña 2012, p. 27, 59
0.43 Evans 1988, Table 1.2
Hodge 20008, Table
B-3
0.07 Smith 2006a, Tables
0.17 C5-1,3
0.20
0.12
0.00 Smith 2006b, Tables
0.85 B5-1,2,5,6
1.04
0.74
0.48
Note: Postclassic Central Mexican sites have approximately 1/10 rim-to-total sherd ratios, so the
ginding stones per total sherds and grinding stones per rims are offset by a factor of ten to produce
equivalent values between the two data types
Table 8.2 Grinding stone frequencies at Calixtlahuaca and comparative central Mexican
sites
371
Manos. In addition to the general frequency of grinding stone, more specific
attributes of manos can also be used to provide information about the intensity of maize
processing. Manos, the mobile half of a pair of grinding stones, can be used to examine
the fineness of grinding. There are two potential variables of interest related to the
fineness of grinding: the form and thickness of the mano. In addition, I also consider the
cross section of the manos, as this is a product of grinding practices, and may be a
habitus-derived variable.
Regional comparisons of the Calixtlahuaca manos with those found at other
Postclassic sites in Central Mexico are primarily subjective. This is due to the lack of
sites with either adequate samples or comparably coded data. Both of the best analyzed
collections of grinding stones in the Basin of Mexico are from sites with issues of
temporal mixing, Otumba (survey collection) (Biskowski 2000), and Cerro Portezuelo
(reconstructed historical excavations) (Biskowski and Watson 2013), which is why they
are not included in the excavation-based frequencies in Table 8.2. These studies do
provide a baseline for identifying the most common forms of manos and metates in the
Basin of Mexico during the Middle-Late Postclassic period – legged metates and handled
manos. However, they also likely underestimate the amount of variability in Middle-Late
Postclassic grinding tool assemblages, as less finely finished pieces may be assigned to
other chronological periods. The ground stone in Smith’s Morelos excavations are coded
comparably to the data from Calixtlahuaca, but show very different distributions for all
mano-related variables, meaning that comparisons divide the cases regionally (Smith
2006a, 2006c).
372
The general form of mano fragments was coded based on end form, with potential
states ranging from no differentiation between the end and body, to differences in wear
but not shape, to overhanging handles resulting from wear, to intentionally shaped
handles (Figure 8.2, A-E).
Figure 8.2 Mano handle (A-E) and cross-section (1-11) forms
373
At Calixtlahuaca, mano handle form shows very little variation (Table 8.3).
Handle form could not be identified for 48% of mano fragments, primarily because they
were medial sections. Of the remainder, the overwhelming majority in all phases are
simple manos (A), with no visible transition in shape or wear between the handle and
body. No manos had intentionally shaped handles, and only two examples show the
pronounced overhang resulting from grinding with a mano wider than the metate.
Unit
315
316
320
323
324
303
307
311
316
307
309
316
317
Dongu
Ninupi
Yata
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Ninupi
Ninupi
Ninupi
Ninupi
Yata
Yata
Yata
Yata
Handle Form
Total Unkn.
A
B
5
2
3
1
1
1
1
3
2
1
5
5
3
3
2
2
1
1
5
1
2
6
4
2
2
1
1
4
2
2
9
8
1
15
11
21
5
7
15
10
2
5
C
2
2
1
Table 8.3 Mano handle forms by household component at Calixtlahuaca, DS-2 sample.
See Figure 8.2 for illustrations of handle forms.
This is in direct contrast with sites in the Basin of Mexico and Morelos, where
mano forms E, and to a lesser extent C and D, are considered characteristic of the
Middle-Late Postclassic (Biskowski 2000; Smith 2006c). In the previously mentioned
374
studies from Morelos, only these three latter forms were recorded. In a chronologically
mixed Epiclassic-Postclassic ground stone assemblage from Cerro Portezuelo in the
Basin of Mexico, approximately half of the manos have pronounced overhangs
(Biskowski and Watson 2013).
Width (the larger of the two cross-section dimensions) and thickness (the smaller
of the two cross-section dimensions) averages by phase show no consistent patterning
(Table 8.4). The coefficient of variation for width drops steadily over time, primarily due
to a decreasing standard deviation. However, the coefficient of variation for thickness
remains approximately the same through all three periods. Based on two-tailed
Kolmogorov-Smirnov tests, neither dimension’s distribution of values differs
significantly at the .10 level between any pair of phases. This indicates that the range and
distribution of variation in Yata phase manos are not particularly different than those
dating to earlier phases.
Dongu
Ninupi
Yata
All Phases
Mean
Width
5.69
5.13
5.62
5.52
St Dev
Width
1.69
0.92
0.73
1.16
Co Var
Width
0.30
0.18
0.13
0.21
Mean
StDev
Co Var
Thickness Thickness Thickness
4.96
1.39
0.28
4.42
1.11
0.25
4.39
1.29
0.29
4.58
1.29
0.28
Table 8.4 Means, standard deviations, and coefficients of variation for mano thickness
and mano width by phase, DS-2 sample
Manos show a wide range of variation in cross section. Of the eleven possible
cross section shapes (Figure 8.2, 1-11), only seven occur in this sample at Calixtlahuaca
(Table 8.5). Either five or six of these seven are present during each phase. The most
375
common mano cross section is square (Form 4), followed by rectangular (5) and roundedsquare (8), during all three phases. At the same time, the individual household
components have diverse assemblages of cross sections; all components with more than
one mano had examples of more than one cross section form. No cross section form that
occurs more than once is present in only one phase. This wide range of variation in mano
cross sections is comparable to other late Postclassic sites in Central Mexico (Smith
2006a, 2006c). However, the specifics of the forms do differ from most other sites.
Manos from Postclassic sites in Morelos are generally dominated by manos with
rounded-to-oval cross sections (Smith 2006a:Table B5-7; 2006c:Table C5-4), as
compared to the square-to-rectangular pieces that dominate the assemblage at
Calixtlahuaca.
376
Unit
315
316
320
323
324
303
307
311
316
307
309
316
317
Dongu
Ninupi
Yata
Phase
Dongu
Dongu
Dongu
Dongu
Dongu
Ninupi
Ninupi
Ninupi
Ninupi
Yata
Yata
Yata
Yata
Total
5
1
1
3
5
3
2
1
5
6
2
4
9
0
1
15
11
21
1
2
4
1
1
5
1
4
2
2
6
1
8
1
10
1
11
1
1
1
1
1
1
3
1
2
2
1
3
1
1
1
1
3
7
4
7
2
3
4
1
1
1
1
2
1
1
1
1
1
1
2
1
2
2
4
1
1
1
Table 8.5 Mano cross-section shapes by household component at Calixtlahuaca, DS-2
sample. For illustrations of cross-section shapes, see Figure 8.2.
In summary, maize grinding tools at Calixtlahuaca show few changes over time.
While there is a significant increase in the frequency of grinding tools during the Yata
phase, this is part of a broader regional trend toward increased maize processing during
the Late Postclassic, rather than a shift toward imitating the practices of a particular
region. This position is supported by the lack of adoption of the overhanging manos
characteristic of the Late Postclassic in the Basin of Mexico and Morelos. There are no
significant changes in mano forms or dimensions at Calixtlahuaca over time. As manos
wear out faster than metates, any pronounced changes in grinding practices should be
more evident in manos than metates. The degree of interhousehold variability in grinding
stones also remains consistent over time.
377
Cooking Ceramics
The third step of maize processing is cooking, which requires a comal to make
tortillas, and large cooking pots for the other four maize foods. I consider two lines of
evidence for this stage. The first is the frequency of comals, measured both as a
percentage of the total ceramic assemblage and as a percentage of rims. The second
artifact type I consider is interior-incised jars, measured as a percentage of total jars.
Interior incised jars are large, wide-mouthed jars, which makes them especially good
candidates for tamale cooking pots. All three analyses use the DS-1 sample.
Comals. While the overall frequency of comals at Calixtlahuaca is very low
compared to most other Postclassic Central Mexican sites, comal frequencies do increase
by an order of magnitude from each phase to the next at the site (Error! Reference
source not found., Table 8.6). Based on t-tests of means, both increases are statistically
significant at the .90 level. The increases in comals also result in decreasing jar/comal
ratios at the site over time. There were only six comal sherds recovered from Dongu
phase contexts in the DS-1. This low of a quantity could easily have been introduced by
stratigraphic mixing or excavated lots that included small corners of later contexts. As a
result, I consider the inhabitants of Calixtlahuaca to generally not have used comals – or
as a result, eaten tortillas – during the Middle Postclassic. By the Ninupi phase, comals
are present in all household components, though at low levels in most households. By the
Yata phase, comals are present in all households and the sitewide average frequency has
increased. Behaviorally, this represents the introduction of a new form of maize food
preparation, and by the Yata phase, its semi-regular use. Learning the physical skills to
378
successfully make tortillas by hand is fairly difficult and was taught to girls during later
childhood in Postclassic Mesoamerica (Mendoza 1992 [1541]):f.60r.
Evaluating the degree of interhousehold variation in comals over time is
complicated by the rapidly increasing mean values. In one sense, the coefficients of
variation for comals actually decrease from each phase to the next, due to the increases in
mean values. In contrast, the ranges of variation in values increase over time, and in this
case, is probably a better measure of behavioral variability.
Figure 8.3 Comal frequencies as a percentage of total sherds by household component
379
Unit
Dongu
307
315
316
320
323
324
Ninupi
303
307
308
311
316
322
Yata
307
309
316
317
324
327
All Sherds
Total Total
%
Sherds Comals Comals
Rims Only
Total Comal
%
Rims Rims Comals
4,756
13,772
2,684
3,951
8,827
892
2
0
3
0
0
1
0.04
0.00
0.11
0.00
0.00
0.11
411
1,374
290
395
834
77
2
0
0
0
0
1
0.49
0.00
0.00
0.00
0.00
1.30
8,991
20,259
3,724
4,986
15,794
1,668
10
6
27
12
43
1
0.11
0.03
0.73
0.24
0.27
0.06
824
2,127
329
570
1,851
211
2
3
9
9
11
0
0.24
0.14
2.74
1.58
0.59
0.00
10,225
3,096
7,492
9,616
3,754
945
191
70
83
135
7
2
1.87
2.26
1.11
1.40
0.19
0.21
1,130
332
921
1,114
373
97
60
11
24
38
5
0
5.31
3.31
2.61
3.41
1.34
0.00
Dongu Mean
Ninupi Mean
Yata Mean
0.04
0.24
1.17
0.30
0.88
2.66
Dongu SD
Ninupi SD
Yata SD
0.05
0.26
0.85
0.53
1.07
1.83
Dongu CoVar
Ninupi CoVar
Yata CoVar
1.24
1.07
0.73
1.77
1.21
0.69
Table 8.6 Comals as percentages of per total sherds and rim sherds by household
component at Calixtlahuaca, with summary statistics by phase.
380
On a regional scale, the low overall frequency of comals at Calixtlahuaca (and its
sister Toluca Valley site of Cerro Toloche) is anomalous for Postclassic Central Mexico.
At most other contemporaneous sites in the broader region, comals make up 10-15% of
all the sherds, or 15-30% of ceramic rims in an assemblage (Table 8.7). The few cases
with values below this range (Cihuatecpan and Tepexpan) were both smaller sites in
Teotihuacan Valley, where maize production was marginal and many inhabitants focused
on maguey cultivation instead. The low frequencies of comals at these sites may be
indicative of a lower reliance on maize in general, rather than a choice to prepare maize
in forms other than tortillas. In contrast, comals are a rare vessel form at Calixtlahuaca,
accounting for an average of .49% of household total classified ceramics and an average
of 1.28% of household classified rims. The appearance of comals at Calixtlahuaca, and
their increasing values during the Yata phase, does represent the introduction of a new
practice at Calixtlahuaca. However, given their relatively low frequencies, it is likely that
tortilla production was either limited to special events, such as feasts, or was adopted by
households as simply one additional way to prepare maize, rather than replacing prior
maize foods.
381
Site
Toluca Valley
Calixtlahuaca
Calixtlahuaca
Calixtlahuaca
Cerro Toloche
Basin of Mexico
Xaltocan
Xaltocan
Xaltocan
Cihuatecpan
Oxtotipac
Tepexpan
Maquixco
Teacalco
Xometla
Tlateloco
Zacatenco
Chalco, Unit A
Chalco, Unit B
Chalco, Unit B
Morelos
Yautepec
Yautepec
Yautepec
Yautepec
Capilco
Capilco
Capilco
Cuexcomate
Cuexcomate
Puebla
Cholula (Well 3)
Cholula (Midden)
Cholula (Well 1)
Cholula (Well 2)
Period
% Rims
MPC
LPC-A
LPC-B
M-LPC
0.3
0.9
2.7
EPC
MPC
LPC
LPC
MPC
LPC
LPC
LPC
LPC
LPC
LPC
PC
EPC
LPC
0.04
0.2
1.2
0.9
16.3
28.7
29.2
4.7
20.0
9.4
20.9
17.6
11.0
22.0
20.0
25.0
28.0
Source
Jaramillo & De la Peña
2012, Table 2
Brumfiel 2005a, Tables
4.2, 4.3, Phases revised
per Overholtzer 2014
Evans 1988, Tbl. 1.2-4
Parsons 1966, Table 28
13.7
13.7
Gonzalez Rul 1988a
Gonzalez Rul 1988b
Hodge 2008, Tables 9.2-7
15.2
14.6
12.7
16.1
16.7
16.0
14.1
11.3
15.2
Smith 2006c, Table C2-3
17.3
23.1
25.1
MPC
LPC-A
LPC-B
COL
MPC
LPC-A
LPC-B
LPC-A
LPC-B
EPC
EPC
LPC
COL
% All
Sherds
Smith 2006a, Table B2-2
Smith 2006a, Table B2-2
McCafferty 2001, Table
5.12
Table 8.7 Comal frequencies as percentages of total sherds and rim sherds at
Calixtlahuaca and comparative Postclassic sites in Central Mexico, arranged by region.
382
In addition to providing evidence for adoption of a non-local method of serving
maize, comals were themselves sometimes non-local items. Based on the results of INAA
and petrography, the comals found at Calixtlahuaca include both imported and locally
produced vessels (Table 8.8). Of the sixteen comal sherds submitted for INAA, five were
assigned to primary or secondary local groups, three were assigned to Basin of Mexico
groups, and eight were unassigned. No other major ceramic type has this high a rate of
unassigned samples, suggesting that comals required a unique blend of materials. Given
that comals are a fragile vessel form, which is then subjected to repeated heating and
cooling, an unusual composition is not overly surprising. The petrographic results are
similar, with seven sherds assigned to local groups, three to Basin of Mexico groups, and
two remaining unassigned. About half of the petrographic samples were selected from the
INAA unassigned sherds, and were successfully placed into petrographic groups,
suggesting that these are mildly anomalous local pastes rather than a unique source group
not otherwise represented in the INAA samples. All of the comal sherds recovered at
Calixtlahuaca are consistent in rim form with Postclassic comals from the Basin of
Mexico (squared rim), rather than Postclassic comals from the Otomi areas in the
northern portion of the Toluca Valley (rounded rim). Even with sourcing results in hand,
I could not visually distinguish between sherds assigned to local and imported groups,
which suggests a highly accurate transfer of specialized knowledge. This in turn suggests
that the appearance of local-sourced comals at Calixtlahuaca does represent the
movement of individuals with potting knowledge from the Basin of Mexico, whether
they lived at Calixtlahuaca or at another site in the region.
383
INAA
Petrography
Total
16
12
Local
5
7
Aztec
3
3
Unknown
8
2
Table 8.8 INAA and petrography sourcing results by geographic macroregion for comals
Interior-Incised Jars. Interior-incised jars are very large, wide-mouthed, plain
jars, which were deeply scored on the interior of the body prior to firing (Figure 8.4).
While this scoring is generally loosely patterned into parallel or crossed zig-zag lines,
these patterns would not have been visible when the vessels were intact. This indicates
that the scoring was functional rather than decorative. The type also does not occur in any
of the whole vessels in museum collections analyzed from the surrounding area (Smith
2003e, 2005), indicating that it is likely a utilitarian type which was rarely included in
burial assemblages. This vessel form is apparently unique to the Toluca Valley, as it has
not been reported at sites in the Basin of Mexico or Morelos, but does occur at the
Postclassic Toluca Valley site of Cerro Toloche (Jaramillo Lunque and De la Peña
Virches 2014:Table 21).
384
Figure 8.4 Interior-incised jar sherds
The general form of these vessels, combined with the ethnohistoric descriptions of
tamale consumption in the Toluca Valley, make these vessels likely candidates for tamale
steaming pots. The incisions could have helped prevent the lattice of sticks supporting the
tamales above the water in the base of the jar from slipping. The other possible use would
be as pulque brewing vessels, such as occur ethnographically in the Mezquital region
(Fournier Garcia 2007). However, these modern vessels do not have interior scoring,
likely because it would be difficult to clean it between batches of pulque.
Interior-incised jar sherds are rare throughout Calixtlahuaca’s history, making up
about 0.5% to 1.5% percent of plain and eroded jar sherds for all households (Table 8.9).
These percentages are almost certainly an undercount of the proportion of plain jars
385
which had interior incising in the original vessel assemblage. Sherds from the un-incised
portions of incised jars would be coded as plain and the plain-and-eroded category also
includes sherds from the undecorated portions of decorated jars, which would have been
serving rather than cooking vessels. Despite this systematic undercounting, incised jar
sherds are present in all households during all phases. The low overall frequency suggests
that these sherds do not represent a maize preparation activity on the scale of tortilla
preparation (as represented by comals at other Postclassic sites).
The average frequency of interior-incised jar sherds increased between the Dongu
and Ninupi phases, before decreasing to its lowest overall level during the Yata phase.
Based on the t-tests of means, both the earlier increase and later decrease are statistically
significant at the .90 level.
386
Unit
Dongu
307
315
316
320
323
324
Ninupi
303
307
308
311
316
322
Yata
307
309
316
317
324
327
Total Eroded & Interior % Incised, % Incised,
Sherds Plain Jars Incised
Total Plain Jars
4,756
13,772
2,684
3,951
8,827
892
3,900
10,779
2,027
3,166
7,024
732
46
123
16
24
36
10
0.97
0.89
0.60
0.61
0.41
1.12
1.18
1.14
0.79
0.76
0.51
1.37
8,991
20,259
3,724
4,986
15,794
1,668
7,000
14,443
2,917
3,717
11,956
1,162
74
189
34
54
167
17
0.82
0.93
0.91
1.08
1.06
1.02
1.06
1.31
1.17
1.45
1.40
1.46
10,225
3,096
7,492
9,616
3,754
945
6,752
1,980
5,544
7,529
2,754
724
55
6
48
50
20
12
0.54
0.19
0.64
0.52
0.53
1.27
0.81
0.30
0.87
0.66
0.73
1.66
Dongu, Mean
Ninupi, Mean
Yata, Mean
0.77
0.97
0.62
0.96
1.31
0.84
Dongu, St. Dev
Ninupi, St. Dev
Yata, St. Dev.
0.27
0.10
0.35
0.32
0.16
0.45
Dongu CoVar
Ninupi CoVar
Yata CoVar
0.35
0.10
0.58
0.34
0.13
0.53
Table 8.9 Interior-incised jars as percentages of total sherds and of plain/eroded jar
sherds by household component at Calixtlahuaca, DS-1 sample, with summary statistics
by phase.
387
There are also some differences in the amount of intra-phase variation over time,
and this is still of interest. Based on both the range of values and the coefficients of
variation, interhousehold diversity begins at an intermediate level during the Dongu
phase, drops during the Ninupi phase and increases again during the Yata phase.
Interestingly, none of the Yata phase values are within two standard deviations of the
preceding Ninupi phase mean, with one case (327-P6) falling above this range, and the
remainder falling below it. This does suggest a fairly widespread reduction in whatever
practice involved interior-incised jars, though one that occurred at different rates in
different households. In addition, the household frequency of incised sherds is basically
independent of the frequency of comals during the Dongu and Ninupi phases, but shows
a strong negative correlation during the Yata phase (Pearson’s r = -.72) (Figure 8.5). One
possible interpretation of this correlation is that after comals appear in significant
quantities at the site, comals and interior-incised jars represent alternative strategies for
producing maize foods for special events.
388
Figure 8.5 Scatterplot of comal and interior-incised jar sherds, as percentages of all
vessel sherds. Each point is a household component.
Maize Preparation Practices: Results
The analyses of maize preparation practices at Calixtlahuaca provide information
about low visibility aspects of food preparation. The results can be used to address the
three questions for this chapter: local patterns of social organization, changes under Aztec
rule, and the probable causes of any changes under Aztec rule.
First, in terms of local social organization, pre-Aztec maize preparation practices
at Calixtlahuaca are both regionally distinct from those seen in other parts of Central
Mexico, and relatively stable over time. There are relatively few changes in maize
processing at Calixtlahuaca between the Dongu and Ninupi phases. There are no
significant changes in the total quantity of grinding stones or in the forms of manos
between these two phases. There are statistically significant increases in the frequencies
of both comals and interior-incised jars between these two phases, but due to the small
overall quantities of both, the total magnitude of these changes is relatively small. The
389
degree of interhousehold variability also remains similar for artifact types related to
maize processing during these two phases. As would be expected from a basic household
necessity, households (with the exception of the possible special-purpose deposit in 324Ph2) during these phases have relatively similar frequencies of grinding stones. They also
have relatively similar frequencies of comals and interior-incised jars with ranges of
variation of less than .75% of the total ceramics among all of the households dating to
these phases. This is consistent with a single, relatively uniform set of food preparation
practices among households dating to these two phases at Calixtlahuaca. Low visibility
practices, such as maize preparation, are not likely to be actively targeted by rulership
strategies. However, the homogeneity in maize preparation practices among households
and over time at Calixtlahuaca suggests a relatively homogenous pre-Aztec population at
the site.
Second, the period under Aztec rule did see some shifts in maize consumption
practices at Calixtlahuaca. There are statistically significant increases in the overall
quantities of grinding stone and comals, and a significant decrease in the quantity of
interior-incised jars, relative to the preceding phases. The increase in grinding stone
frequencies at Calixtlahuaca is part of a broader Central Mexican trend toward increased
quantities of grinding tools over the course of the Postclassic. The increase in comal
frequencies moves Calixtlahuaca toward a more characteristically Aztec pattern of comal
use, but even the highest-frequency households at Calixtlahuaca remain substantially
below the comal quantities seen at sites in other parts of the Basin of Mexico. The shifts
toward more characteristically Aztec patterns of maize processing are unevenly
distributed among the Yata phase households, with larger ranges of overall variation for
390
comal and interior-incised jars driven by some household values outside of the previously
observed ranges of variation, while others remain at levels characteristic of the preceding
local phase. As a relatively low visibility activity, changes in maize processing are not
likely to be the direct result of imperial strategies. The small, but significant shifts toward
more Aztec maize preparation techniques, paired with the increasing ranges of variation
suggest that individual households at Calixtlahuaca experimented with culinary practices
in different ways. This pattern is consistent with some breakdown in conformity to local
collective social norms and household attempts to individually negotiate their positions
relative to the Aztec Empire’s more network oriented rulership strategies.
Third, the specific causal factors behind the Yata phase changes at Calixtlahuaca
are ambiguous. Generally, maize processing is a low-visibility activity, which would
make changes more indicative of immigration than of emulation. However, within the
various lines of evidence for maize processing, those associated with the least conscious
decision-making – mano cross-section and end form – show the most stability over time.
Additionally, while there is a trend toward more Basin of Mexico-style practices during
the Yata phase, these never reach the level that one or more of the household components
could be confused with one from the Basin. The relatively limited adoption of comals,
and their inverse relationship with the frequency of interior-incised jars, suggests that
tortillas were added to the repertoire of ways of serving maize, rather than replacing prior
ways. The relatively low quantities of both forms of cooking ceramics suggests that both
may have been used for maize preparation for special events such as feasts. Based on
negative evidence, two Yata phase households at Calixtlahuaca, 324-Ph6 and 327-Ph6,
which remain firmly within the preceding Ninupi phase frequency ranges for both comals
391
and incised jars, can be considered cases of neither local emulation nor migration.
However, the presence of locally produced comals, which could not be distinguished
from imported pieces without technical analyses, strongly suggests the migration of some
specialized comal-producers into the area around Calixtlahuaca, though they are probably
not represented in the sample of excavated households.
In general, the lack of change in groundstone forms may be due to their longer use
life, smaller transportation radius, and increased potential for specialized production.
While new ceramic forms could circulate over longer distances and from new producers,
ground stone production may simply have remained in the hands of the same group of
producers throughout the site’s history, thereby limiting the potential for the appearance
of new forms after the Aztec conquest.
Food Serving Practices
As in the previous chapter on ritual, I contrast two aspects of food serving:
serving practices (based on serving vessel form) and potential group affiliation (based on
the use of different local and non-local decorative groups). These are largely independent
classifications, as the primarily local decorative groups occur in multiple vessel forms at
Calixtlahuaca. The analysis of vessel form allows me to identify households that are
serving food in non-local ways. The analysis of decoration identifies both the level of
inter-household variation in decorated ceramic use over the site’s history, as well as
specific households with substantial amounts of non-local ceramics. I use the
combination of these two dimensions of food serving to determine the degree to which
392
ideas about food serving practices are travelling in conjunction with imported serving
wares.
In general, the analyses in this section show that households at Calixtlahuaca were
relatively homogenous during the Dongu and Ninupi phases, with no evidence for
multiple local cultural groups. The analyses of vessel form did not produce consistent
groupings of household components, and the analyses of decoration divided households
primarily by phase. Because food serving is a high visibility activity, households
diverging from the local pattern may be either cases of emulation or immigration.
Serving Vessel Form Variability at Calixtlahuaca
Serving vessel form is a medium-high visibility trait, given that it involves public
aspects of food consumption, but one with less overt opportunities for information
signaling than the decoration on ceramics. Decorated ceramics usually convey stylistic
information, but food serving practices are only visible when they are actively taking
place.
393
Unit
Dongu
307
315
316
320
323
324
Ninupi
303
307
308
311
316
322
Yata
307
309
316
317
324
327
Rim Sherds
N. Sherds Bowls Jars Copas
All Sherds
N. Sherds Bowls
Jars Copas
231
725
165
226
507
47
97.84
96.28
94.55
93.81
92.11
93.62
2.16
3.59
5.45
5.75
7.69
6.38
0.00
0.14
0.00
0.44
0.20
0.00
632
2,129
505
607
1,358
124
83.54
76.61
74.26
83.53
72.39
66.13
16.46
23.34
25.74
16.31
27.54
33.87
0.00
0.05
0.00
0.16
0.07
0.00
524
1,382
209
315
1,029
119
96.37
95.08
96.17
94.92
94.75
94.12
3.63
4.78
3.83
4.76
5.25
5.88
0.00
0.14
0.00
0.32
0.00
0.00
1,406
3,986
595
908
2,678
363
87.20
75.21
80.84
86.01
80.92
68.32
12.80
24.71
19.16
13.88
18.97
31.68
0.00
0.08
0.00
0.11
0.11
0.00
703
207
534
699
187
54
96.59
92.27
94.57
95.99
97.86
98.15
3.13
3.38
3.56
2.72
2.14
1.85
0.28
4.35
1.87
1.29
0.00
0.00
2,010
570
1,359
1,496
482
171
86.07
84.74
81.02
82.55
83.40
76.02
13.83
13.68
17.88
16.38
16.60
23.98
0.10
1.58
1.10
1.07
0.00
0.00
94.70
95.24
95.90
5.17
4.69
2.80
0.13
0.08
1.30
76.08 23.88
79.75 20.20
82.30 17.06
0.05
0.05
0.64
Dongu SD
Ninupi SD
Yata SD
2.05
0.87
2.21
1.99
0.85
0.69
0.17
0.13
1.68
6.74
7.05
3.53
6.77
7.06
3.77
0.07
0.06
0.69
Dongu CoVar
Ninupi CoVar
Yata CoVar
0.02
0.01
0.02
0.38
0.18
0.25
1.35
1.70
1.29
0.09
0.09
0.04
0.28
0.35
0.22
1.37
1.13
1.08
Dongu Range
Ninupi Range
Yata Range
5.73
2.26
5.88
5.53
2.26
1.71
0.44
0.32
4.35
17.42 17.56
18.88 18.88
10.05 10.29
0.16
0.11
1.58
Dongu Mean
Ninupi Mean
Yata Mean
Table 8.10 Relative serving vessel form (Bowl, Jar, Copa) frequencies by rims sherds and
total sherds, by household component at Calixtlahuaca, with summary statistics by phase
394
Food serving assemblages in Postclassic Central Mexico can be broadly grouped
into three categories of vessels: open vessels, restricted vessels and copas. Open vessels
include simple bowls, tripod bowls, dishes, plates, and basins. They can be used to serve
dry foods (tamales, tortillas), or semi-liquid foods, such as stews or salsas. Restricted
vessels include necked jars, pitchers, and tecomates (very rare!). Restricted vessels would
have been used primarily to serve liquids, such as atole, pulque, or cacao-based
beverages. Because many restricted vessels were used for purposes other than serving,
such as cooking or storage, I only include decorated restricted vessels in my analyses.
Copas are specialized goblets used for drinking liquids, such as those listed for restricted
vessels. I use these generalized vessel form categories to allow for both comparisons
among households at Calixtlahuaca and among sites in Central Mexico more broadly.
At Calixtlahuaca, there is little change in the relative frequencies of the three
categories of serving vessel forms over time, with open vessels always making up the
majority of serving vessels (Table 8.10). While both the rim and total sherd based data
sets show a small decrease in the frequencies of jars relative to both bowls and copas over
time, only the rim-based Ninupi-to-Yata phase difference for jars is statistically
significant at the .90 level, based on t-tests of means. This indicates that the majority of
the changes seen over time could easily be due to random chance. There is also little
change in the amount of variation among households over time. The coefficients of
variation either remain both low and very similar (open vessels) or are inconsistent
between the rim and total sherd based frequencies for the form (restricted vessels and
copas). Based on this analysis, food serving practices likely served a socially integrative
395
function at Calixtlahuaca, promoting the maintenance of a collective identity based on
shared practices in both private and public food consumption.
Regional Variation in Serving Vessel Frequencies
In order to determine whether the lack of change in the serving vessel assemblage
over time at Calixtlahuaca is consistent with the maintenance of local practices of food
serving, I compare the serving vessel form frequencies at Calixtlahuaca with those from
other sites in the Basin of Mexico, Morelos, and Puebla. This comparison serves to
establish whether food serving practices were similar across all of Postclassic Central
Mexico, or whether there were alternative, regionally distinctive practices which were not
adopted by the residents of Calixtlahuaca. I compare the serving vessel assemblage at
Calixtlahuaca to those from the Toluca Valley site of Cerro Toloche (Jaramillo Lunque
and De la Peña Virches 2012), the Basin of Mexico sites of Tlateloco (González Rul
1988a), Iztatlan and Hualquila (González Rul 1988b), Xaltocan (Brumfiel 2005b),
Chiconautla and Nonoalco (Elson and Smith 2002), and the rural Teotihuacan Valley
sites excavated by Parsons (1966), as well as the Morelos sites of Yautepec (Smith
2006c), Cuexcomate and Capilco (Smith 2006a) and the UA-1 compound at Cholula
(McCafferty 2001). These comparative cases include a mix of commoner and elite
contexts, but Smith et al. (2003) demonstrated that there is little difference between
commoner and elite households in the proportions of vessel forms within serving
assemblages at sites in Morelos. Due to the way the data are reported for the various sites,
I perform two sets of comparisons, one based on rim sherd counts (Table 8.11) and one
based on total sherd counts (Table 8.12). While the regional split is not exact, the rim
396
sherd based values generally involve comparisons with the northern Basin of Mexico and
Puebla, and the total sherd count based values generally involve comparisons with the
southern Basin of Mexico, Morelos, and other Toluca Valley sites.
In both Calixtlahuaca and all of the comparative cases, open vessels make up the
majority of the serving assemblage. There is a general trend for Toluca Valley sites
(Calixtlahuaca and Cerro Toloche) to have the highest frequency of restricted serving
vessels, followed by the sites in Morelos, the Basin of Mexico, and finally Puebla. The
frequencies of copas are generally lowest in the Toluca Valley, followed by overlap for
the Northern Basin of Mexico, Morelos, and Puebla, and the highest values occurring in
the Southern Basin of Mexico. Copa frequency may be related to a site’s relative ranking
as larger, more urban sites tend to have higher proportions of copas than rural sites in the
same region.
In the rim-sherd based analysis the cases divided strongly geographically, in both
the hierarchical and k-means cluster analyses (Table 8.11, Figure 8.6). The two methods
also produce similar lower-level clustering. The primary division occurs between the
cases from Calixtlahuaca and all of the comparative cases, and the second division is
between the first two and the third phase at Calixtlahuaca. The comparative cases then
divide along roughly urban/rural lines with a general division between the rural
Teotihuacan Valley sites and urban Xaltocan and Cholula. This suggests that Late
Postclassic-B Calixtlahuaca does represent a relatively large shift in serving vessel use,
on a regional scale, but not one that specifically matches a comparative region.
397
Period Bowls
Toluca Valley
Calixtlahuaca
Calixtlahuaca
Calixtlahuaca
Basin of Mexico
Xaltocan
Xaltocan
Xaltocan
Oxtotipac
Tepexpan
Maquixco
Teacalco
Xometla
Puebla
Cholula (Well3)
Cholula (Midden)
Cholula (Well1)
MPC
LPC-A
LPC-B
K-Means Clusters
Jars Copas 2 3 4 5 Source
94.70 5.17
95.24 4.69
95.90 2.80
0.13
0.08
1.30
1
1
1
1
1
3
1
1
3
1
1
3
Brumfiel 2005a,
Tables 4.2-3,
Phases per
Overholtzer 2014
Parsons 1966,
Tables 28-40
EPC
MPC
LPC
MPC
LPC
LPC
LPC
LPC
99.60
98.69
98.32
99.73
98.34
99.37
99.36
99.51
0.00
0.00
0.00
0.27
0.91
0.56
0.56
0.24
0.40
1.31
1.68
0.00
0.76
0.07
0.08
0.24
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
4
2
2
4
2
4
4
4
4
5
5
4
2
4
4
4
EPC
EPC
LPC
98.20 0.00
98.29 0.00
99.55 0.00
1.80
1.71
0.45
2
2
2
2
2
2
2
2
4
5 McCafferty 2001,
5 Table 5.12
4
Table 8.11 Serving vessel form (Bowl, Jar, Copa) frequencies from rim sherd counts at
Calixtlahuaca and comparative sites, with K-means clustering results for 1-5 clusters.
Bowls include both plain and decorated vessels, Jars include decorated vessels only.
398
Figure 8.6 Rim sherd based hierarchical clustering results for serving vessel forms at
Calixtlahuaca and comparative sites
In the total sherd based analysis (Table 8.12, Figure 8.7), the hierarchical and kmeans cluster analyses produce the same general pattern. Both methods first divide sites
in the Toluca Valley (Calixtlahuaca and Cerro Toloche) from all of the comparative
cases. Both methods then generally divide cases from the Basin of Mexico from those in
Morelos, though this is not a perfect division.
399
Figure 8.7 Total sherd count based hierarchical clustering results for serving vessel
forms at Calixtlahuaca and comparative sites
400
Toluca Valley
Calixtlahuaca
Calixtlahuaca
Calixtlahuaca
Cerro Toloche
Basin of Mexico
Chiconautla
Nonoalco
Iztatlan
Hualquila
Tlateloco
Morelos
Yautepec
Yautepec
Yautepec
Capilco
Capilco
Capilco
Cuexcomate
Cuexcomate
Phase
Bowls
MPC
LPC-A
LPC-B
M-LPC
76.08
79.75
82.30
83.28
MPC-LPC
LPC
LPC
LPC
LPC
MPC
LPC-A
LPC-B
MPC
LPC-A
LPC-B
LPC-A
LPC-B
Jars Copas
23.88
20.20
17.06
16.72
K-Means Clusters
2 3 4 5
0.05
0.05
0.64
0.00
1
1
1
1
1
1
1
1
1
1
1
1
95.33
88.09
89.13
92.32
90.00
2.84 1.84
8.03 3.88
0.00 10.87
1.24 6.43
0.20 9.80
2
2
2
2
2
3
2
2
2
2
4
3
2
2
2
93.32
94.84
92.89
93.66
96.21
96.52
96.34
97.61
6.55
4.71
6.33
6.10
3.11
3.04
3.13
1.93
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
4
4
3
4
4
4
4
4
0.14
0.45
0.77
0.24
0.68
0.44
0.53
0.45
Source
1
1
1
1 Jaramillo & De la
Peña 2012, Table 2
5 Elson & Smith
4 2002, Table 2
3 Gonzalez Rul
2 1988b, Tbl 1
3 Gonzalez Rul
1988a, p 186-7
5 Smith 2006c, Table
5 C2-3
4
5 Smith 2006a, Table
5 B2-2
5
5 Smith 2006a, Table
5 B2-2
Table 8.12 Serving vessel form frequencies from total sherd counts at Calixtlahuaca and
comparative sites, with K-means clustering results. Bowls include both plain and
decorated vessels, Jars include decorated vessels only.
Serving Vessel Form Analysis: Results
The results of the analyses of serving vessel form relate to activities at a medium
level of visibility – higher than food preparation, but lower than ceramic decoration. The
results of these analyses demonstrate that serving vessel use at Calixtlahuaca was similar
over time, both in terms of the relative proportions of different forms of serving vessels,
and in the low amount of variation among contemporaneous households. This pattern is
consistent with a relatively collective local organization of power, which was not
substantially affected by Aztec rule.
401
On a regional level, serving vessel use does vary noticeably among major regions
of Central Mexico, though site rank may be a confounding variable. Serving vessel use at
Calixtlahuaca is distinctive throughout the site’s history, with the only case ever grouped
with Calixtlahuaca in any of the cluster analyses being the nearby Toluca Valley site of
Cerro Toloche. However, while the Yata phase changes in serving vessel use at
Calixtlahuaca were not statistically significant due to the small number of households in
the sample, the shifts are toward a pattern more characteristic of other regions of Central
Mexico, though not necessarily the Basin of Mexico in particular. This is again consistent
with highly locally-driven decisions about which aspects of Aztec culture to adopt and
how.
Analysis of Decorative Groups
Decorated ceramic serving vessels are the highest visibility form of material
culture under consideration in this chapter. Because of this, they are the most likely to be
consciously used to create and manipulate identities, such as Bowser (2000) demonstrates
in her study of pottery decoration in Amazonian groups. I analyze decorated ceramics at
Calixtlahuaca to determine the degree of interhousehold variation over time, and the
degree to which individual households were displaying connections with non-local
regions. The discussion of non-local groups draws heavily on the prior analysis of
exchange in Chapter 4. This consideration of decoration complements the preceding
analysis of vessel form, as decoration and form are cross-cutting categories at
Calixtlahuaca.
402
Due to the flexibility of the ceramic classification system used at Calixtlahuaca,
the ceramic types used in actual classification can be combined in several ways. One of
these is to group types into decorative families, which approximate the way “ware” is
used as a category in some other systems. All types grouped into a decorative family have
stylistically similar decoration in similar colors. Decorative families cross-cut vessel
forms; most occur in both bowl and jar forms, and some also include miscellaneous other
vessel forms. Strictly speaking, decorative families do not take paste into consideration,
though the correlation between decoration and paste mean that most decorative families
only include pieces from one INAA or petrographic macroregion (Local, Basin of
Mexico, S/SW of the Toluca Valley, Other).
The decorated serving ceramics at Calixtlahuaca can be divided into eleven
decorative families on the basis of surface decoration (Table 8.13, Figure 8.8, Figure 8.9).
These are local families A, B, C, and E, Aztec families Aztec Black-on-Orange, Aztec
Black-on-Red, and Other Aztec, and S/SW Toluca Valley families D, Incised Red,
Negative, and a miscellaneous other non-local decorated types category. While these
decorative families generally parallel the source regions discussed in Chapter 4, there is a
moderately higher degree of geographic mixing within decorative families due to local
imitations (which are sorted with their model type here) and families where types fell into
multiple source regions (Families C, D). Table 8.14 presents the percentage of each group
out of the total decorated ceramics for that household component. Because these values
only include the decorated portion of the ceramic assemblage, they minimize any
variation due to household crafting activities involving plainwares (which would decrease
decorated percentages across the board) or wealth (which might increase the proportion
403
of decorated ceramics relative to the assemblage as a whole). As a result, it measures how
households chose to divide their servingware purchases among stylistic options, relative
to their purchasing ability.
Family
Local
A
B
Description
Forms
Red-rim on buff
Local redwares. Black or Black and White on Red.
Excludes B-types sourced to other regions.
Local polychrome. Red/Black/Brown on White
Local Red on Buff with line segments
Bowls
Bowls, Jars, Pitchers
C
E
Basin of Mexico
Black on Orange Aztec-style B/O regardless of paste source
Black on Red
Aztec-style guinda polished redwares
Other Aztec
Misc. other Aztec decorative styles, mostly polychromes
South/Southwest of Toluca Valley
D
Red or Red/Brown on White, various styles
Incised Red
Incised Redwares (B-5 and B-11 types)
Negative
All types with negative decoration (Neg/Buff,
Neg+Red/Buff, Neg+Red/White)
Other
Other Imported All other non-local and non-Aztec decorative types
Bowls
Bowls, Jars, Copas
Bowls
Bowls, Pitchers, Copas
Bowls, Jars
Bowls, Pitchers
Bowls, Jars, Pyriform
Vessels
Bowls, Jars
Bowls, Jars
Table 8.13 Major ceramic decorative groups and associated vessel forms at
Calixtlahuaca
The general trend in interhousehold variation is toward decreasing variation
between the Dongu and Ninupi phases, followed by an increase in variation between the
Ninupi and Yata phases. The coefficients of variation reduce for eight out of the eleven
decorative families between the first two phases of the site’s history, and nine out of
eleven then increase between the final two phases of the site’s occupation. The decrease
in diversity between the first two phases is primarily driven by increased access to nonlocal types. All households dating to the Dongu and Ninupi phases have all of the local
404
decorative families present, but are strongly dominated by local E family ceramics, with
the local B family as the second most common group. However, the ubiquity of the
Aztec-associated ceramic families increases between the two phases, and access to
imports from other regions remains about the same. The increase in variation between the
Ninupi and Yata phases is driven both by more variation in the relative importance of
local groups – two households reverse the positions of B and E family frequencies – as
well as more uneven access to imports from all regions. This pattern of increasingly
uneven access to imports has been previously discussed in Chapter 4.
Figure 8.8 Primarily Local Ceramic Decorative Families
405
Figure 8.9 Primarily Non-Local Ceramic Decorative Families
The major decorative family groups were clustered using hierarchical (Figure
8.10) and k-means (Table 8.14) clustering procedures, to identify whether there were
longstanding group divisions within the households and/or whether there were specific
406
outlier households driving the increased variability during the Yata phase. No
comparative cases from outside of Calixtlahuaca were included in this analysis.
Decorative ceramic assemblages in the Toluca Valley and the Basin of Mexico are highly
distinctive. As no case at Calixtlahuaca even approaches 50% Aztec decorated ceramics,
any attempt to include comparative cases would simply divide them based on region of
origin.
Figure 8.10 Hierarchical Clustering Results for Ceramic Decorative Family Frequencies
The two clustering methods produced similar results. Both methods first separate
four of the Yata phase households (307-Ph6, 317-Ph6, 309-Ph6, and 316-Ph6), though
the hierarchical clustering removed the cases one by one, in the order listed, while Kmeans separates them as a single group. Both methods also divide the remaining cases
407
along roughly temporal lines, generally separating the Dongu (MPC) cases from the
Ninupi (LPC-A) and remaining Yata (LPC-B) cases at higher cluster solutions. In the kmeans clustering, one of the previously separated cases, 316-Ph6 flips back to join the
other LPC-A and B cases. Both clustering methods temporally flip the same pair of cases;
303-Ph4 is placed in the Dongu cluster, and 320-Ph2 is placed in the Ninupi/Yata cluster.
At higher numbers of clusters, k-means splits off one or two cases at a time from the
Ninupi/Yata cluster, while otherwise leaving the existing clusters intact. The further
subdivisions produced by the hierarchical clustering do not provide additional useful
information; the first two cases separated out of the primarily Ninupi cluster are the
cross-grouped Dongu cases, and one of the two Yata cases. The hierarchical clustering
divisions below this level do not split along similar variable lines in the primarily Dongu
and primarily Yata clusters suggesting that longstanding group affiliations prior to Aztec
rule are not visible in this analysis.
The results of the cluster analysis parallel those produced by the general
considerations of inter-phase variability. The Dongu and Ninupi phase cases are
relatively uniform within each phase. In the Yata phase cases there is a more pronounced
division between the three cases which form a distinctively Aztec-oriented group and the
two cases which continue along predominately local lines.
408
Unit
Dongu
307
315
316
320
323
324
Ninupi
303
307
308
311
316
322
Yata
307
309
316
317
324
327
A
Local
B
C
E
Basin of Mexico
S/SW St. Mex.
Other K-Means Clust.
AzOR AzRED OTH IncRED D NEG Import 2 3 4 5
8.37
1.62
3.31
2.06
1.20
4.30
16.33
12.67
17.17
31.56
11.74
6.45
2.39 68.13
1.38 81.15
3.31 74.10
1.18 61.95
2.51 77.60
1.08 77.42
0.40
0.08
0.00
0.29
0.00
0.00
0.80 0.00
0.08 0.24
0.30 0.00
1.47 0.00
0.00 0.00
0.00 0.00
0.40 0.80
0.89 1.06
0.00 0.90
0.29 0.88
0.96 1.20
1.08 3.23
1.99
0.16
0.60
0.29
3.83
6.45
0.40
0.65
0.30
0.00
0.96
0.00
1
1
1
1
1
1
3
3
3
1
3
3
3
3
3
1
3
3
3
3
3
1
3
3
1.52
5.24
8.80
3.23
2.86
5.85
17.65
23.96
23.59
27.05
26.96
26.83
2.66 71.54
3.52 63.72
4.58 59.15
1.24 61.29
3.99 58.79
3.90 52.68
0.76
0.27
0.70
3.47
2.93
5.85
0.76 0.00
1.41 0.12
2.11 0.00
0.25 0.25
2.00 0.07
3.41 0.00
3.80 0.57
0.35 1.06
0.00 0.35
0.99 0.74
0.27 1.40
0.00 0.00
0.00
0.20
0.35
0.74
0.00
0.49
0.76
0.16
0.35
0.74
0.73
0.98
1
1
1
1
1
1
3
1
1
1
1
1
3
1
1
1
1
4
3
1
1
1
1
4
4.32
2.20
8.83
7.80
3.15
3.03
35.26
40.66
22.20
27.23
24.77
25.25
1.37 37.58
0.37 20.15
2.86 46.90
1.43 26.11
0.45 64.41
2.02 67.68
15.58 5.05 0.00
18.32 15.75 1.83
10.74 7.52 0.00
20.06 6.69 0.00
4.05 1.35 0.00
1.01 1.01 0.00
0.11 0.42
0.00 0.73
0.12 0.60
0.00 0.80
1.80 0.00
0.00 0.00
0.32
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.24
9.87
0.00
0.00
2
2
2
2
1
1
2
2
1
2
1
1
2
2
4
2
1
1
2
5
4
2
1
1
Dongu Mean 3.48 15.99 1.98 73.39 0.13
Ninupi Mean 4.58 24.34 3.32 61.20 2.33
Yata Mean
4.89 29.23 1.42 43.80 11.63
0.44 0.04
1.66 0.07
6.23 0.31
0.60 1.34 2.22
0.90 0.69 0.30
0.34 0.42 0.05
0.38
0.62
1.69
Dongu SD
Ninupi SD
Yata SD
2.65 8.53 0.90 7.12
2.61 3.63 1.20 6.26
2.76 7.16 0.95 19.57
0.17
2.16
7.78
0.59 0.10
1.12 0.10
5.38 0.75
0.43 0.93 2.49
1.46 0.50 0.29
0.72 0.35 0.13
0.37
0.30
4.01
Dongu CoVar 0.76 0.53 0.46 0.10
Ninupi CoVar 0.57 0.15 0.36 0.10
Yata CoVar
0.56 0.24 0.67 0.45
1.35
0.93
0.67
1.33 2.45
0.68 1.37
0.86 2.45
0.72 0.69 1.12
1.62 0.73 0.98
2.13 0.83 2.45
0.97
0.49
2.38
7.17 25.11 2.24 19.21 0.40 1.47 0.24
7.28 9.40 3.34 18.85 5.58 3.17 0.25
6.63 18.46 2.50 47.53 19.05 14.74 1.83
1.08 2.43 6.29
3.80 1.40 0.74
1.80 0.80 0.32
0.96
0.82
9.87
Dongu Rng
Ninupi Rng
Yata Rng
Table 8.14 Ceramic decorative group frequencies by household component, with KMeans clustering results and summary statistics by phase. For decorative group
descriptions see Table 8.13.
409
Analysis of Decorative Groups: Results
These analyses of major decorative grouping of ceramics provide information
about the trajectories of cultural change under both local and Aztec rulership at
Calixtlahuaca. As an analysis of very high visibility material culture, it provides
information on how the residents of the households at Calixtlahuaca chose to present
themselves to the world around them.
First, there were not pronounced divisions in social identity spanning the site’s
entire history. Time is by far the most significant factor in producing the clustering
pattern seen in the results. In addition, subdivisions within the major temporal clusters do
not consistently divide the clusters based on the same set of variables. This suggests that
either there were not strongly marked social groups at the site, or that there were enough
groups that many of them are represented by single households in the excavated sample.
For the majority of types, the interhousehold variability decreases between the Dongu
and Ninupi phases. This is consistent with a relatively collective local organization of
power resulting in the benefit of increased interregional access being relatively evenly
distributed across the population, and individual households seeking to demonstrate
similar, rather than distinctive, foreign ties.
Second, Aztec rule during the Yata phase saw changes in both the content and
interhousehold variation in decorated ceramics. On average, Yata phase households have
less local E family ceramics and imports from the south-southwest State of Mexico, and
more local B family ceramics and imports from the Aztec area. The increase in local B
family ceramics may be linked to the rise in the use of Aztec wares, as of all the local
decorative families, B is the closest to types produced in the Basin of Mexico. This
410
overall shift toward an increased use of Aztec-affiliated decorative groups is unevenly
distributed across households, which creates an increased degree of interhousehold
variation for most types relative to the preceding Ninupi phase. Of the Yata phase
household components, two remain firmly grounded in the pattern characteristic of the
preceding phases (324-Ph6, 327-Ph6), three trend very strongly toward the new highAztec pattern, though in individually distinctive ways (307-Ph6, 309-Ph6, and 317-Ph6),
and one is an intermediate case (316-Ph6). Due to the high visibility of decorated
ceramics, households showing an increased preference for Aztec ceramics may be doing
so either due to emulation or immigration. However, in either case, the increase in
diversity in decorated ceramics indicates less homogenous social identities at the site, as
individual households pursued different strategies vis-à-vis the Aztec Empire.
Conclusions
This chapter has presented data on food related practices at Calixtlahuaca over
time, particularly maize processing and general food serving practices, with the goal of
evaluating how changes in local identities related to broader changes in rulership at the
site. This evaluation was framed around three questions: the pre-Aztec pattern of
rulership and trajectory of cultural change at the site, the pattern of cultural change under
Aztec rule and its implication concerning rulership, and potential specific mechanisms of
change under Aztec rule. In this section, I combine the evidence on foodways with that
on ritual practices from the previous chapter. The results of the analyses in this chapter
411
are summarized in Table 8.15 and can be compared to the results of the prior chapter in
Table 7.10.
Ground Stone
Dongu-Ninupi
Shift toward
Basin-pattern
Evenness of
variation
Ninupi-Yata
Shift toward
Basin-pattern
Evenness of
variation
Cooking Ceramics
Quantity
Use
Use
No shift
No shift
Yes, small increase in
comal use
n/a
No shift?
Moderately uneven
Increases,
but no
No shift
n/a
No shift?
Yes, additional
increase in comal use
High degree of
variability
Serving Vessels
Use
Decoration
Increasing nonlocal ceramics
Low varation
Increased
both phases
eveness
No shift
No shift
Low varation
both phases
Increasing
stylistically
Increasing
variability
Table 8.15 Summary of results of analyses of foodways showing shifts in interhousehold
variation and local vs. Basin of Mexico cultural patterns under local and Aztec imperial
rule
First, in terms of local patterns of rulership, while Calixtlahuaca was under local
rule, foodways were both stable over time and relatively homogenous among households.
In terms of maize processing, general grinding stone frequencies and more specific
aspects of mano form remain comparable. Comal and interior-incised jars both show
small increases in frequency between the Dongu and Ninpui phases. Serving vessels
show some minor shifts in preferred decorative families, particularly among imported
wares, but continue to be used in similar ways, based on vessel form ratios. The degree of
interhousehold variation is low for most variables, and either remains comparable
between the two phases (total ground stone, mano forms, comals, serving vessel forms),
or reduces (interior-incised jars, decorative families). There is no evidence for multiple
412
local subgroups during these phases, based on the lack of similar subdivisions among
households over time or across multiple artifact types within a phase.
Multiple aspects of food processing and serving, including different mano shapes,
a general lack of comals, and high proportions of jars in serving vessel assemblages,
differentiate the Toluca Valley from other parts of highland Central Mexico during these
phases. As a result, temporal and interhousehold stability in foodways in the Toluca
Valley occurred despite the knowledge of other potential patterns of maize preparation
and food serving in adjacent areas, with whom the residents of Calixtlahuaca were
trading. The patterns seen in foodways over time match well with the results of the
previous chapter which found that ritual practices also had few inter-phase differences,
low interhousehold variability, and distinctive regional practices during the Dongu and
Ninupi phases. The low interhousehold variability and the trend toward increasingly even
access to non-local ceramics as these became more widely available, suggests a relatively
collective social organization of power at the site prior to the Aztec conquest.
Second, in terms of changes in rulership under Aztec rule, the Yata phase does see
some changes toward more Aztec foodways and an increase in interhousehold variability
at the site. Some aspects of foodways (increased comals, reduced use of decorated
serving jars, more use of Aztec-style ceramics, possibly total grinding stone frequencies)
move toward patterns more characteristic of the Basin of Mexico, while others remain
similar to the local patterns characteristic of the preceding phases (mano forms and
dimensions, interior-incised jar use). However, the changes fall far short of reaching
levels where the values seen at Calixtlahuaca could be confused with those from the
Basin of Mexico. The changes toward more Aztec foodways are also unevenly
413
distributed among households at Calixtlahuaca, leading to increased interhousehold
variation during the Yata phase. Both the pattern of shifts toward a more Aztec pattern of
use in some but not all aspects of material culture, and the increase in interhousehold
variability, are the same as the patterns seen in ritual items in the preceding chapter. The
incomplete nature of the shift toward Aztec practices at Calixtlahuaca is consistent with
relatively indirect rule by the Aztec Empire. The increasing interhousehold variation is
consistent with a less collective imperial rulership strategy, where households could seek
individual gains from fostering particular imperial connections.
Third, in terms of specific causal mechanisms for change under Aztec rule, the
uneven distribution of Aztec practices and objects among Yata phase households at
Calixtlahuaca can be used to evaluate the potential causes for their appearance,
particularly local emulation and the immigration of people from the Basin of Mexico. Of
the practices considered in this chapter, visibility increases from maize grinding to maize
cooking to food serving practices to food serving vessels. In cases of migration, a nonlocal pattern should be present in relatively low visibility practices and may also be
present in higher visibility activities. The maintenance of high visibility traits among
migrants will be more common where distinct group identities benefit either the host or
immigrant groups. In contrast, in cases of local emulation, local households should adopt
primarily high visibility aspects of non-local culture and/or use foreign object in local
ways. In the previous chapter on ritual practices, three households (307-Ph6, 309-Ph6,
and 317-Ph6) had practices consistent with immigrants downplaying their differences
from the local population (Aztec low visibility practices, but relatively local high
visibility practices), two households remained consistent with prior local practices (324414
Ph6 and 327-Ph6), and one household was intermediate between the two groups,
potentially interpreted as either an ethnically mixed household or a case of local
emulation.
However, the picture presented by the foodways based evidence is less clear. The
general division remains between households that have relatively high shifts toward
Aztec style practices or objects in some domains and those that do not show shifts in any
domains. Which cases assign to which group are almost identical to those produced by
the analyses of ritual items. However, the lines of evidence which do show changes are
not cleanly divided along lines of low versus high visibility or objects versus practices. In
food preparation practices, maize grinding does show increases in overall intensity, but
these are not associated with changes in mano form. Some households partially replace
interior-incised jars, likely used for tamale production, with comals for tortilla
production, but the latter remains at levels far below what is seen in other parts of Central
Mexico. In terms of food serving, some households shift toward using substantial
amounts of Aztec style ceramics and their closest local analogs, but this shift is not
accompanied by a change in the food serving practices that these vessels were used for.
As a result, when food serving practices are considered in conjunction with ritual
practices, there are two households that retain local practices and generally have low
levels of Aztec style objects as well (324-Ph6 and 327-Ph6). These are best interpreted as
traditionalist local households that actively avoided integrating Aztec practices into their
daily lives. Additionally, there are two households that have more mixed use of Aztec
objects and practices, with either intermediate adoption of Aztec practices, or the
adoption of some practices but not others (316-Ph6 and 317-Ph6). These are interpreted
415
as cases of local emulation, with Aztec objects and practices integrated into local
contexts. Finally, there are two households which consistently show the most Aztec
practices and the most Aztec objects, for markers where there is any shift toward a Basin
pattern (307-Ph6 and 309-Ph6). The interpretation of these two cases is ambiguous, as
they may be either aggressive cases of local emulation, well integrated cases of
immigration, or ethnically mixed households. In terms of implications for rulership, the
variability among households suggests relatively indirect rulership, as some households
successfully avoided almost all changes toward more Aztec lifestyles. Both highly
variable local emulation or highly integrated immigrants would support an argument for
relatively network oriented rule by the Aztec Empire. Either local households were
variably successful in maintaining patronage relationships, or the empire was not
sufficiently concerned with immigrants that they felt comfortable maintaining distinctive
identities.
The variability among the Yata phase households at Calixtlahuaca demonstrates
the complexity of social processes operating in Postclassic Central Mexico. While there
is an overall trend toward increased interaction with the Basin of Mexico over time, such
interaction varied widely on a household-by-household level. Some households may have
been occupied by immigrants from the Basin of Mexico who brought Basin-style
practices and social connections with them. Among the local residents of the site, some
chose to integrate some Basin-style practices into their lifestyles. Others chose to actively
reject Basin-style practices, emphasizing local practices with strong ties to preceding
periods at the site. As a result, it is overly simplistic to speak of cultural changes at
416
Calixtlahuaca as being strictly driven by immigration or emulation or resistance, as all
three processes were occurring simultaneously.
417
CHAPTER 9
CHAPTER 9. CONCLUSIONS
In this dissertation I set out to determine the effects of Aztec rule on provincial
commoners, using the site of Calixtlahuaca as a case study. The Aztec empire has
traditionally been considered a hegemonic state, which interfered little with the day-today lives of its members (Hassig 1988). Under this perspective, the effects of Aztec rule
can generally be characterized as slightly negative due to an increased tax burden. More
recently, this perspective has been challenged by comparative studies of collective action
in ancient states (Blanton and Fargher 2008; Fargher and Blanton 2007), which found
that the Aztec Empire was relatively collective, with associated services provided to the
greater population. Archaeologists have also argued whether the presence of the Aztec
Empire was exploitative, imposing imperial ideology along with taxes, or positive,
providing increased economic opportunities and cultural connections (Nichols, et al.
2009). Much of the difficulty in addressing these questions has been due to the issue of
separating the effects of Aztec rule from those of broader processes of cultural and
economic integration occurring during the Postclassic.
I structured my investigation of the effects of rulership at Calixtlahuaca along two
axes of variation: the directness and the collectiveness of rule. The directness of rule
measures the degree to which rulers chose to directly administer the lives of the ruled, as
opposed to working through proxies (Gerring, et al. 2011; Hassig 1988). In the Aztec
case, the empire had a choice between establishing its own bureaucracy in provincial
areas and leaving local rulers in place as proxies. The directness of rule was generally
418
expected to correlate with the magnitude of changes occurring after the Aztec conquest of
Calixtlahuaca.
The collectiveness of rule measures the degree to which leaders were accountable
to their subjects (Blanton and Fargher 2008; Fargher and Blanton 2007; Fargher, et al.
2011). Under more collective rulership, rulers build powerbases by inclusion, promoting
group solidarity across social classes, and providing public services. More collective
rulership may be linked to economic development, as taxes are often collected in staple
goods, leading the state to have a vested interest in the development of a market system.
In contrast, in less collective (network-oriented) rulership strategies, leaders’ power is
based on exclusive control of material or spiritual sources of power and patron-client
social networks. At Calixtlahuaca, I expect the collectiveness of rule, both local and
Aztec, to be visible as the degree of variation among contemporaneous households, with
more collective rulership resulting in greater interhousehold similarity. More collective
rule is also expected to result in a general trend toward economic growth and increased
market integration.
Based on the factors which these two dimensions of rulership share, there is likely
to be some degree of correlation between the two dimensions. Relatively indirect rule is
almost always relatively non-collective, depending heavily on provincial elites. In
contrast, more direct rule can occur along a continuum of collectiveness. Additionally,
the preexisting local organization of power may differ from imperial strategies of
rulership, leading to mixed strategies at different levels of control. I argue that this was
the case at Calixtlahuaca, with relatively collective local rulership overlaid by an
additional layer of relatively indirect, non-collective imperial rulership.
419
I address these two dimensions of variation more specifically by examining the
economic and cultural impacts of the Aztec conquest at Calixtlahuaca separately, before
combining the two lines of evidence. I compared the periods prior to the formation of the
Aztec Empire (Dongu phase/Middle Postclassic), prior to the Aztec conquest of
Calixtlahuaca (Ninupi phase/Late Postclassic-A) and after the Aztec conquest of the site
(Yata phase/Late Postclassic-B), with a sample of six excavated households from each
phase. The comparisons of the first two phases provide a pre-Aztec baseline for the social
organization of power and general trajectories of change under local rule. The
comparisons of the final phase to this baseline provide a measure of the changes resulting
from Aztec rule.
My results show that the Aztec Empire did have a noticeable effect on conquered
populations, independent of broader, preexisting Postclassic trends toward macroregional
economic and cultural integration. Prior to Aztec rule, regional economic integration and
commoner wealth levels were increasing at Calixtlahuaca and this growth was relatively
evenly distributed among households. The economic effects of Aztec conquest generally
resulted in a cessation of further economic growth and an increase in interhousehold
variation. Cultural change at Calixtlahuaca shows a similar pattern. Prior to Aztec rule,
households at the site were gradually shifting toward more pan-Central Mexican cultural
practices, as opposed to practices distinct to the Toluca Valley, but maintained a high
degree of interhousehold similarity. Following the Aztec conquest of the site, the degree
of Aztec and/or more generally Central Mexican practices increases. However, these
cultural changes become increasingly unevenly distributed among households during this
phase. Both the economic and cultural trends are consistent with relatively collective
420
local rule prior to the conquest of the site by the Aztec Empire. The relatively small
changes under Aztec rule and their uneven distribution among households are most
consistent with both moderately indirect and relatively non-collective imperial rule.
Economic Hypotheses and Expectations
Based on previous case studies of Aztec provincial sites, I initially proposed two
scenarios for the economic effects of Aztec rule at Calixtlahuaca: 1) less direct, more
network oriented rule by the Aztec Empire, 2) more direct, more collective rule by the
Aztec Empire. I recognized that these two positions represented positions on a
continuum, that intermediate positions were quite possible, and that some current models
of Aztec economics include elements of both scenarios (i.e., Fargher 2009). The first
scenario proposed less direct, more network-oriented rule by the Aztec Empire,
conforming to traditional models of Aztec rule (Hassig 1985). Archaeologically, it is
generally characteristic of the evidence seen in the outer provinces of the empire. The
second scenario proposed both more direct and more collective rule by the Aztec Empire,
in keeping with Blanton and Farger’s (2008) argument for more collective governance. In
past archaeological research, this second scenario is generally best supported by sites in
the Basin of Mexico.
In the first scenario, Aztec rule at Calixtlahuaca was relatively indirect, with a
network-oriented rulership strategy focused on co-opting local elites. Under these
conditions, I expect the conquest by the Aztec Empire to have had little effect on existing
economic patterns, which were generally toward increasing market integration over the
421
course of the Postclassic. I expect to see the continuation or intensification of preconquest patterns of local craft production and market exchange, and a relatively low
frequency of artifacts from the Basin of Mexico which would have served simply as one
foreign trading partner among many. Overall wealth should remain steady or follow preAztec trajectories. This pattern of minimal change has been observed at Totogal (Venter
2012) in Veracruz and Coixtlahuaca (Kowalewski, et al. 2010) in Oaxaca. Calixtlahuaca
might be expected follow this pattern because, while part of a tributary rather than
strategic province, the Toluca Valley still formed a significant segment of the buffer zone
between the Aztec and Tarascan empires.
In the second scenario, Aztec rule was more direct, with a more collective
rulership strategy that worked to incorporate both commoners and elites directly into the
imperial system. In this case, I would expect to see more extensive changes in the local
economic system under Aztec rule, as is seen at sites in the Basin of Mexico. I expect the
Basin of Mexico (or even the imperial capitals more specifically) to dominate the local
economic system. This would be expected to result in a reduction in either the diversity
or total quantity of local craft production, and an increased volume of trade with the
Basin. Overall wealth should decrease. Previous case studies falling toward this end of
the economic continuum include Huexotla (Brumfiel 1980), and Xaltocan (Brumfiel
2005b) in the Basin of Mexico, and Yautepec, Cuexcomate and Capilco in Morelos
(Smith 2003c, 2004b, 2010). Calixtlahuaca might be expected show this pattern due to its
geographic proximity to the Basin of Mexico, the post-conquest rearrangement of local
power known from the codices (Chimalpahin 1965 [1606-1631]:105), and the importance
422
of the region in supplying maize to the Basin of Mexico (Paso y Tronasco 19051906:7[2]:6-7).
Economic Findings
Chapters 4 through 6 of this dissertation focused on the economic aspects of life
at Calixtlahuaca, with emphases on characterizing the pre-Aztec pattern and its
relationship to local strategies of rulership and on the changes that occurred after the
Aztec conquest and incorporation of the Toluca Valley into the Triple Alliance Empire.
The three chapters focused on evidence for exchange, craft production, and wealth,
respectively. They generally show similar temporal patterns to those seen in other parts of
Central Mexico during the Middle and Late Postclassic, though the overall levels of
economic integration into the Postclassic World System are lower than at
contemporaneous sites in the Basin of Mexico and Morelos. The general pattern is one of
increasing integration into the Postclassic Mesoamerican world system prior to Aztec
rule, with a lack of interhousehold variation consistent with relatively collective local
rule. This is followed by a levelling-off of economic growth and an increase in
interhousehold variation under Aztec rule during the Yata phase. This is consistent with
moderately direct but primarily network oriented rule by the Aztec Empire.
In chapter 4, I analyzed ceramic and lithic evidence for trade patterns over time at
Calixtlahuaca. Ceramic exchange was evaluated using INAA, petrography, and typebased classifications. Lithic exchange was evaluated using a combination of INAA and
XRF sourcing. Both artifact classes showed an increasing quantity and diversity of
423
imports between the Dongu and Ninupi phases, prior to the Aztec conquest, with access
to non-local goods relatively evenly distributed among households. During the Yata
phase, the diversity of both obsidian and ceramic sources decreases, and sources become
more unevenly distributed among households. However, the quantity of obsidian
reaching the site decreases, while the total frequency of imported ceramics increases. For
both artifact types, there is a shift in the source regions of foreign goods over time, with
an increasing proportion of imported goods coming from the Basin of Mexico.
In chapter 5, I examined evidence for craft production in the form of ceramic
production, direct lithic production using various technologies, and textile production.
Textile production increases over time, while most other forms of craft production
decrease, though overall levels of craft production are all low for all crafts during all
phases. Local ceramic production remains stable over time, measured both as the
percentage of local source groups in the INAA sample and as the degree of
standardization of common vessel forms. Evidence for blade-core and bifacial lithic
production decreased over time, while evidence for bipolar lithic production increased
during the Yata phase. When these trends are compared to the evidence for overall lithic
imports at the site over time, they suggest that bipolar lithic production is primarily a
reaction to a scarcity of obsidian. In contrast, multiple lines of evidence for textile
production (cotton and maguey spindle whorls, and basalt maguey scrapers) increase over
time. This shows an increase in both cotton and maguey textile production over time,
especially during the Yata phase. Interestingly, higher levels of cotton spinning occur
primarily in households displaying higher levels of other Aztec practices, while both
Aztec and locally oriented households show higher levels of maguey textile production.
424
The timing of the most dramatic increase in textile production suggests that the increase
in production levels was driven primarily by increased levels of taxation, rather than
increased market opportunities. The overall levels of evidence for all forms of craft
production at Calixtlahuaca are both low and unspecialized when compared to other
regional sites, which is unexpected considering that it was the dominant city-state in the
Toluca Valley prior to the area’s conquest by the Aztec Empire.
In chapter 6, I presented various measures of wealth at Calixtlahuaca and
discussed these within a broader context of their implications for the quality of life at the
site over time. Most of the measured wealth markers increased between the Dongu and
Ninupi phases. Following the Aztec conquest of the site, the overall pattern is toward a
levelling-off in households’ wealth with levels remaining similar to those seen during the
preceding Ninupi phase. I also calculated a combined wealth index, with various artifact
classes (ceramics, lithics, ground stone, and jewelry/rare items) assigned relative values
based on the prices listed in historic Nahuatl wills and other documents. Based on this
combined index, the average household’s wealth increased between the Dongu and
Ninupi phases, before remaining statistically equivalent during the Ninupi and Yata
phases. The range of variation in household wealth, measured as the coefficient of
variation or as absolute range, increased from each phase to the next, suggesting that
inequality among households was increasing over time. During the Yata phase, this
inequality is largely driven by the two most heavily culturally-Aztec households, which
are substantially wealthier than the other contemporaneous households at the site, based
on a wide range of artifact classes. This indicates that the Yata phase saw a decrease in
425
the standard of living under Aztec rule, both in terms of the household averages, but
likely even more so in terms of local households’ perceptions of inequality.
Taken together, the findings in these three chapters paint a picture of increasing
economic and social integration into the Postclassic Mesoamerican World System
(Kepecs and Kohl 2003) over time, partially interrupted and redirected by Calixtlahuaca’s
incorporation into the Triple Alliance Empire. During the Dongu phase, Calixtlahuaca
was culturally and economically isolated. The material culture of the Toluca Valley
diverges from that of the Basin of Mexico during the Early Postclassic and does not
obviously fit within that of any other surrounding region. (As an illustration of this
increasing divergence from regional trends, compare Epiclassic/Early Postclassic Santa
Cruz Atizapán (Kabata 2010) to Early/Middle Postclassic Huamango (Piña Chán 1981),
to Middle/Late Postclassic Calixtlahuaca.) Trade in both key necessities (obsidian) and
luxury goods (imported ceramics, bronze items) was relatively limited, but
geographically ecumenical, extending to the east, west, and south. Corresponding to this
lack of evidence for imported goods, there is also no evidence that anything was being
produced for export from Calixtlahuaca at this time. Interhousehold variation is at its
lowest during this phase, largely due to the lack of anything to vary.
During the subsequent Ninupi phase, the site reached its economic peak. Imported
ceramics from a diversity of sources are present in all households. Obsidian and copper,
relatively high-value/low-weight goods, occur in their highest frequencies during this
phase. At the same time, evidence for craft production remains relatively low, suggesting
that increased market opportunities did not provoke dramatic changes in the organization
or intensity of local craft production. This may be due to a lack of potential trade goods in
426
the Toluca Valley; the area lacks a high quality obsidian source, salt springs, or a climate
appropriate for cotton cultivation. The increased foreign goods that do appear at the site
are evenly distributed among households, both in terms of access to the full range of
source regions and in terms of relatively comparable frequencies of most goods in most
households. This broad-based economic growth is consistent with relatively collective
local rule providing economic opportunities to a wide portion of the commoner
population. The timing of the primary period of economic growth prior to Aztec rule also
demonstrates the independence of the market system developing in Postclassic Central
Mexico and the political development of the Triple Alliance.
During the Yata phase, and the associated Aztec rule of Calixtlahuaca, economic
growth at the site was interrupted, to the detriment of the residents of the site. Foreign
connections were re-oriented toward the Basin of Mexico, at the expense of trade with
other parts of Central Mexico. This finding contrasts with the argument that the Aztec tax
(tribute) requirements and/or a general “Pax Azteca” promoted lateral market
development (Blanton 1996; Hicks 1987). The associated reduction in obsidian at the site
curtailed an already very limited amount of local lithic production. In contrast, textile
production increases, likely to meet tribute requirements. The net effect of these
economic shifts is mildly negative, with no further growth in average household wealth
and a pronounced wealth difference between culturally local and culturally Aztec
households. Local households decrease to almost Dongu-phase wealth levels, while
households featuring higher levels of Aztec cultural practices have wealth levels at least
50% higher, driven by higher values for almost all artifact classes examined. The net
effect is consistent with a medium directness of rule based on the magnitude of economic
427
changes; prior patterns of exchange modified toward the Basin of Mexico (though no
former trading partner-regions were completely lost), resulting in a cessation of further
economic growth, despite the modest increase in textile production. The Yata phase
trends are also consistent with relatively non-collective rule by the Aztec Empire, as they
were almost uniformly negative for the residents of Calixtlahuaca.
Cultural Hypotheses and Expectations
Empires may seek to reduce the cost of controlling provincial areas via the
promotion of shared ideologies. This is well documented Aztec strategy for co-opting
provincial elites into supporting the empire (Berdan, et al. 1996; Stark and Chance 2012),
but the degree to which it effected commoners is less understood (Brumfiel 1998). At the
same time, provincial peoples may choose to adopt or reject foreign symbols or objects as
a means of bolstering their own social or economic positions within local contexts of
power. There is a distinction between using foreign objects, which may be manipulated in
a number of ways, and adopting foreign practices, which take a greater degree of
knowledge. The study of local choices concerning the adoption of foreign culture can
also be complicated by immigration from culturally distinctive areas. As a result, in
addition to considering evidence for the directness and collectiveness of rule at
Calixtlahuaca, I also consider the likelihood of emulation and immigration as alternative
explanations for the observed cultural changes.
Generally speaking, more direct rule should provide both greater top-down
pressure to adopt imperial practices and more opportunities for bottom-up interaction
428
between residents of the provinces and the imperial core. The collectiveness of rule will
also influence the adoption of non-local culture. The higher degree of regional interaction
produced by collective rule will result in more overall exposure to foreign cultural
practices. However, the patron-client relationships used in network-based rulership are
likely to produce more uneven adoption of foreign items and practices as individual
households acquire these from patrons, or use them to bolster specific relationships with
members of the imperial state.
In looking at more specific explanations for cultural changes under Aztec rule, I
used a framework of high- and low-visibility traits, based on Clark (2001) and related
work (Cowgill 2013; Ortman 2012) to differentiate between potential explanations for the
observed changes in material culture under Aztec rule. In this model, low-visibility traits,
such as pottery forming techniques or butchering practices, are learned from members of
one’s own community and are not likely to be subject to conscious manipulation, making
them good markers of immigrant populations. In contrast, high-visibility traits, such as
pottery decoration, are much more likely to be actively manipulated, whether as markers
of group identity (Clark, et al. 2013), local emulation of foreign locations, or immigrant
references to a distant homeland (Manzanilla 2004). In market based economies, such as
Postclassic Central Mexico, the use of objects in low/high visibility contexts is expected
to be a more useful marker than production techniques in identifying cultural origins.
I framed my investigation of cultural change at Calixtlahuaca around three
questions. First, how much cultural diversity was there at Calixtlahuaca prior to Aztec
rule and what can be inferred about local rule from this? More collective rule is generally
expected to produce more strongly shared cultural practices, due to both more inclusive
429
state-sponsored practices, and stronger social norms concerning sanctioning. Based on
the ethnohistoric documentation of multiple linguistic groups in the Postclassic Toluca
Valley, I considered it possible that there would be multiple cultural groups present
throughout Calixtlahuaca’s history. If this was the case, I expected there to be similar
dimensions of variation in particular forms of material culture that continued across the
two periods prior to Aztec rule, and the same household components would group
together based on multiple lines of evidence. I did not find evidence for multiple
distinctive cultural groups at Calixtlahuaca prior to Aztec rule; the homogeneity of
cultural practices at the site is consistent with a single cultural group under relatively
collective rule.
Second, what cultural changes occurred at Calixtlahuaca while it was under Aztec
rule and what can be inferred about Aztec rulership practices from these changes? What
is the overall magnitude of change toward the use of more Aztec style objects and
practices and how evenly are these distributed among households? As with the previously
discussed economic topics, I expect the directness of rule to be visible in the magnitude
of changes at the site, and the collectiveness of Aztec rule (and the degree to which it
interrupted prior local rulership) to be visible in the interhousehold variability in the
adoption of Aztec objects and/or practices. I found that there were some moderate
changes toward more Aztec cultural patterns and an increased use of Aztec style items
during the Yata phase. However, these changes were highly unevenly distributed among
households. This would be consistent with a moderately direct rule but primarily noncollective rulership strategy on the part of the Aztec Empire.
430
Third, can the changes seen under Aztec rule be explained by particular
mechanisms, namely local emulation or foreign immigration, for particular households?
It is possible that households at Calixtlahuaca could have actively emulated Aztec
culture, both in the form of goods and practices. This would likely be the result of a
combination of imperial and local choices, with both sides seeing advantages to the local
adoption of Aztec culture. This scenario would be visible archaeologically as an
appearance of Aztec practices in high visibility contexts, as well as a likely increase in
the use of Aztec style objects. In contrast, it is also possible that Aztec settlers from the
Basin of Mexico moved into Calixtlahuaca, bringing their traditional practices and
stylistic tastes with them. Ethnohistoric sources for the Toluca Valley do indicate some
degree of population replacement after the Aztec conquest, including both Matlatzincas
fleeing across the frontier to the Tarascan Empire, and immigrants from the Basin of
Mexico moving into the Toluca Valley (García Castro 1999). This would be visible as the
use of Aztec practices in low visibility contexts. Depending on the relationship between
the Aztec settlers and the local population, Aztec settlers might also emphasize their nonlocal origins in high visibility contexts, though this is not an essential component of this
scenario. A high degree of population replacement would also be expected to be
accompanied by a disruption of preexisting market networks and local craft production. I
found that households at Calixtlahuaca covered the full range of causal mechanisms,
from households that remained strongly culturally local, to those that adopted some Aztec
objects or practices in piecemeal ways, to those that adopted a wider range of Aztec style
goods and practices and could be strong local emulators, well integrated immigrants, or
ethnically mixed households.
431
Cultural Findings
Chapters 7 and 8 focused on the identification of cultural changes at Calixtlahuaca
over time. My analyses centered on questions of local rulership, the implications of
cultural change for Aztec rulership, and potential specific causes for the observed cultural
changes. Chapter 7 examined evidence for these social processes in ritual activities, and
Chapter 8 did the same using food preparation and serving methods. I selected these two
cultural domains because both have the potential to serve in both inclusive (inclusive
public ritual, inclusive feasting) and exclusive (exclusive feasting, exclusive rituals)
leadership strategies (Carballo 2015), and both include lower- and higher-visibility
aspects.
Chapter 7 addressed questions of cultural identity as expressed in ritual practice,
with a focus on two artifact types: figurines and censers. Figurines are a low-visibility
artifact type used almost exclusively in domestic contexts. Censers are medium visibility
artifact type, used in both domestic and state ritual activities (Huster, et al. 2015; Smith
2002). For both artifact types, the presence of non-local items (based on paste type) was
analyzed separately from the presence of non-local practices (based on figurine subject
matter and general censer form). Both artifact types were examined for patterns of
variation within Calixtlahuaca and for broad similarities with other contemporaneous
Central Mexican sites, using hierarchical and k-means clustering analyses. Prior to Aztec
rule, ritual practices at Calixtlahuaca were distinct from those seen in most other parts of
Central Mexico and relatively homogenous among households. Following the Aztec
conquest of the site, the analyses show far more Aztec influence in figurines, both in
432
terms of imported items and in terms of use, than in censers. They also showed
mismatches between households with physically Aztec items and some of those with
practices consistent with other parts of Central Mexico, which would be consistent with
local households integrating foreign objects into local frameworks of use on one hand
and immigrant households making due with locally available items on the other. The
appearance of Aztec objects and practices primarily in low-visibility domains not
associated with state policy is best explained through bottom up processes either of
migration or local adoption for reasons other than state sponsorship.
Chapter 8 looked at issues of cultural identity through food preparation and
serving methods. Food preparation is a low-visibility activity, while food serving
practices range from medium to high visibility, depending on the context of the meal. The
examination of food preparation focused on methods for preparing maize, including
grinding stones and cooking ceramics. Food serving methods were examined using
serving vessel form and decoration. Two of the datasets, ground stone and ceramic
decoration, could not be compared to regional datasets due to a lack of comparably coded
data (ground stone) or fundamental regional differences (ceramic decoration). The
remaining two datasets, cooking and serving vessel forms, were compared to other
contemporaneous sites in surrounding areas of Central Mexico, using the same
hierarchical and k-means clustering methods applied in the previous chapter. The results
of these analyses of foodways were less cleanly divided along low and high visibility
lines than those for ritual items. Some Yata phase households had somewhat non-local
patterns of both maize cooking vessels and serving vessel decoration, while the other
lines of evidence do not include any especially non-local cases.
433
On a phase-average level, the results of both of these chapters parallel the general
economic trends seen in the previous set of chapters. Evidence for cultural interaction –
both in terms of objects and in terms of practices - with the Basin of Mexico increases
over time. This is visible both in terms of the quantity of Aztec style material culture
entering circulation at Calixtlahuaca, and in the increasing use of the Aztec style, in both
local and imported objects. Cultural practices at the site are distinctive from those in most
of Central Mexico during these phases, indicating a distinct local cultural tradition.
During the final Yata phase, the site average for most traits remains more similar to
preexisting local practices than to non-local ones, but there is evidence for small shifts
toward a pattern of material culture more similar to the Basin of Mexico across a wide
range of lines of evidence.
Interhousehold variation also increases during Aztec rule. Together, the analyses
in these two chapters demonstrated that Dongu and Ninupi phase households were
relatively homogenous within phases. The ceramic phase usually, though not perfectly,
determined the primary cluster assignments for households dating to these two time
periods. In contrast, the Yata phase households did not consistently cluster as a single
group in most of the analyses, with individual households fluctuating between clustering
with earlier local cases, some contemporaneous local cases, and the foreign comparative
cases. The patterning in this variation is consistent with two households actively
maintaining a local identity (324-Ph6 and 327-Ph6), two local households emulating
some Aztec practices (316-Ph6 and 317-Ph6), and two households which were either
very strong cases of emulation, ethnically mixed, or well-integrated immigrants (307-Ph6
and 309-Ph6).
434
This patterning has implications for considerations of both immigrants and locals
at the site. First, if there were immigrants at Calixtlahuaca, they were a well-integrated
minority. The majority of the items in the immigrant households were consistent with
items made in the preexisting local tradition, indicating that the majority of suppliers of
basic goods in the local market system remained local peoples. However, despite this,
immigrant or ethnically mixed households chose to continue visibly marking their
cultural differences, using Aztec items in both some high and low visibility contexts. The
potential Aztec immigrant or strong emulator households were significantly wealthier
than any of the other contemporaneous local households at the site. This suggests it was
both safe and socially advantageous for immigrants to maintain and display social ties to
the Basin of Mexico.
Second, the local population could choose to opt in or out of emulating Aztec
practices. This can be seen in the contrast between the emulator and locally focused
households during the Yata phase. The presence of both imported and locally produced
versions of Aztec ceramics, especially figurines and Black-on-Orange ceramics, in local
households indicates that there were few cultural restrictions on the acquisition of such
goods. Neither were households missing Aztec goods restricted by economic causes. One
of the households which shows a rejection of Aztec culture is comparably wealthy to the
two households that did emulate aspects of Aztec culture, demonstrating that it was not
an inability to purchase such goods that limited the household’s access. At the same time,
the uneven distribution of different types of Aztec goods in individual households
indicates that there was no master promotion of such goods as an overall cultural
complex, to be adopted in full. The intermediate-level emulator households use a subset
435
of the same items and practices used by the Aztec immigrant/strong emulator households
at the site. Thus it is a reasonable possibility that these cases of intermediate-level local
emulation were primarily influenced by bottom-up non-state interactions with people
from the Basin of Mexico, such as their immigrant commoner neighbors, rather than by
official state policy or even local elites.
Cultural practices at Calixtlahuaca are most consistent with a relatively collective
form of pre-Aztec rulership. Most cultural traits show relatively low levels of
interhousehold variation prior to Aztec rule, and while access to non-local goods
increases, they are incorporated into local frameworks of use. Under Aztec rule, cultural
practices show a modest shift toward more Basin of Mexico practices, but also show
much higher levels of interhousehold variation. Given that many of the cultural practices
which shift toward more Aztec-characteristic patterns are not those that would be
expected to be associated with imperial cult activity (e.g. figurines), these shifts are more
likely the result of bottom-up processes than official Triple Alliance imperial policy. As a
result, the cultural shifts under Aztec rule are most consistent with moderately indirect
rule, and with a network-oriented rulership strategy.
Methodological Implications
In analyzing the data for this dissertation, I have identified three methodological
points of particular importance. These are the value of reasonable sample sizes, the
importance of comparing multiple lines of evidence, and the need to publish basic
436
excavation data for comparative purposes. While none of these are particularly new or
unique points, they do benefit from a brief discussion here.
First, in an era of reduced funding for archaeological research, there is a tendency
to excavate a relatively small sample of contexts at any given site, and extrapolate
extensively from them. This may be seen in any number of recent projects in
Mesoamerica, where two or three houses are taken to be representative of the site as a
whole. While archaeological interpretation will always be based on a sample, this trend
seems to be increasing. My analysis of households at Calixtlahuaca demonstrates the
flaws in such a method, as a random subsample of two or three of the excavated
household contexts from each phase (let alone two or three total contexts), would have
had a reasonably high probability of producing differing overall patterns of exchange,
craft production, or ethnic diversity over time simply due to the amount of variation in
the dataset as a whole. One example of this can be seen in the comparison of INAA (17
components) and petrography-based analyses (6 components) of imported ceramics over
time, where petrography showed the majority of the increase in imports earlier than the
INAA samples did. However, a comparison of the specific household components
selected for petrography showed that they did correspond with the INAA results from the
same contexts, but that these were not necessarily representative of the broader, sitewide
pattern. I am aware that, as a dissertation, the scope of my research would not have been
possible without access to the data produced by the greater Calixtlahuaca Archaeological
Project. However, this does point to the value of returning to a previously excavated site,
or coordinating multiple dissertation-level projects in order to provide a reasonable
sample of contexts.
437
Second, in a similar vein, my analyses also demonstrated the need to compare the
results of multiple lines of analysis. There is often a tendency for specialists associated
with a larger project to publish their results more or less independently of each other. In a
best-case scenario, this leads simply leads to similar conclusions weakened by not
considering supporting evidence, while in the worst-case scenario this leads to active
conflicts of interpretation based on what should be a single body of evidence. The data
from Calixtlahuaca could easily have produced the latter pattern, if the analyses had been
divided upon more traditional lines. For example, the evidence for the total quantity of
lithic and ceramic imports during the Yata phase show drastically different patterns, with
the former dropping and the latter rising.
Finally, my work on this dissertation has led me to emphasize the value of
publishing basic data, such as artifact counts, either in print or in a widely accessible
digital repository. This is not the sort of sexy publishing of hot theoretical ideas that will
get you talked about, but it is good, practical archaeology. If archaeology seeks to move
beyond the descriptive level, with explanations based on single sites, then it is imperative
that comparative data be widely available. As a fairly basic example, I compiled
comparative data on a number of items in this dissertation, including lithic/sherd ratios,
spindle whorl frequencies, and percentages of imported ceramics at various sites. In each
case, these comparative data allowed me to contextualize the results at Calixtlahuaca in a
regional context. For example, while the amount of evidence for textile production at
Calixtlahuaca goes up quite across the three study phases, even the final, highest levels
are far below what is seen at most contemporary sites. Given the amount published on
many of the sites for which I collected data, it was surprisingly hard to compile basic
438
data. The publication of such information allows for the analyses that the original
researcher may not have anticipated, and that are often not possible from summary
statistics alone.
Implications for Understanding the Aztec Empire
Based on the rulership strategies identified at Calixtlahuaca, this dissertation
provides three insights about the relationship between the Aztec Empire and the
commoner households which made up the majority of its population. First, as a general
characterization, local rule at Calixtlahuaca was relatively collective, while imperial
Aztec rule was both moderately indirect and primarily non-collective. This provides a
means of resolving the apparent contradiction between the characterizations of the Aztec
Empire provided by Hassig (1988) and Blanton and Fargher (2008). The former is
focused primarily on actions taken by the Triple Alliance proper, acting above the level
of the city-state. The latter conflates actions at the level of the city-state, which in the
Basin of Mexico highly overlapped the imperial hierarchy, with those of the Triple
Alliance. In shifting the focus of research to outside of the Basin, this distinction becomes
more apparent.
Second, as a result of this relatively indirect rule outside of the Basin of Mexico,
the Aztec Empire was not particularly accountable to its provincial subjects. Based on the
data from Calixtlahuaca, the Aztec Empire did not act for the benefit of the majority of its
population, especially the provincial commoner majority. (As a caveat, the commoner
population of the Basin of Mexico may have been a separate case.) Aztec rule curtailed
439
the growth of foreign trade connections and caused a cessation in economic growth at the
site. This is not a particularly new insight, as the Triple Alliance has traditionally been
characterized as predominately hegemonic in nature, more interested in what could be
extracted from its provinces than what could be invested in them (Hassig 1988). This net
negative effect of Aztec rule helps account for the frequent revolts against the Aztec
Empire, as commoners would have seen benefits in participating in attempts to free their
community from imperial control.
Third, while the Aztec empire did not act for the benefit of its provincial
population, neither did it actively oppress them. The combination of the economic and
cultural changes at Calixtlahuaca demonstrate that—at least in this case—the changes felt
by commoner households were more likely to be the secondary consequences of Aztec
imperial actions, rather than the direct result of policies targeting commoners themselves.
The loss of diversity in trading partners was likely a secondary consequence of the
establishment of the frontier between the Aztec and Tarascan Empires (Silverstein 2000).
In addition to official policies limiting trade across the border, ongoing conflict between
the two empires likely limited how much people living near the border felt comfortable
travelling. Both empires limited official cross-border trade and given the more centralized
nature of the Tarascan state (Pollard 2016), it may have applied the more successful
barriers to trade. Commoner access to non-local goods at Calixtlahuaca may also have
been affected by Aztec efforts to remove control of the local market system from local
political authorities (Berdan, et al. 1996; Hirth 2013). Similarly, there does not appear to
have been any attempt to introduce Aztec culture into commoner lifeways as a means of
political control. Some local households did adopt specific aspects of Aztec culture, but
440
these are idiosyncratic among households and often involve the incorporation of Aztec
style objects into local use practices. In addition, all of the emphasized items are part of
stylistic traditions that predate the Aztec Empire in the Basin of Mexico.
Future Research Directions
This work provides valuable insight into the effects of Aztec rule on provincial
commoners at Calixtlahuaca. The insights produced in this research demonstrate the
value of a nuanced consideration of the effects of imperial-provincial interaction. As a
result, case studies of Aztec imperialism should move beyond simply finding evidence
for an Aztec presence in provincial areas. This is an important first step, and the
distribution and types of evidence of interaction can provide valuable information about
the types of interaction occurring between the two regions. However, the next logical step
is to ask what the particular imperial and provincial strategies that produced this pattern
were, and how these combined to produce the observed effects of the interaction. There
may be cases, especially beyond the relatively highly integrated market network of
Central Highland Mexico, where Aztec rule had far fewer economic effects than at
Calixtlahuaca because there was less previous economic development to interrupt. In
looking at the directness and collectivity of rule, this dissertation provides a useful
framework for looking at the interplay between imperial and local organizations of
power. It should be extended to sites in other settings, particularly rural sites and sites
where there is little evidence for interaction, in order to further develop our understanding
of the Aztec Empire.
441
REFERENCES CITED
2009 Amoxcalli: La Casa de Los Libros. vol. 2014. CIESAS,
http://www.amoxcalli.org.mx/.
Abbott, Dave
2006 Hohokam Ritual and Economic Transformation: Ceramic Evidence from
the Phoenix Basin, Arizona. North American Archaeologist 27(4):285-310.
Abu-Lughod, Janet L.
1991 Before European hegemony: the world system AD 1250-1350. Oxford
University Press, Oxford.
Adams, Jenny L.
1999 Refocusing the Role of Food-Grinding Tools as Correlates for Subsistence
Strategies in the U.S. Southwest. American Antiquity 64(3):475-498.
Alcock, Susan E., Terence N. D'Altroy, Kathleen D. Morrison and Carla M. Sinopoli
(editors)
2001 Empires: Perspectives from Archaeology and History. Cambridge
University Press, New York.
Allen, Robert C.
2001 The Great Divergence in European Wages and Prices from the Middle
Ages to the First World War. Explorations in Economic History 38(4):411-447.
2009 How Prosperous were the Romans?: Evidence from Diocletian's Price
Edict (AD 301). In Quantifying the Roman Economy: Methods and Problems, edited by
A. Bowman and A. Wilson, pp. 327-345. Oxford University Press, Oxford.
Allen, Robert C., Jean-Pascal Bassino, Debin Ma, Christine Mull-Murata and Jan Luiten
van Zanden
2011 Wages, prices, and living standards in China, 1738-1925: in comparison
with Europe, Japan, and India. Economic History Review 64(S1):8-38.
Alva Ixtlilxochitl, Fernando de
1975-77 [1600-1640] Obras Históricas, edited by E. O'Gorman. 2 vols.
Universidad Nacional Autónoma de México, Mexico City.
Anderson, Arthur J. O., Frances Berdan and James Lockhart (editors)
1976 Beyond the codices: the Nahua view of colonial Mexico. University of
California Press, Berkeley.
Andrews, Bradford W.
442
2013 Calixtlahuaca Stone Tools: Technological Trends and Their
Socioeconomic Implications. Paper presented at the 78th Annual Meeting,
Society for American Archaeology, Honolulu, HI.
Anthony, David W.
1990 Migration in Archeology: The Baby and the Bathwater. American
Anthropologist 92(4):895-914.
Appadurai, Arjun
1988 How to Make a National Cuisine: Cookbooks in Contemprary India.
Comparative Studies in Society and History 30(1):3-24.
Arana, Raúl M.
1990 Proyecto Coatlán, área Tonatico-Pilcaya. Instituto Nacional de
Antropología e Historia.
Arias, Luz Marina and Desha M. Girod
2014 Indigenous Origins of Colonial Institutions. Quarterly Journal of Political
Science 9(3):371-406.
Arnold, Phillip J.
1991 Domestic Ceramic Production and Spatial Organization: A Mexican case
study in ethnoarchaeology. New Studies in Archaeology. Cambridge University
Press, Cambridge.
Arroyo Abad, Leticia, Elwyn Davies and Jan Luiten van Zanden
2012 Between conquest and independence: Real wages and demographic
change in Spanish America, 1530–1820. Explorations in Economic History
49(2):149-166.
Balkansky, Andrew K., Gary M. Feinman and Linda M. Nicholas
1997 Pottery Kilns of Ancient Ejutla, Oaxaca, Mexico. Journal of Field
Archaeology 24(2):139-160.
Barlow, Robert H.
1949 The Extent of the Empire of the Culhua Mexica. Ibero-Americana Series,
vol. 28. University of California Press, Berkeley.
1951 El codice de los alfareros de Cuauhtitlán. Revista Mexicana de Estudios
Antropológicos 12:5-12.
Basalenque, Fray Diego
1975 [1642] Arte y vocabulario de la lengua matlaltzingo vuelto a la
castellana. Biblioteca Enciclopédica del Estado de México 33. Estado de México,
Toluca.
443
Beekman, Christopher S. and Alexander F. Christensen
2003 Controlling for Doubt and Uncertainty through Multiple Lines of
Evidence: A New Look at the Mesoamerican Nahua Migrations. Journal of
Archaeological Method and Theory 10(2):111-164.
Berdan, Frances F.
2003 The Economy of Postclassic Mesoamerica. In The Postclassic
Mesoamerican World, edited by M. E. Smith and F. F. Berdan, pp. 93-95.
University of Utah Press, Salt Lake City.
2008 Concepts of Ethnicity and Class in Aztec-Period Mexico. In Ethnic
Identity in Nahua Mesoamerica: the View from Archaeology, Art History, Ethnohistory,
and Contemporary Ethnography, edited by F. F. Berdan, J. K. Chance, A. R. Sandstrom,
B. J. Stark, J. Taggart and E. Umberger, pp. 105-132. University of Utah Press, Salt Lake
City.
Berdan, Frances F. and Patricia Rieff Anawalt (editors)
1992 [1541] The Codex Mendoza. 4 vols. University of California Press,
Berkeley.
Berdan, Frances F., Richard E. Blanton, Elizabeth H. Boone, Mary G. Hodge, Michael E.
Smith and Emily Umberger (editors)
1996 Aztec Imperial Strategies. Dumbarton Oaks, Washington.
Berdan, Frances F. and Michael E. Smith
1996 Imperial Strategies and Core-Periphery Relations. In Aztec Imperial
Strategies, edited by F. F. Berdan, R. E. Blanton, E. H. Boone, M. G. Hodge, M.
E. Smith and E. Umberger, pp. 209-217. Dumbarton Oaks, Washington.
2003 The Aztec Empire. In The Postclassic Mesoamerican World, edited by M.
E. Smith and F. F. Berdan, pp. 67-72. University of Utah Press, Salt Lake City.
Biskowski, Martin
2000 Maize Preparation and the Aztec Subsistence Economy. Ancient
Mesoamerica 11:293-306.
2008 Maize-grinding tools in prehispanic Mexico. In New Approaches to Old
Stone: Recent Studies of Ground Stone Artifacts., edited by M. Y. Rowan and J. R.
Ebling, pp. 144-158. Equinox, London.
Biskowski, Martin and Karen D. Watson
2013 Changing Approaches to Maize Preparation at Cerro Portezuelo. Ancient
Mesoamerica 24(1):213-223.
Blake, Michael
444
2010 Dating the Intial Spread of Zea mays. In Histories of Maize in
Mesoamerica, edited by J. E. Staller, R. H. Tykot and B. F. Benz, pp. 45-62. Left
Coast Press, Walnut Creek, CA.
Blanton, Richard E.
1983 The Aztec Garrison of "Acatepec". In The Cloud People: Divergent
Evolution of the Zapotec and Mixtec Civilizations, edited by K. V. Flannery and J.
Marcus, pp. 318. Academic Press, New York.
1996 The Basin of Mexico Market System and the Growth of Empire. In Aztec
Imperial Strategies, edited by F. F. Berdan, R. E. Blanton, E. H. Boone, M. G. Hodge, M.
E. Smith and E. Umberger, pp. 47-84. Dumbarton Oaks, Washington.
2010 Collective Action and Adaptive Socioecological Cycles in Premodern
States. Cross-Cultural Research 44(1):41-59.
Blanton, Richard E. and Lane F. Fargher
2008 Collective Action in the Formation of Pre-Modern States. Springer, New
York.
Blanton, Richard E., Lane F. Fargher and Vernice F. Heredia Espinoza
2005 The Mesoamerican World of Goods and Its Transformations. In
Settlement, subsistence, and social complexity : essays honoring the legacy of
Jeffrey R. Parsons, edited by R. E. Blanton, pp. 314. Cotsen Institute of
Archaeology, University of California, Los Angeles.
Blanton, Richard E., Gary M. Feinman, Stephen A. Kowalewski and Peter E. Peregrine
1996 A Dual-Processual Theory for the Evolution of Mesoamerican
Civilization. Current Anthropology 37(1):1-14.
Blanton, Richard E., Stephen A. Kowalewski, Gary M. Feinman and Jill Appel
1993 Ancient Mesoamerica: A Comparison of Change in Three Regions.
Cambridge University, Cambridge.
Blanton, Richard E. and Peter N. Peregrine
1997 Main Assumptions and Variables for Economic Analysis Beyond the
Local System. In Economic Analysis Beyond the Local System, edited by R. E.
Blanton, P. N. Peregrine, D. Winslow and T. D. Hall, pp. 3-12. University Press
of America, Lanham.
Bodley, John H.
2003 The Power of Scale: A Global History Approach. M. E. Sharpe, Armonk,
NY.
Boone, Elizabeth Hill
445
1987 Templo Mayor Research 1521-1978. In The Aztec Templo Mayor, edited
by E. H. Boone, pp. .5-70. Dumbarton Oaks, Washington, D.C.
1996 Manuscript Painting in the Service of Imperial Ideology. In Aztec Imperial
Strategies, edited by F. F. Berdan, R. E. Blanton, E. H. Boone, M. G. Hodge, M. E. Smith
and E. Umberger, pp. 181-209. Dumbarton Oaks, Washington.
2003 A Web of Understanding. In The Postclassic Mesoamerican World, edited
by M. E. Smith and F. F. Berdan, pp. 207-221. University of Utah Press, Salt Lake City.
Boone, Elizabeth Hill and Michael E. Smith
2003 Postclassic International Styles and Symbol Sets. In The Postclassic
Mesoamerican World, edited by M. E. Smith and F. F. Berdan, pp. 186-193.
University of Utah Press, Salt Lake City.
Borejsza, Aleksander, Isabel Rodríguez López and Charles D. Frederick
n.d.
Informe de Los Análisis de Laboratorio del Proyecto Calixtlahuaca 20072015. The Calixtlahuaca Archaeological Project, v. 3. BAR.
Bourdieu, Pierre
1977 Outline of a Theory of Practice. Cambridge University Press, Cambridge.
1984 Distinction: A Social Critique of the Judgement of Taste. Harvard
University Press, Cambridge.
Bove, Frederick J., José Vicente Genovex and Carlos A. Batres
2012 Pipil Archaeology of Coastal Guatemala. In Fanning the Sacred Flame:
Mesoamerican Studies in Honor of H. B. Nicholson, edited by M. A. Boxt and B.
D. Dillon, pp. 231-268. University Press of Colorado, Boulder, CO.
Bowles, Samuel and Herbert Gintis
2004 Persistent parochialism: trust and exclusion in ethnic networks. Journal of
Economic Behavior & Organization 55(1):1-23.
2006 Social preferences, homo economicus, and zoon politikon. In The Oxford
Handbook of Contextual Political Analysis edited by R. E. Goodin and C. Tilly, pp. 172186. Oxford University Press, Oxford.
Bowser, Brenda J.
2000 From Pottery to Politics: An Ethnoarchaeological Study of Political
Factionalism, Ethnicity, and Domestic Pottery Style in the Ecuadorian Amazon.
Journal of Archaeological Method and Theory 7(3):219-248.
Bradby, Hannah
446
1997 Health, Eating and Heart Attacks: Glaswegian Punjabi women's thinking
about everyday food. In Food, Health and Identity, edited by P. Caplan, pp. 213233. Routledge, London and New York.
Bragdon, Kathleen J.
1988 The material culture of the Christian Indians of New England, 1650-1775.
In Documentary Archaeology in the New World, edited by M. C. Beaudry, pp.
126-131. New Directions in Archaeology. Cambridge University Press,
Cambridge.
Braswell, Geoffrey E.
2003 Obsidian Exchange Spheres. In The Postclassic Mesoamerican World,
edited by M. E. Smith and F. F. Berdan, pp. 131-158. University of Utah Press,
Salt Lake City.
2010 The Rise and Fall of Market Exchange: A Dynamic Approach to Ancient
Maya Economy. In Archaeological Approaches to Market Exchange in Ancient Societies,
edited by C. P. Garraty and B. L. Stark, pp. 127-139. University Press of Colorado,
Boulder.
Brown, Clifford T., April A. Watson, Ashley Gravlin-Berman and Larry S. Liebovitch
2012 Poor Mayapan. In The Ancient Maya of Mexico: Reinterpreting the Past of
the Northern Maya Lowlands, edited by G. E. Braswell, pp. 306-324. Equinox,
Briston, CT.
Brown, Marley R. III
1988 The behavioral context of probate inventories: an example from the
Plymouth Colony. In Documentary Archaeology in the New World, edited by M.
C. Beaudry, pp. 79-82. New Directions in Archaeology. Cambridge University
Press, Cambridge.
Brown, Michael F.
1996 On Resisting Resistance. American Anthropologist 98:729-735.
Brumfiel, Elizabeth M.
1976 Specialization and Exchange at the Late Postclassic (Aztec) Community
of Huexotla, Mexico, Department of Anthropology, University of Michigan, Ann
Arbor.
1980 Specialization, Market Exchange and the Aztec State: the View from
Huexotla. Current Anthropology 21(4):459-478.
1991 Weaving and Cooking: Women's Production in Aztec Mexico. In
Engendering Archaeology: Women and Prehistory, edited by J. M. Gero and M. W.
Conkey, pp. 224-251. Basil Blackwell, Oxford.
447
1996 Figurines and the Aztec State: Testing the Effectiveness of Ideological
Domination. In Gender and Archaeology, edited by R. P. Wright. University of
Pennsylvania Press, Philadelphia.
1997 Tribute Cloth Production and Compliance in Aztec and Colonial Mexico.
Museum Anthropology 21(2):55-71.
1998 Huitzilopochtli's Conquest: Aztec Ideology in the Archaeological Record.
Cambridge Archaeological Journal 8(01):3-13.
2005a Conclusions: Production and Power at Xaltocan. In La producción local y
el poder en el Xaltocan Posclásico / Production and Power at Postclassic Xaltocan,
edited by E. M. Brumfiel, pp. 348-368. Arqueología de México. Instituto Nacional de
Antropología e Historia, Mexico City and University of Pittsburgh, Pittsburgh.
2005b La producción local y el poder en el Xaltocan Posclásico / Production and
Power at Postclassic Xaltocan. Arqueología de México. Instituto Nacional de
Antropología e Historia, Mexico City and University of Pittsburgh, Pittsburgh.
2007 Solar Disks and Solar Cycles: Spindle Whorls and the Dawn of Solar Art
in Postclassic Mexico. Treballs d' Arqueologia 13:91-113.
Brumfiel, Elizabeth M. and Lisa M. Overholtzer
2009 Alien Bodies, Everyday People, and Hollow Spaces: Embodiment,
Figurines, and Social Discourse in Postclassic Mexico. In Mesoamerican
Figurines: Small-Scale Indices of Large-Scale Social Phemonena, edited by C. T.
Halperin, K. A. Faust, R. Taube and A. Giguet, pp. 297-326. University Press of
Florida, Gainsville, FL.
Brumfiel, Elizabeth M. and Cynthia Robin
2012 Class and Ethnicity in Ancient Mesoamerica. In The Oxford Handbook of
Mesoamerican Archaeology, edited by D. L. Nichols and C. Pool, pp. 673-683.
Oxford University Press, New York.
Brumfiel, Elizabeth M., Tamara Salcedo and David K. Schafer
1993 The lip plugs of Xaltocan: Function and meaning in Aztec archaeology. In
Economies and Polities in the Aztec Realm, edited by M. G. Hodge and M. E.
Smith, pp. 113-130. Institute for Mesoamerican Studies, Albany.
Bullock Kreger, Meggan M.
2010 Urban population dynamics in a preindustrial New World city: Morbidity,
mortality, and immigration in postclassic Cholula, Department of Anthropology,
Pennsylvania State University, State College, PA.
Burke, Adrian L. and Gilles Gauthier
448
2013 Geochemical Sourcing of Obsidian from the Toluca Valley using XRF.
Paper presented at the 78th Annual Meeting, Society for American Archaeology,
Honolulu, HI.
Burmeister, Stefan
2000 Archaeology and Migration: Approaches to an Archaeological Proof of
Migration. Current Anthropology 41(4):539-567.
Bustamante, Maurico and Domingo Garcia
2015 Distinction's framework*: A tool to understand the space of cultural
practices in Mexico. In The Routledge Companion to Bourdieu’s 'Distinction',
edited by P. Coulangeon and J. Duval. CRESC. Routledge.
Camposeco M., Jose Balvino
1994 An Ch'ech, Sajchi' o Ki: El Maguey y Sus Usos en Guatemala. Yax Te'
Press, Rancho Palos Verdes.
Carbajal Correa, María del Carmen and Luis Alfonso González Miranda
2003 Cerro de los Magueyes: un centro funerario para matlatzincas y mexicas
durante el Postclásico tardío. Arqueología (INAH) 29:65-114.
Carballo, David M.
2009 Household and Status in Formative Central Mexico: Domestic Structures,
Assemblages, and Practices at La Laguna, Tlaxcala. Latin American Antiquity
20:473-501.
2015 Urbanization and Religion in Ancient Central Mexico. Oxford Studies in
the Archaeology of Ancient States. Oxford University Press, Oxford.
Carballo, David M., Paul Roscoe and Gary M. Feinman
2014 Cooperation and Collective Action in the Cultural Evolution of Complex
Societies. Journal of Archaeological Method and Theory 21:98-133.
Card, Jeb J. (editor)
2013 The Archaeology of Hybrid Material Culture. Southern Illinois University
Carbondale, Carbondale, IL.
Carr, Christopher and Jill E. Neitzel (editors)
1995 Style, Society, and Person: Archaeological and Ethnological Perspectives.
Plenum Press, New York.
Carrasco, Pedro
1950 Los Otomíes: cultura e historia prehispánica de los pueblos
mesoamericanos de habla otomiana. Universidad Nacional Autónoma de México,
Mexico City.
449
1964 Family Structure of Sixteenth-Century Tepoztlan. In Process and Pattern
in Culture: Essays in Honor of Julian H. Steward, edited by R. A. Manners. Aldine
Publishing Company, Chicago.
1999 The Tenochca Empire of Ancient Mexico: The Triple Alliance of
Tenochtitlan, Tetzcoco, and Tlacopan. University of Oklahoma Press, Norman.
Cashdan, Elizabeth (editor)
1990 Risk and Uncertainty in Tribal and Peasant Economies. Westview Press,
Boulder, CO.
Castanzo, Ronald A.
2004 Tepeaca Kiln Project. Report submitted to the Foundation for the
Advancement of Mesoamerican Studies (http://www.famsi.org/reports/02021).
Castillo Romero, Guizella
1996 Enterramientos rituales en el Valle de Toluca: los sitios arqueológicos de
Calixtlahuaca, Teotenango, y Metepec. Tesis de Maestría, Tesis de Maestría,
Historia y Etnohistoria, Escuela Nacional de Antropología e Historia, Mexico
City.
Chapman, Anne C.
1957 Port of Trade Enclaves in Aztec and Maya Civilizations. In Trade and
Market in Early Empires, edited by K. Polanyi, C. M. Arensberg and H. W.
Pearson, pp. 114-153. Free Press, Chicago, IL.
Charlton, Cynthia Otis, Thomas H. Charlton and Deborah L. Nichols
1993 Aztec Household-Based Craft Production: Archaeological Evidence from
the City-State of Otumba. In Prehispanic Domestic Units in Western
Mesoamerica: Studies of the Household, Compound, and Residence, edited by R.
S. Santley and K. G. Hirth, pp. 147-172. CRC Press, Boca Raton.
Charlton, Thomas H.
1996 Early Colonial Period Ceramics: Decorated Red Ware and Orange Ware
Types of the Rural Otumba Aztec Ceramic Complex. In Arqueología
Mesoamericana: Homenaje a William T. Sanders, edited by A. G. Mastiche, J. R.
Parsons, R. S. Santley and M. C. Serra Puche, pp. 461-479. vol. 1. INAH,
Mexico.
Charlton, Thomas H. and Patricia Fournier Garcia
2011 Pots and Plots: The Multiple Roles of Early Colonial Red Wares in the
Basin of Mexico (Identity, Resistence, Negotiation, Accomodation, Aesthetic
Creativity, or Just Plain Economics). In Enduring Conquests: Rethinking the
Archaeology of Resistance to Spanish Colonialism in the Americas, edited by M.
Liebmann and M. S. Murphy, pp. 127-148. SAR Press, Santa Fe, NM.
450
Charlton, Thomas H., Patricia Fournier Garcia and Juan Cervantes
1995 La cerámica del periodo Colonial Temprano en Tlateloco: el caso de la
Loza Roja Bruñida. In Presencias y encuentros: Investigaciones arqueológicas de
Salvamento, pp. 135-156. Dirección de Salvamento Arqueológico, Mexico City.
Charlton, Thomas H. and Deborah L. Nichols
1992 Late Postclassic and Colonial Period Elites at Otumba, Mexico: The
Archaeological Dimensions. In Mesoamerican Elites: An Archaeological
Assessment, edited by D. Z. Chase and A. F. Chase, pp. 242-258. University of
Oklahoma Press, Norman, OK.
Charlton, Thomas H., Deborah L. Nichols and Cynthia Otis Charlton
1991 Aztec craft production and specialization: archaeological evidence from
the city-state of Otumba, Mexico. World Archaeology 23:p. 98-114.
Charlton, Thomas H., Deborah L. Nichols and Cynthia Otis Charlton
2000 Otumba and Its Neighbors: Ex oriente lux. Ancient Mesoamerica 11:247265.
Chase-Dunn, Christopher and Thomas D. Hall
1997 Rise and Demise: Comparing World-Systems. Westview Press, Boulder.
Chase, Brad, P. Ajithprasad, S.V. Rajesh, Ambika Patel and Bhanu Sharma
2014 Matieralizing Harappan identities: Unity and diversity in the borderlands
of the Indus Civilization. Journal of Anthropological Archaeology 35:63-78.
Cheal, David
1989 Strategies of Resource Management in Household Economies: Moral
Economy or Political Economy? In The Household Economy: Reconsidering the
Domestic Mode of Production, edited by R. R. Wilk, pp. 11-22. Westview Press.,
Boulder, CO.
Chimalpahin, Francisco de San Antón Muñon
1965 [1606-1631]
Relaciones Originales de Chalco Amequemecan.
Translated by S. Rendón. Fondo de Cultura Económica, Mexico City.
Clark, Jeffery J.
2001 Tracking Prehistoric Migrations: Pueblo Settlers among the Tonto Basin
Hohokam. Anthropological Papers of the University of Arizona. vol. 65.
University of Arizona Press, Tucson.
Clark, Jeffery J., Deborah L. Huntley, J. Brent Hill and Patrick D. Lyons
2013 The Kayenta Diaspora and Salado Meta-Identity in the Late Pre-Contact
U.S. Southwest. In The Archaeology of Hybrid Matieral Culture (Occasional
Paper No. 39), edited by J. J. Card, pp. 399-424. Southern Illinois University
Press, Carbondale.
451
Clark, Jeffery J., Patrick Lyons, J. Brett Hill, Stacey Lengyel and Mark Slaughter
2009 Migrants and Mounds in the Lower San Pedro Valley, A.D. 1200-1450. In
Between Mimbres and Hohokam: Exploring the Archaeology and History of
Southeastern Arizona, edited by H. Wallace. University of Arizona Press, Tucson.
Clark, John E.
1987 Politics, Prismatic Blades and Mesoamerican Civilization. In The
Organization of Core Technology, edited by J. K. Johnson and C. A. Morrow, pp.
259-284. Westview Press, Boulder, CO.
Cline, S. L. and Miguel León-Portilla
1993 The Testaments of Culhuacan (modified first edition). UCLA Latin
American Center Nahuatl Studies Series, No. 1.
Cobean, Robert H.
2002 Un Mundo Obsidiana: Minería y comercio de un vidrio volcánico en el
México antiguo/ A World of Obsidian: the mining and trade of a volcanic glass in
ancient Mexico. Arqueología de México, Instituto Nacional de Antropología e
Historia, Mexico and University of Pittsburgh, Pittsburgh.
Cortés, Hernando
1962 Five Letters of Cortés to the Emperor. Translated by J. B. Morris. Norton,
New York.
Costin, Cathy
1991 Craft Specialization: Issues in Defining, Documenting, and Explaining the
Organization of Production. In Archaeology Method and Theory, edited by M. B.
Schiffer, pp. 1-56. University of Arizona Press, Tucson.
Cowgill, George L.
2008 An Update on Teotihuacan. Antiquity 82:962-975.
2013 Possible Migrations and Shifting Identities in the Central Mexican
Epiclassic. Ancient Mesoamerica 24(1):131-149.
Cresswell, R.
1976 Avant-Propos. Techniques et Culture 1(1):5-6.
Crider, Destiny L.
2011 Epiclassic and Early Postclassic Interaction in Central Mexico as
Evidenced by Decorated Pottery, School of Human Evolution and Social Change,
Arizona State University, Tempe.
Crown, Patricia L. and Suzanne K. Fish
452
1996 Gender and Status in the Hohokam Pre-Classic to Classic Transition.
American Anthropologist 98(4):803-817.
Cuauhtitlán, Anales de
1985 [1606-1631]
Mexico.
Códice Chimalpopoca o Anales de Cuauhtitlán. UNAM,
Cuauhtitlán, Códice de los alfareros de
1568 Pieza de contaduría de una fábrica de cerámica indígena, posterior a la
Conquista. vol. Manuscritos mexicanos 103-118. Mesoamerican Codex 109 vols.
Biblioteca Nacional de París.
Cucina, A. and V. Tiesler
2003 Dental caries and antemortem tooth loss in the Northern Peten area,
Mexico: A biocultural perspective on social status differences among the Classic
Maya. American Journal of Physical Anthropology 122:1-10.
Curet, L. Antonio
1993 Regional Studies and Ceramic Production Areas: An Example from La
Mixtequilla, Veracruz, Mexico. Journal of Field Archaeology 20:427-440.
Cutright, Robyn E.
2015 Eating Empire in the Jequetepeque: A Local View of Chimú Expansion on
the North Coast of Peru. Latin American Antiquity 26(1):64-86.
D'Altroy, Terence N. and Timothy K. Earle
1985 Staple Finance, Wealth Finance, and Storage in the Inka Political
Economy. Current Anthropology 26(2):187-206.
D'Altroy, Terence N. and Christine A. Hastorf (editors)
2001 Empire and Domestic Economy. Plenum, New York.
De Lucia, Kristin
2011 Domestic Economies and Regional Transition: Household Production and
Consumption in Early Postclassic Mexico, Department of Anthropology,
Northwestern University, Evanston, IL.
2014 Everyday Practice and Ritual Space: the Organization of Domestic Ritual
in Pre-Aztec Xaltocan, Mexico. Cambridge Archaeological Journal 24(3):379-403.
De Lucia, Kristin and Lisa M. Overholtzer
2014 Everyday Action and the Rise and Decline of Ancient Polities: Household
Strategy and Political Change in Postclassic Xaltocan, Mexico. Ancient
Mesoamerica 25(02):441-458.
Deagan, Kathleen A.
453
1973
Mestizaje in colonial St. Augustine. Ethnohistory 20(1):55-65.
Deagan, Kathleen A. and Joan K. Koch
1983 Spanish St. Augustine: The archaeology of a colonial Creole community.
Academic Press, New York.
Deal, Michael
1998 Pottery Ethnoarchaeology in the Central Maya Highlands. Foundations of
Archaeological Inquiry. University of Utah Press, Salt Lake City.
Deetz, James
1977 In Small Things Forgotten: the Archaeology of the Early American Life.
Anchor Press, Garden City, NY.
Díaz del Castillo, Bernal
1963 The Conquest of New Spain. Translated by J. M. Cohen. Penguin, New
York.
Doyle, Michael W.
1986 Empires. Cornell University Press, Ithaca.
Drennan, Robert D.
1984 Long-Distance Transport Costs in Pre-Hispanic Mesoamerica. American
Anthropologist 86:105-112.
1985 Porters, Pots, and Profit: The Economics of Long-Distance Exchange in
Mesoamerica. American Anthropologist 87(4):891-893.
Durán, Diego
1951 [1581] Historia de las Indias de Nueva España e Islas de la Tierra Firme,
edited by A. M. Garibay K. 2 vols. Editorial Porrua, Mexico.
Earle, Timothy K.
1987 Specialization and the production of wealth: Hawaiian chiefdoms and the
Inka empire. In Specialization, exchange and complex societies, edited by E. M.
Brumfiel and T. K. Earle, pp. 64-75. Cambridge University Press, Cambridge.
Earle, Timothy and Michael E. Smith
2012 Household Economies Under the Aztec And Inka Empires: A
Comparison. In The Comparative Archaeology of Complex Societies, edited by
M. E. Smith, pp. 238-284. Cambridge University Press, Cambridge.
Eerkens, Jelmer W. and Robert L. Bettinger
2001 Techniques for Assessing Standardization in Artifact Assemblages: Can
We Scale Material Variability. American Antiquity 66(3):493-504.
454
Eisenstadt, Shmuel Noah
1993 [1963] The Political Systems of Empires New Brunswick. Transaction
Publishers.
Elam, J. Michael, Michael D. Glascock and Hector Neff
2008 Obsidian Source Analysis. In Un Lugar de Jade: Sociedad y Economía en
el Antiguo Chalco/ Place of Jade: Society and Economy in Ancient Chalco, edited
by M. G. Hodge, pp. 259-268. Instituto Nacional de Antropología e Historia,
Mexico City and University of Pittsburgh, Pittsburgh.
Ellis, Simon P.
2000 Roman Housing. Duckworth, London.
Elson, Christina M.
1995 Aztec Chiconautla: An Analysis of Dr. George Vallant's 1935 Excavation,
MA thesis, Department of Anthropology, Hunter College, The City Univerity of
New York.
1999 An Aztec Palace at Chiconautla, Mexico. Latin American Antiquity
10(2):151-167.
Elson, Christina M. and R. Alan Covey (editors)
2006 Intermediate Elites in Pre-Columbian States and Empires. University of
Arizona Press, Tucson.
Elson, Christina M. and Michael E. Smith
2002 Archaeological Deposits from the Aztec New Fire Ceremony. Ancient
Mesoamerica 12(02):157-174.
Emerson, Thomas E. and Kristin M. Hedman
2016 The Dangers of Diversity: The Consolidation and Dissolution of Cahokia,
Native North America's First Urban Polity. In Beyond Collapse: Archaeological
Perspectives of Resilience, Revitalization, and Transformation in Complex
Societies, edited by R. K. Faulseit, pp. 147-176. Center for Archaeological
Investigation, Occasional Paper No. 42. Southern Illinois University, Carbondale,
IL.
Eriksson, Lina
2011 Rational Choice Theory: Potential and Limits. Palgrave Macmillian, New
York.
Evans, Susan Toby
1988 Excavations at Cihuatecpan: an Aztec Village in the Teotihuacan Valley.
Vanderbilt University Publications in Archaeology, Nashville, TN.
455
1992 The Productivity of Maguey Terrace Agriculture in Central Mexico
During the Aztec Period. In Gardens of Prehistory: The Archaeology of Settlement
Agriculture in Greater Mesoamerica, edited by T. J. Killion, pp. 92-116. University of
Alabama Press, Tuscaloosa.
2006 Antecedents of the Aztec Palace: Palaces and Political Power in Classic
and Postclassic Mexico. In Palaces and power in the Americas : from Peru to the
northwest coast, edited by J. J. Christie and P. J. Sarro. University of Texas Press, Austin.
Fargher, Lane F.
2009 A Comparison of the Spatial Distribution of Agriculture and Craft
Specialization in Five State-Level Societies Journal of Anthropological Reseach
65(3):353-387.
Fargher, Lane F. and Richard E. Blanton
2007 Revenue, Voice, and Public Goods in three Pre-Modern States.
Comparative Studies in Society and History 49:848-882.
Fargher, Lane F., Verenice Y. Heredia Espinoza and Richard E. Blanton
2011 Alternative pathways to power in late Postclassic Highland Mesoamerica.
Journal of Anthropological Archaeology 30:306-326.
Faust, Katherine A. and Christina T. Halperin
2009 Approaching Mesoamerican Figurines. In Mesoamerican Figurines:
Small-Scale Indices of Large-Scale Social Phemonena, edited by C. T. Halperin,
K. A. Faust, R. Taube and A. Giguet, pp. 1-24. University Press of Florida,
Gainsville, FL.
Feinman, Gary M.
1995 The emergence of inequality. In Foundations of social inequality, pp. 255279. Springer.
1996 The Changing Structure of Macroregional Mesoamerica: With Focus on
the Classic-Postclassic Transition in the Valley of Oaxaca. Journal of World-Systems
Research 2(1).
1997 Macro-Scale Perspectives on Settlement and Production in Ancient
Oaxaca. In Economic Analysis Beyond the Local System, edited by R. E. Blanton, P. N.
Peregrine, D. Winslow and T. D. Hall, pp. 13-42. University Press of America, Lanham.
1998 Scale and Social Organization: Perspectives on the Archaic State. In
Archaic States, edited by G. M. Feinman and J. Marcus, pp. 95-134. School of American
Research Press, Santa Fe, NM.
Feinman, Gary M. and Christopher P. Garraty
456
2010 Preindustrial Markets and Marketing: Archaeological Perspectives.
Annual Review of Anthropology 39:167-191.
Feinman, Gary M. and Linda M. Nicholas
1991 The Monte Albán State: A Diachronic Perspective on an Ancient Core and
its Periphery. In Core / Periphery Relations in Precapitalist Worlds, edited by C.
Chase-Dunn and T. D. Hall, pp. 240-276. Westview Press, Boulder.
Feinman, Gary M., Steadman Upham and Kent G. Lightfoot
1981 The Production Step Measure: An Ordinal Index of Labor Input in
Ceramic Manufacture. American Antiquity 46:871-884.
Ferwerda, Jeremy and Nicholas L. Miller
2014 Political Devolution and Resistance to Foreign Rule: A Natural
Experiment. American Political Science Review 108(3):642-660.
Flannery, Kent V. and Joyce Marcus (editors)
2003 The Cloud People: Divergent Evolution of the Zapotec and Mixtec
Civilizations. Percheron Press, Clinton Corners, NY.
Flenniken, J. Jeffrey
1981 Replicative Systems Analysis: a Model Applied To the Vein Quartz
Artifacts from the Hoko River Site. Washington State University Laboratory of
Anthropology Reports of Investigation. Washington State University.
Fournier Garcia, Patricia
1998 El complejo nixtamal/comal/tortilla en Mesoamerica. Boletin de
Antropologia Americana 32:13-40.
2007 Los Hñähñü del Valle del Mezquital: Maguey, Pulque y Alfarería. INAH,
Escuela Nacional de Antropologia e Historia, Mexico City.
Fowles, Severin M.
2013 An archaeology of doings: secularism and the study of Pueblo religion.
School for Advanced Research Press, Santa Fe, NM.
Frank, A. G.
1966 The development of underdevelopment. Monthly Review (September):1731.
García Castro, René
1999 Indios, territorio y poder en la provincia matlatzinca: la negociación del
espacio político de los pueblos otomianos, siglos XV-XII. CIESAS, Instituto
Nacional de Antropología e Historia, Mexico City and El Colegio Mexiquense,
Toluca, Mexico.
457
2000 Tradición, matrimonio y patrimonio. Una familia de caciques mazahuas
en el siglo XVII, vista a traves de sus testamentos. In Jocotitlán, edited by R. Hernández
Rodríguez. El Colegio Mexiquense, Toluca.
2006 Davíd contra Goliat: o de cómo la pequeña counidad de San Mateo Atenco
venció jurídicamente al gran Marquesado de Valle en los siglos XVI y XVII. In La
proeza histórica de un pueblo, San Mateo Atenco en el valle de Toluca, siglos VIII-XIX,
edited by R. García Castro and M. T. Jarquín Ortega, pp. 37-75. El Colegio Mexiquense
and Universidad Autónoma del Estado de México, Toluca.
García Payón, José
1932 Calixtlahuaca Monumental. Acción Social 3(349):8-13.
1936 La zona arqueológica de Tecaxic-Calixtlahuaca y los matlatzincas:
etnología y arqueología (primera parte). Talleres Gráficas de la Nación, Mexico City.
1938 La escultura de Ehecatl-Quetzalcoatl encontrada en Calixtlahuaca. Revista
de Revistas _____:__________-.
1941a La cerámica del valle de Toluca. Revista Mexicana de Estudios
Antropológicos 5:209-238.
1941b Manera de disponer de los muertos entre los matlatzincas del Valle de
Toluca. Revista Mexicana de Estudios Antropológicos 5:64-78.
1942
Matlatzincas o pirindas. El Nacional, Ediciones Encuadernables, Mexico
City.
1956/57
Síntesis de las investigaciones estratigráficas practicadas en
Tecaxic-Calixtlahuaca. Revista Mexicana de Estudios Antropológicos 14(2):161-165.
1979 La zona arqueológica de Tecaxic-Calixtlahuaca y los matlatzincas:
etnología y arqueología (textos de la segunda parte), edited by Wanda Tommasi de
Magrelli and Leonardo Manrique Castañeda. Biblioteca Enciclopédica del Estado de
México 30. Estado de México, Toluca.
1981 La zona arqueológica de Tecaxic-Calixtlahuaca y los matlatzincas:
etnología y arqueología (tablas, planos e ilustraciones de la segunda parte), edited by
Leonardo Manrique Castañeda. Biblioteca Enciclopédica del Estado de México 31.
Estado de México, Toluca.
García Sánchez, Magdalena A.
2008 Petates, peces y patos: pervivencia cultural y comercio entre México y
Toluca. El Colegio de Michoacán, Centro de Investigaciones y Estudios
Superiores en Antropología Social, Zamora, Mexico.
458
Garraty, Christopher P.
2000 Ceramic Indices of Aztec Eliteness. Ancient Mesoamerica 11(02):323340.
2006 The Politics of Commerce: Aztec Pottery Production and Exchange in the
Basin of Mexico, AD 1200-1650, School of Human Evolution and Social Change
(Anthropology), Arizona State University, Tempe, AZ.
2007 Intercambio de mercado y consolidación en el corazón del Imperio
Azteca. Revista Española de Antropología Americana 37:139-164.
2009 Evaluating the Distributional Approach in Inferring Marketplace
Exchange: A Test Case from the Mexican Gulf Lowlands. Latin American Antiquity
20(1):157-174.
2013 Market Development and Pottery Exchange Under Aztec and Spanish
Rule in Cerro Portezuelo. Ancient Mesoamerica 24(1):151-176.
Garraty, Christopher P. and Michael A. Ohnersorgen
2009 Negotiating the Imperial Landscape: The Geopolitics of Aztec Control in
the Outer Provinces of the Empire. In The Archaeology of Meaningful Places,
edited by B. J. Bowser and M. N. Zedeno, pp. 107-131. University of Utah Press,
Salt Lake City.
Garraty, Christopher P. and Barbara L. Stark
2002 Imperial and Social Relations in Postclassic South-Central Veracruz. Latin
American Antiquity 13:3-33.
2010 Archaeological Approaches to Market Exchange in Ancient Societies.
University Press of Colorado, Boulder.
Gasco, Janine
1992 Documentary and Archaeological Evidence for Household Differentiation
in Colonial Soconusco, New Spain. In Text-Aided Archaeology, edited by B. J.
Little, pp. 83-94. CRC Press, Boca Raton.
Gerring, John, Daniel Ziblatt, Johan Van Gorp and Julián Arévalo
2011 An Institutional Theory of Direct and Indirect Rule. World Politics
63(3):377-433.
Given, Michael
2004 The Archaeology of the Colonized. Routledge, London and New York.
Glascock, Michael D.
459
2012 Report on X-ray Fluorescence Analysis of Obsidian Artifacts from
Calixtlahuaca, State of Mexico, Mexico. Missouri University Research Reactor
Archaeometry Laboratory.
Global Price and Income History Group
2013 Global Price and Income History Group, edited by L. Arroyo Abad and P.
H. Lindert. vol. 2014, University of California-Davis.
Goldstone, Jack A. and John F. Haldon
2009 Ancient States, Empires, and Exploitation: Problems and Perspectives. In
The dynamics of ancient empires: state power from Assyria to Byzantium, edited
by I. Morris and W. Scheidel. Oxford University Press, Oxford and New York.
Golitko, Mark and Gary M. Feinman
2015 Procurement and Distribution of Pre-Hispanic Mesoamerican Obsidian
900 BC-AD 1520: a Social Network Analysis. Journal of Archaeological Method
and Theory 2015(22):206-247.
González, Roberto J.
2001 Zapotec science: farming and food in the Northern Sierra of Oaxaca
University of Texas Press, Austin.
González Rul, Francisco
1988a La cerámica en Tlatelolco. Colección Científica 172. Instituto Nacional de
Antropología e Historia, Mexico City.
1988b La cerámica postclásica y colonial en algunos lugares de la ciudad de
México y el área metropolitana. In Ensayos de alfarería Prehispánica e histórica de
Mesoamérica: Homenaje a Eduardo Noguera Auza, edited by M. C. Serra Puche and C.
Navarrete Cáceres, pp. 387-416. Instituto de Investigaciones Antropológicas, Universidad
Nacional Autónoma de México, Mexico City.
Green Carr, Lois and Lorena S. Walsh
1980 Inventories and the Analysis of Wealth and Consumption Patterns In St.
Mary's County, Maryland, 1658–1777. Historical Methods: A Journal of
Quantitative and Interdisciplinary History 13(2):81-104.
Green, Judith Strupp
2010 Feasting with Foam: Ceremonial Drinks of Cacao, Maize and Pataxte
Cacao. In Pre-Columbian Foodways: Interdisciplinary Approaches to Food,
Culture, and Markets in Ancient Mesoamerica, edited by J. E. Staller and M.
Carrasco, pp. 315-343. Springer, New York.
Haines, Helen R., Gary M. Feinman and Linda M. Nicholas
2004 Household Economic Specialization and Social Differentiation: The stonetool assemblage at El Palmillo, Oaxaca. Ancient Mesoamerica 15(02):251-266.
460
Hall, Thomas D.
1997 The Millennium Before the "Long Sixteenth Century:" How Many WorldSystems Were There? In Economic Analysis Beyond the Local System, edited by
R. E. Blanton, P. N. Peregrine, D. Winslow and T. D. Hall, pp. 43-70. University
Press of America, Lanham.
1998 The effects of incorporation into world-systems on ethnic processes:
lessons from the ancient world for the contemporary world. International Political
Science Review 19(3):251-267.
Hall, Thomas D. and Christopher Chase-Dunn
1996 Comparing World-Systems: Concepts and Hypotheses. In Pre-Columbian
World Systems, edited by P. N. Peregrine and G. M. Feinman, pp. 11-26.
Prehistory Press, Madison, WI.
Hall, Thomas D., P. Nick Kardulias and Christopher Chase-Dunn
2011 World-Systems Analysis and Archaeology: Continuing the Dialogue.
Journal of Archaeological Research 19:233-279.
Haller, Mikael J., Gary M. Feinman and Linda M. Nicholas
2006 Socieconomic inequality and differential access to faunal resources at El
Palmillo, Oaxaca, Mexico. Ancient Mesoamerica 17(01):39-55.
Hassig, Ross
1985 Trade, Tribute, and Transportation: The Sixteenth-Century Political
Economy of the Valley of Mexico. University of Oklahoma Press, Norman, OK.
1988 Aztec Warfare: Imperial Expansion and Political Control. University of
Oklahoma Press, Norman.
Hay, Conran Alexander
1978 Kaminaljuyu Obsidian: Lithic Analysis and the Economic Organization of
a Prehistoric Mayan Chiefdom, Department of Anthropology, The Pennsylvania
State University.
Healan, Dan M. (editor)
1989 Tula of the Toltecs. University of Iowa Press, Iowa City.
Healan, Dan M., Janet M. Kerley and George J. Bey III
1983 Excavation and Preliminary Analysis of an Obsidian Workshop in Tula,
Hidalgo, Mexico. Journal of Field Archaeology 10(2):127-145.
Hechter, Michael
2013 Alien Rule. Cambridge Studies in Comparative Politics. Cambridge
University Press, New York, NY.
461
Hegmon, Michelle, Margaret C. Nelson and Mark J. Ennes
2000 Corrugated Pottery, Technological Style, and Population Movement in the
Mimbres Region of the American Southwest. Journal of Anthropological
Research 56(2):217-240.
Hendon, Julia A.
2003 Feasting at Home: Community and House Solidarity Among the Maya of
Southeastern Mesoamerica. In The Archaeology and Politics of Food and
Feasting in Early States and Empires, edited by T. L. Bray, pp. 203-234. Kluwer
Publshers, New York.
Henrich, Joseph, Robert Boyd, Samuel Bowles, Colin Camerer, Ernst Fehr and Herbert
Gintis
2004 Foundations of Human Sociality-Economic Experiments and
Ethnographic: Evidence From Fifteen Small-Scale Societies.
Hernández Rodríguez, Rosaura
1988 El Valle de Toluca: Epoca Prehispanica y siglo XVI. Colegio Mexiquense,
Toluca.
1998 La conquista Mexica del valle de Toluca. In Historia general del Estado
de México, vol. 2: época prehispnáica y siglo XVI, edited by R. Hernández Rodríguez,
pp. 403-415. Gobierno del Estado de México and El Colegio Mexiquense, Toluca.
2011 La tradición oral del Valle de Toluca. In Historia general ilustrada del
Estado de México, tomo 2, etnohistoria, edited by R. Hernández Rodríguez and R. C.
Martínez García, pp. 179-197. El Colegio Mexiquense, Toluca.
Hicks, Frederic
1987 First Steps Toward a Market-Integrated Economy in Aztec Mexico. In
Early State Dynamics, edited by H. J. M. Classen and P. van de Velde, pp. 91107. E. J. Brill, New York.
1994 Cloth in the Political Economy of the Aztec State. In Economies and
Polities in the Aztec Realm, edited by M. G. Hodge and M. E. Smith, pp. 89-111. Institute
for Mesoamerican Studies, SUNY Albany, NY.
1996 Class and state in Aztec Official Ideology. In Ideology and the Formation
of Early States, edited by H. J. M. Claessen and J. G. Oosten, pp. 256-277. E.J. Brill,
Leiden.
Hirth, Kenneth G.
1993 Identifying Rank and Socioeconomic Status in Domestic Contexts: An
Example from Central Mexico. In Prehispanic Domestic Units in Western
462
Mesoamerica: Studies of the Household, Compound, and Residence, edited by R.
S. Santley and K. G. Hirth, pp. 121-146. CRC Predd, Boca Raton.
1998 The Distributional Approach: A New Way to Identify Marketplace
Exchange in the Archaeological Record. Current Anthropology 39(4):451-476.
2006a Modeling a Prehistoric Economy: Mesoamerican Obsidian Systems and
Craft Production at Xochicalco. In Obsidian Craft Production in Ancient Central Mexico,
edited by K. G. Hirth, pp. 287-300. University of Utah Press, Salt Lake City.
2006b Modeling Domestic Craft Production at Xochicalco. In Obsidian Craft
Production in Ancient Central Mexico, edited by K. G. Hirth, pp. 275-286. University of
Utah Press, Salt Lake City.
2008 The Economy of Supply: Modeling Obsidian Procurement and Craft
Provisioning at a Central Mexican Urban Center. Latin American Antiquity 19(4):435457.
2009a Craft Production, Household Diversification, and Domestic Economy in
Prehispanic Mesoamerica. In Housework: Craft Production and Domestic Economy in
Ancient Mesoamerica, edited by K. G. Hirth, pp. 13-32. Archaeological Papers of the
American Anthropological Association, Number 19, C. Costin, general editor.
2009b Intermittent Crafting and Multicrafting at Xochicalco. In Housework:
Craft Production and Domestic Economy in Ancient Mesoamerica, edited by K. G. Hirth,
pp. 75-91. Archaeological Papers of the American Anthropological Association, Number
19, C. Costin, general editor.
2013 The Merchant's World: Commerical Diversity and the Economics of
Interregional Exchange in Highland Mesoamerica. In Merchants, Markets, and Exchange
in the Pre-Columbian World, edited by K. G. Hirth and J. Pillsbury, pp. 85-112.
Dumbarton Oaks, Washington D.C.
Hirth, Kenneth G. and Ronald A. Castanzo
2006 Production for Use or Exchange: Obsidian Consumption at the Workshop,
Household, and Regional Levels. In Obsidian Craft Production in Ancient
Central Mexico, edited by K. G. Hirth, pp. 218-240. University of Utah Press, Salt
Lake City.
Hodge, Mary G. (editor)
2008 Un Lugar de Jade: Sociedad y Economía en el Antiguo Chalco/ Place of
Jade: Society and Economy in Ancient Chalco. Instituto Nacional de
Antropología e Historia, Mexico City and University of Pittsburgh, Pittsburgh.
Hodge, Mary G. and Leah D. Minc
463
1990 The Spatial Patterning of Aztec Ceramics: Implications for Prehispanic
Exchange Systems in the Valley of Mexico. Journal of Field Archaeology
17:415-437.
1991 Aztec-Period Ceramic Distribution and Exchange Systems. National
Science Foundation.
Hodge, Mary G., Hector Neff, M. James Blackman and Leah D. Minc
1992 A Compositional Perspective on Ceramic Production in the Aztec Empire.
In Chemical Characterization of Ceramic Pastes in Archaeology, edited by H.
Neff, pp. 203-220. Prehistory Press, Madison, WI.
Hodos, Tamar
2009 Colonial Engagements in the Global Mediterranean Iron Age. Cambridge
Archaeological Journal 19(2):221-241.
Hollander, Jocelyn A. and Rachel L. Einwohner
2004 Conceptualizing Resistance. Sociological Forum 19(4):533-554.
Huster, Angela C.
2009 Postclassic Maguey Cloth Specialization in Urban Contexts. MA Paper,
MA Portfolio Paper, School of Human Evolution and Social Change, Arizona
State University, Tempe.
2013 Assessing Systematic Bias in Museum Collections: A Case Study of
Spindle Whorls. Advances In Archaeological Practice 1(2):77-90.
2015 Postclassic Ceramic Exchange at Aztec Calixtlahuaca and Surrounding
Areas of Central Mexico. Paper presented at the 80th Annual Meeting, Society for
American Archaeology, San Francisco, CA.
Huster, Angela C. and Michael E. Smith
2015 A New Archaeological Chronology for Aztec-Period Calixtlahuaca,
Mexico. Latin American Antiquity 26(1):3-25.
Huster, Angela C., Michael E. Smith and Juliana Novic
2015 Artefactos Rituales de Contextos Públicos y Domésticos en Calixtlahuaca.
In Bajo el volcán. Vida y ritualidad en torno al Nevado de Toluca, edited by P.
Luna, A. Montero and R. Junco, pp.?? INAH, México.
Hutson, Scott R.
2002 Built Space and Bad Subjects: Domination and Resistance at Monte
Alban, Oaxaca, Mexico. Journal of Social Archaeology 2(1):53-80.
Insoll, Timothy
2004 Archaeology, ritual, religion. Psychology Press.
464
Jaramillo Lunque, R. and R. De la Peña Virches
2012 Proyecto: Investigación y Conservación del Sito Arqueológico del Cerro
Toloche y Área Circunvecina: Informe Parcial, 2a Temporada, Abril-Junio 2012.
Instituto Mexiquense de Cultura.
2014 Proyecto: Investigación y Conservación del Sitio Arqueológico del Cerro
Toloche y Área Circunvecina: Informe de la Tercera Temporada, Noviembre 2013Diciembre 2014. Instituto Mexiquense de Cultura.
Jennings, Justin, Kathleen L. Antrobus, Sam J. Atencio, Erin Glavich and Rebecca
Johnson
2005 "Drinking beer in a blissful mood": alcohol production, operational chains,
and feasting in the ancient world. Current Anthropology 46(2):275-303.
Johnson, Kathryn A., Andrew E. White, Brenna M. Boyd and Adam B. Cohen
2011 Matzah, Meat, Milk, and Mana: Psychological Influences on ReligioCultural Food Practices. Journal of Cross-Cultural Psychology 42:1421-1436.
Jones, Michael Owen
2007 Food Choice, Symbolism, and Identity: Bread and Butter Issues for
Folkloristics and Nutrition Studies (American Folklore Society Presidential
Address, October 2005). Journal of American Folklore 120(476):129-177.
Kabata, Shigeru
2009 La industria de obsidiana y su abasticimiento a Santa Cruz Atizapán. In La
gente de la ciénaga en tiempos antiguos: la historia de Santa Cruz Atizapán,
edited by Y. Sugiura Yamamoto, pp. 243-260. Universidad Nacional Autónoma
de México, Instituto de Investigaciones Antropológicas, DGAPA.
2010 La Dinámica Regional Entre el Valle de Toluca y las Áreas Circundantes:
Intercambio Antes y Después de la Caída de Teotihuacan, Instituto de Investigaciones
Antropológicas, UNAM, Mexico, D.F.
Karabowicz, Amy
2009 Wattle and Daub Architecture at Calixtlahuaca, Mexico. Paper presented
at the Society for American Archaeology Annual Meeting, Atlanta, Georgia.
Kedourie, Elie
1960 Nationalism. Praeger, London.
Kelly, Isabel Truesdell
1947 Excavations at Apatzingan. Viking Fund Publications in Archaeology,
New York.
Kepecs, Susan and Philip Kohl
465
2003 Conceptulizing Macroregional Interaction: World-Systems Theory and the
Archaeological Record. In The Postclassic Mesoamerican World, edited by M. E.
Smith and F. F. Berdan. University of Utah Press, Salt Lake City.
King, Stacie M.
2011 Thread Production in Early Postclassic Coastal Oaxaca, Mexico:
Technology, Intensity, and Gender. Ancient Mesoamerica 22:323-343.
Klein, Cecelia F. and Naoli Victoria Lona
2009 Sex in the City: A Comparison of Aztec Ceramic Figurines to Copal
Figurines from the Templo Mayor. In Mesoamerican Figurines: Small-Scale
Indices of Large-Scale Social Phemonena, edited by C. T. Halperin, K. A. Faust,
R. Taube and A. Giguet, pp. 327-377. University Press of Florida, Gainsville, FL.
Kohl, Philip L.
1987 The Use and Abuse of World Systems Theory: The Case of the Pristine
West Asian State. Advances in Archaeological Method and Theory 11(1):1-35.
Kowalewski, Stephen, Luis Barba Pingarrón, Gabriela García Ayala, Benjamin A. Steere,
Jorge Blancas Vásquez, Marisol Yadira Cortés Vilchis, Leonardo López Zárate, Agustín
Ortiz Butrón, Thomas J. Pluckhahn and Blanca Vilchis Flores
2010 La Presencia Azteca en Oaxaca: La Provincia de Coixtlahuaca. Anales de
Antropología 44:77-103.
Kuznar, Lawrence A.
2001 Risk Sensitivity and Value among Andean Pastoralists: Measures, Models,
and Empirical Tests1. Current Anthropology 42(3):432-440.
La Ferrara, E.
2003 Kin Groups and Reciprocity: A Model of Credit Transactions in Ghana.
American Economic Review 93(5):1730-1751.
Laitin, David D.
1998 Identity in formation: The Russian-speaking populations in the near
abroad 22. Cambridge University Press.
Landa, Diego de
1941 Landa's Relación de las Cosas de Yucatan. Translated by A. M. Tozzer.
Papers, No. 18. Peabody Museum of Archaeology and Ethnology, Harvard
University, Cambridge.
Langenwalter, P. E.
1980 The Archaeology of 19th-Century Chinese Subsistence at the Lower
China Store, Madera County, California. In Archaeological Perspectives on
Ethnicity in America, edited by R. L. Schuyler, pp. 102-112. Baywood,
Farmingdale, NY.
466
Lemmonier, Pierre
1986 The Study of Material Culture Today: Toward an Anthropology of
Technical Systems. Journal of Anthropological Archaeology 5:147-186.
Lemonnier, Pierre (editor)
1993 Technological Choices: Transformation in Material Culture Since the
Neolithic. Routledge, New York.
Lentz, David L.
1991 Maya Diets of the Rich and Poor: Paleoethnobotanical Evidence from
Copan. Latin American Antiquity 2:269-287.
Lévi-Strauss, Claude
1969 The Raw and the Cooked. Harper and Row, New York.
Levine, Marc N.
2011 Negotiating Political Economy at Late Postclassic Tututepec (Yucu Dzaa),
Oaxaca, Mexico. American Anthropologist 113(1):22-39.
Levine, Marc N., Lane F. Fargher, Leslie G. Cecil and Jamie E. Forde
2015 Polychrome Pottery Economics and Ritual Life in Postclassic Oaxaca,
Mexico. Latin American Antiquity 26(3):319-340.
Liebmann, Matthew, T. J. Ferguson and Robert W. Preucel
2005 Pueblo settlement, architecture, and social change in the Pueblo revolt era,
A.D. 1680-1696. Journal of Field Archaeology 30(1):45-60.
Liebmann, Matthew and Melissa S. Murphy (editors)
2011 Enduring Conquests: Rethinking the Archaeology of Resistance to Spanish
Colonialism in the Americas. SAR Press, Santa Fe, NM.
Livarda, Alexandra
2011 Spicing up life in northwestern Europe: exotic food plant imports in the
Roman and medieval world Vegetation History and Archaeobotany 20:143-164.
Lockhart, James
1992 The Nahuas After the Conquest. Stanford University Press, Stanford.
López Austin, Alfredo and Leonardo López Luján
2009 Monte Sagrado- Templo Mayor. INAH/UNAM, Mexico.
López Luján, Leonardo
2005 The Offerings of the Templo Mayor of Tenochtitlan. Revised ed.
Translated by B. R. Ortiz de Montellano and T. Ortiz de Montellano. University
of New Mexico Press, Albuquerque.
467
Luttwak, Edward N.
1976 The Grand Strategy of the Roman Empire From the First Century A.D. to
the Third. Johns Hopkins University Press, Baltimore.
Mahoney, James
2010 Colonialism and postcolonial development: Spanish America in
comparative perspective. Cambridge University Press.
Malkin, Irad
2002 A Colonial Middle Ground: Greek, Etruscan, and Local Elites in the Bay
of Naples. In The Archaeology of Colonialism, edited by C. L. Lyons and J. K.
Papadopoulos, pp. 151-181. Getty Publications, Los Angeles.
2005 Herakles and Melqart: Greeks and Phoenicians in the Middle Ground. In
Cultural Borrowings and Ethnic Appropriations in Antiquity, edited by E. S. Gruen, pp.
238-257. F. Steiner, Stuttgart
Manzanilla, Linda
2004 Social Identity and Daily Life at Classic Teotihuacan. In Mesoamerican
Archaeology: Theory and Practice, edited by J. A. Hendon and R. A. Joyce, pp.
124-147. Blackwell, Malden.
2011 Sistemas de Control de Mano de Obra y del Intercambio de Bienes
Suntuarios en el Corredor Teotihuacano Hacia la Costa del Golfo en el Clásico. Anales de
Antropología 45:9-32.
2015 Cooperation and tensions in multiethnic corporate societies using
Teotihuacan, Central Mexico, as a case study. Proceedings of the National Academy of
Sciences:201419881.
Mata-Míguez, Jaime, Lisa M. Overholtzer, Enrique Rodríguez-Alegría, Brian M. Kemp
and Deborah A. Bolnick
2012 The Genetic Impact of Aztec Imperialism: Ancient Mitochondrial DNA
Evidence from Xaltocan, Mexico. American Journal of Physical Anthropology
149(4):504-516.
Matrícula de Tributos
1980 Matrícula de Tributos (Códice de Moctezuma). Akademische Druck u
Verlagsanstalt, Graz.
McCafferty, Geoffrey G.
2001 Ceramics of Postclassic Cholula, Mexico: typology and seriation of
pottery from the UA-1 domestic compound. Monograph. Cotsen Institute of
Archaeology at University of California, Los Angeles.
468
McCafferty, Sharisse D. and Geoffrey G. McCafferty
2000 Textile Production in Postclassic Cholula, Mexico. Ancient Mesoamerica
11:39-54.
McGuire, Randall H.
1982 The Study of Ethnicity in Historical Archaeology. Journal of
Anthropological Archaeology 1:159-178.
1993 The structure and organization of Hohokam exchange. In The American
Southwest and Mesoamerica, pp. 95-119. Springer.
McVicker, Donald, Laurene Lambertino-Urquizo, Michael D. Glascock and Hector Neff
n.d.
Regional Production for the International Market: Aztec Black-on-Orange
Ceramics from Tlacotepec, Valley of Toluca, Mexico. In The Archaeology of
Production, Distribution, and Consumption in Postclassic Central and Western
Mexico: Contributions from Materials Composition Analysis, edited by T. H.
Charlton, D. L. Nichols and H. A. Neff. (book in preparation).
Meanwell, Jennifer
2013a A Petrographic Analysis of Domestic Pottery Consumption at
Calixtlahuaca. Paper presented at the 78th Annual Meeting, Society for American
Archaeology, Honolulu, HI.
2013b Petrographic Analysis of Surface Collected Sherds from Calixtlahuaca.
Submitted to Report to the Calixtlahuaca Archaeological Project.
2014 Final Report: Calixtlahuaca INAA-Petrography Comparison. Submitted
to Report to the Calixtlahuaca Archaeological Project.
Mendieta, Gerónimo de
1945 [ca. 1571-1596] Historia eclesiástica indiana, compuesta por el padre fray
Gerónimo de Mendieta . . . con algunas advertencias del p. fray Joan de
Domanyquíar el padre fray Gerónimo de Mendieta . . . sacadas de cartas y otros
borradores del autor. 4 vols. Salvador Chávez Hayhoe, Mexico City.
Mendoza Cerón, Isaías and Una Canger
1993 In Tequil de Morrales: Working With Maguey. Bianco Luno, Copenhagen.
Mendoza, Codex
1992 [1541] The Codex Mendoza, edited by F. F. Berdan and P. R. Anawalt. 4
vols. University of California Press, Berkeley.
Miller, Daniel
1985 Artifacts as Categories: A study of ceramic variability in Central India.
New Studies in Archaeology. Cambridge University Press, Cambridge.
469
Miller, George L.
1988 Classification and economic scaling of nineteenth-century ceramics. In
Documentary Archaeology in the New World, edited by M. C. Beaudry, pp. 172183. New Directions in Archaeology. Cambridge University Press, Cambridge.
Miller, Katherine A.
2015 Family, ‘Foreigners’, and Fictive Kinship: A Bioarchaeological Approach
to Social Organization at Late Classic Copan. PhD in Anthropology, School of
Human Evolution and Social Change, Arizona State University, Tempe.
Millett, P.
2001 Productive to Some Purpose? The Problem of Ancient Economic Growth.
In Economies Beyond Agriculture in the Classical World, edited by D. J.
Mattingly and J. Salmon. Routledge, London.
Millhauser, John
2005 Classic and Postclassic Chipped-Stone at Xaltocan. In La producción local
y el poder en el Xaltocan Posclásico / Production and Power at Postclassic
Xaltocan, edited by E. M. Brumfiel, pp. 267-318. Arqueología de México.
Instituto Nacional de Antropología e Historia, Mexico City and University of
Pittsburgh, Pittsburgh.
Millhauser, John K., Lane F. Fargher, Verenice Y. Heredia Espinoza and Richard E.
Blanton
2015 The geopolitics of obsidian supply in Postclassic Tlaxcallan: A portable
X-ray fluorescence study. Journal of Archaeological Science (in press).
Mills, Barbara J.
2016 Communities of Consumption. In Knowledge in Motion: Constellations of
Learning Across Time and Place, pp. 247-270.
Millward, R.
1981 The Emergence of Wage Labor in Early Modern England. Explorations in
Economic History 18:21-39.
Minc, Leah D.
2009 Style and Substance: Evidence for Regionalization within the Aztec
Market System. Latin American Antiquity 20:343-374.
Minc, Leah D., Mary G. Hodge and M. James Blackman
1994 Stylistic and Spatial Variability in Early Aztec Ceramics: Insights into
Pre-Imperial Exchange Systems. In Economies and Polities in the Aztec Realm,
edited by M. G. Hodge and M. E. Smith, pp. 133-174. Studies on Culture and
Society. vol. 6. Institute for Mesoamerican Studies, Albany.
Mintz, Sindey W. and Christine M. Du Bois
470
2002 The Anthropology of Food and Eating. Annual Review of Anthropology
31:99-119.
Morris, Ian
2004a Archaeology, Standards of Living, and Greek Economic History. In The
Ancient Economy: Evidence and Models, edited by J. G. Manning and I. Morris,
pp. 91-126. Stanford University Press, Stanford.
2004b Economic Growth in Ancient Greece. Journal of Institutional and
Theoretical Economics 160:709-742.
Morrison, Kathleen D.
1994 States of Theory and States of Asia: Regional Perspectives on States in
Asia. Asian Perspectives 33(2):183-196.
Mortensen, Justin
2014 Grinding Stones in Post-Classic Morelos and the Toluca Valley, Senior
Honors Thesis, Barrett Honors College, Arizona State University.
n.d.
Piedra de moler. In Calixtlahuaca, organización de un centro urbano
posclásico: Informe final, edited by M. E. Smith. Report in preparation for submission to
the Consejo de Arqueología, Instituto Nacional de Antropología e Historia.
Murakami, Tatsuya
2010 Power relations and urban landscape formation: A study of construction
labor and resources at Teotihuacan, School of Human Evolution and Social
Change, Arizona State University, Tempe.
Murillo Rodríguez, Silvia
2002 La vida a través de la muerte: estudio biocultural de las funerarias en el
Temazcaltepec prehispánico. Instituto Nacional de Antropología e Historia,
Mexico City.
Neff, Hector
2014 Pots as signals: Explaining the enigma of long-distance ceramic exchange.
In Craft and science: International perspectices on archaeological ceramics,
edited by M. Martinón-Torres, pp. 1-11. UCL Qatar Series in Archaeology and
Cultural Heritage. Bloomsbury Qatar Foundation, Doha, Qatar.
Nichols, Deborah L.
2013 Merchants and Mechandise: the Archaeology of Aztec Commerce at
Otumba, Mexico. In Merchants, Markets, and Exchange in the Pre-Columbian
World, edited by K. G. Hirth and J. Pillsbury, pp. 49-83, J. Pillsbury and M. E.
Pye, general editor. Dumbarton Oaks, Washington D.C.
471
Nichols, Deborah L., Elizabeth M. Brumfiel, Hector Neff, Mary G. Hodge and Thomas
H. Charlton
2002 Neutrons, Markets, Cities and Empires: a 1000-year perspective on
ceramic production and distribution in the Postclassic Basin of Mexico. Journal of
Anthropological Archaeology 21(1):25-82.
Nichols, Deborah L., Christina Elson, Leslie G. Cecil, Nina Neivens de Estrada, Michael
D. Glascock and Paula Mikkelsen
2009 Chiconautla, Mexico: A Crossroads of Aztec Trade and Politics. Latin
American Antiquity 20(3):443-472.
Nichols, Deborah L., Mary Jane McLaughlin and Mara Benton
2000 Production Intensification and Regional Specialization: Maguey fibers and
textiles in the Aztec city-state of Otumba. Ancient Mesoamerica 11(2):267-291.
Nichols, Deborah L., Wesley D. Stoner and Destiny L. Crider
2014 A Geospatial Approach to the Development of Postclassic Markets:
Ceramic Production and Exchange. Paper presented at the 79th Annual Meeting,
Society for American Archaeology, Austin, TX.
Novic, Juliana
2015 Neighborhood Dynamics at Calixtlahuaca, Mexico, School of Human
Evolution and Social Change, Arizona State University, Tempe, AZ.
Ohnersorgen, Michael A.
2001 Social and Economic Organization of Cotaxtla in the Postclassic Gulf
Lowlands, Department of Anthropology, Arizona State University, Tempe,
Arizona.
2006 Aztec Provincial Administration at Cuetlaxtlan, Veracruz. Journal of
Anthropological Archaeology 25:1-32.
Oka, Rahul and Chapurukha M. Kusimba
2008 The Archaeology of Trading Systems, Part 1: Towards a New Trade
Synthesis. Journal of Archaeological Research 16:339-395.
Olson, Jan Marie
2001 Unequal Consumption: A Study of Domestic Wealth Differentials in
Three Late Postclassic Mexican Communities. PhD dissertation, Department of
Anthropology, University at Albany, SUNY, Albany.
2007 A Socioeconomic Interpretation of Figurine Assemblages from Late
Postclassic Morelos, Mexico. In Commoner ritual and ideology in ancient Mesoamerica
edited by N. Gonlin and J. C. Lohse. University Press of Colorado, Boulder.
Olson, Jan Marie and Michael E. Smith
472
2016 Material Expressions of Wealth and Social Class at Aztec-Period Sites in
Morelos, Mexico. Ancient Mesoamerica 27:in press.
Olson, Mancur
1965 The Logic of Collective Action: Public Goods and The Theory of Groups.
Harvard University Press, Cambridge.
Ortman, Scott G.
2012 Winds from the North: Tewa orgins and historical anthropology.
University of Utah Press, Salt Lake City.
Osborne, Robin
1991 The Potential Mobility of Human Populations. Oxford Journal of
Archaeology 10(2):231-252.
Ostrom, Elinor
1990 Governing the commons: the evolution of institutions for collection action.
Cambridge University Press, Cambridge.
2007 Collective Action Theory. In The Oxford Handbook fo Comparative
Politics, edited by S. C. Stokes, pp. 186-208. Oxford University Press, Oxford.
Otis Charlton, Cynthia, Thomas H. Charlton and Deborah L. Nichols
1993 Aztec Household-Based Craft Production: Archaeological Evidence from
the City-state of Otumba, Mexico. In Prehispanic Domestic Units in Western
Mesoamerica: Studies of the Household, Compound, and Residence, edited by R.
S. Santley and K. G. Hirth, pp. 147-171. CRC Press, Boca Raton.
Overholtzer, Lisa M.
2012 Empires and Everyday Material Practices: A Household Archaeology of
Aztec and Spanish Imperialism at Xaltocan. PhD dissertation, Department of
Anthropology, Northwestern University, Evanston, IL.
2013 Archaeological Interpretation and the Rewriting of History:
Deimperializing and Decolonizing the Past at Xaltocan, Mexico. American
Anthropologist 115(3):481-495.
2014 A New Bayesian Chronology for Postclassic and Colonial Occupation at
Xaltocan, Mexico Radiocarbon 56(3):1077-1092.
2015 Agency, practice, and chronological context: A Bayesian approach to
household chronologies. Journal of Anthropological Archaeology 37:37-47.
Parker, Bradley J.
2003 Archaeological manifestations of empire: Assyria's imprint on
southeastern Anatolia. American journal of archaeology 107(4):525-557.
473
Parry, William J.
1994 Prismatic blade technologies in North America. In The organization of
North American prehistoric chipped stone tool technologies, edited by P. J. Carr,
pp. 87-98, Ann Arbor, MI.
2001 Production and exchange of obsidian tools in late Aztec city-states.
Ancient Mesoamerica 12(1):101-111.
2002 Aztec Blade Production Strategies in the Eastern Basin of Mexico. In
Pathways to Prismatic Blades: A Study in Mesoamerican Obsidian Core-Blade
Technology, edited by K. G. Hirth and B. W. Andrews, pp. 37-45. Cotsen Institute of
Archaeology, University of California, Los Angeles.
Parsons, Jeffrey R.
1966 The Aztec Ceramic Sequence in the Teotihuacan Valley, Mexico. 2 vols.
PhD, Anthropology, University of Michigan, Ann Arbor.
1971 Prehistoric settlement patterns in the Texcoco region, Mexico. University
of Michigan Press.
2008 Prehispanic Settlement Patterns in the Northwestern Valley of Mexico:
The Zumpango Region Memoirs, No. 45. Museum of Anthropology, University of
Michigan, Ann Arbor, Michigan.
Parsons, Jeffrey R. and Andrew J. Darling
2000 The role of maguey in the Mesoamerican tierra fría: ethnographic, historic
and archaeological perspectives. In The Archaeology of Drylands: Living at the
Margin, edited by G. Barker and D. Gilbertson, pp. 288-312. Routledge, London.
Parsons, Jeffrey R. and Mary Hrones Parsons
1990 Maguey Utilization in Highland Central Mexico: An Archaeological
Ethnography. Anthropological Papers 82. Museum of Anthropology, University
of Michigan, Ann Arbor.
Parsons, Jeffrey R. and Michael E. Whalen
1982 Prehispanic settlement patterns in the southern Valley of Mexico: the
Chalco-Xochimilco region 14. University of Michigan Museum.
Parsons, Mary Hrones
1972 Spindle Whorls from the Teotihuacan Valley, Mexico. In Miscellaneous
Studies in Mexican Prehistory, edited by M. W. Spence, J. R. Parsons and M. H.
Parsons, pp. 45-80. Anthropological Papers. vol. 45. Museum of Anthropology,
University of Michigan, Ann Arbor.
Paso y Tronasco, Francisco del (editor)
474
1905-1906
Papeles de Nueva España. Segunda Serie, geografía y estadistica. 7
vols. Tipográfico Sucesores de Rivandeneyra, Mexico.
Pastrana, Alejandro
1998 La explotación azteca de la obsidiana en la Sierra de las Navajas.
Colección Científica 383. Instituto Nacional de Antropología e Historia, Mexico
City.
Pastrana Cruz, Alejandro
2007 La Produción de la Obsidiana de la Triple Alizana en la Cuenca de
México. INAH, Mexico D.F.
Pennington, Campbell W.
1963 The Tarahumara of Mexico: their environment and material culture
University of Utah Press, Salt Lake City.
Peregrine, Peter N.
1996 Introduction: World-Systems Theory and Archaeology. In Pre-Columbian
World Systems, edited by P. N. Peregrine and G. M. Feinman, pp. 1-10. Prehistory
Press, Madison, WI.
2012 Power and Legitimation: Political Strategies, Typology, and Cultural
Evolution. In The Comparative Archaeology of Complex Societies, edited by M. E.
Smith, pp. 165-191. Cambridge University Press, Cambridge.
Pérez Rodríguez, Verónica
2003 Household Intensification and Agrarian States: Excavation of Houses and
Terraced Fields in a Mixtec Cacicazgo. Ph.D., Anthropology, University of
Georgia, Athens.
2006 States and Households: The Social Organization of Terrace Agriculture in
Postclassic Mixteca Alta, Oaxaca, Mexico. Latin American Antiquity 17:3-22.
Perodie, James R.
2001 Feasting for Prosperity: a Study of Southern Northwest Coast Feasting. In
Feasts: Archaeological and Ethnographic Perspectives on Food, Politics and
Power, edited by M. Dietler and B. Hayden, pp. 185-214. Smithsonian Institute
Press, Washington.
Phillips, Catherine Anderson
2002 Neglected Artifacts: A Study of Re-Worked Ceramic Sherds from the
Lake Pátzcuaro Basin, Mexico, MA Thesis, Department of Anthropology,
Michigan State University, Ann Arbor, MI.
Piña Chán, Román (editor)
475
1975 Teotenango: el antiguo lugar de la muralla. 2 vols. Gobierno del Estado
de México, Mexico City.
1981 Investigaciones sobre Huamango y región vecina (Memoria del Proyecto).
2 vols. Dirección de Turismo del Gobierno del Estado de México, Toluca.
Piperata, Barbara Ann
2008 Forty days and forty nights: A biocultural perspective on postpartum
practices in the Amazon. Social Science and Medicine 67(7):1094-1103.
Pizzigoni, Caterina
2007 Testaments of Toluca. UCLA Latin American Studies 90. Stanford
University Press, Stanford.
2013 The Life Within: Local Indigenous Society in Mexico's Toluca Valley,
1650-1800. Stanford University Press, Stanford.
Polanyi, Karl, Conrad M. Arensberg and Harry W. Pearson (editors)
1957 Trade and Market in the Early Empires. Henry Regnery, Chicago.
Pollard, Helen Perlstein
1993 Tariacuri's Legacy: The Prehispanic Tarascan State. University of
Oklahoma Press, Norman.
2000 Tarascan External Relationships. In Greater Mesoamerica: The
Archaeology of West and Northwest Mexico, edited by M. S. Foster and S. Gorenstein,
pp. 71-80. University of Utah Press, Salt Lake City.
2016 Ruling "Purépecha Chichimeca" in a Tarascan World. In Political
Strategies in Pre-Columbian Mesoamerica, edited by S. Kurnick and J. Baron, pp. 217240. University Press of Colorado, Boulder, CO.
Pope, Kevin O., Mary E. D. Pohl, John G. Jones, David L. Lentz, Christopher von Nagy,
Francisco J. Vega and Irvy R. Quitmyer
2001 Origin and Environmental Setting of Ancient Agriculture in the Lowlands
of Mesoamerica. Science 292:1370-1373.
Price, T. Douglas, Linda Manzanilla and William D. Middleton
2010 Immigration and the Ancient City of Teotihuacan in Mexico: a Study
Using Strontium Isotope Ratios in Human Bone and Teeth. Journal of
Archaeological Science 27:903-913.
Rebnegger, Karin J.
2010 Obsidian Production and Changing Consumption in the Lake Patzcuaro
Basin, Michoacan, Mexico. Ancient Mesoamerica 21(01):79-89.
476
2013 Obsidian Production and Consumption in the Tarascan State, Department
of Anthropology, Michigan State University.
Reinhold, Manfred
1981 Arqueología de Valle de Bravo, México. Biblioteca Enciclopédica del
Estado de México, vol. CVIII. Estado de México, Toluca.
Renfrew, Colin
1994 The archaeology of religion. In The ancient mind: Elements of cognitive
archaeology, pp. 47-54.
Restall, Matthew
1995 Life and death in a Maya community: The Ixil testaments of the 1760s.
Labrynthos, Lancanster, CA.
Restall, Matthew, Lisa Sousa and Kevin Terraciano
2005 Mesoamerican voices: native-language writings from Colonial Mexico,
Oaxaca, Yucatan, and Guatemala Cambridge University Press, Cambridge, New
York.
Rice, Prudence M.
1987 Pottery Analysis: A Sourcebook. The University of Chicago Press,
Chicago.
Ringle, William M., Tomas Gallareta Negron and George J.III Bey
1998 The Return of Quetzalcoatl: Evidence for the spread of a world religion
during the Epiclassic period. Ancient Mesoamerica 9:183-232.
Robles García, Nelly M.
2012 Mexico's National Archaeology Programs. In The Oxford Handbook of
Mesoamerican Archaeology, edited by D. L. Nichols and C. A. Pool, pp. 47-54.
Oxford University Press, Oxford, UK.
Rodríguez-Alegría, Enrique
2008 Narratives of Conquest, Colonialism, and Cutting-Edge Technology.
American Anthropologist 110:33-43.
Rogers, J. D.
2005 Archaeology and the interpretation of colonial encounters. In The
Archaeology of Colonial Encounters: Comparative Perspectives, edited by G.
J. Stein, pp. 331-354. School of American Research Press, Santa Fe, NM.
Rojas Rabiela, Teresa, Elsa Leticia Rea López and Constantino Medina Lima
1999-2000
Vidas y bienes olvidados: testamentos indígenas novohispanos. 3
vols. CIESAS, Mexico, D.F.
477
Romero Padilla, Laura A.
2015 El componente cerámico en los actos litúrgicos del Nevado de Toluca. In
Bajo el volcán. Vida y ritualidad en torno al Nevado de Toluca, edited by P.
Luna, A. Montero and R. Junco, pp. 171-189. INAH, México.
Rosenblum, Jordan D.
2010 Food and Identity in Early Rabbinic Judaism. Cambridge University
Press, Cambridge.
Rouse, Irving
1986 Migrations in Prehistory: Inferring Population Movememt from Cultural
Remains. Yale University Press, New Haven.
Roux, Valentine
2003 Ceramic Standardization and Intensity of Production: Quantifying Degrees
of Specialization. American Antiquity 68(4):768-782.
Ruiz Medrano, Ethelia and Xavier Noguez (editors)
2004 Códice de Santiago Tlacotepec (Municipio de Toluca, Estado de México).
El Colegio Mexiquense and Instituto Mexiquense de Cultura, Toluca.
Sackett, J. R.
1977 The meaning of style: a general model. American Antiquity 42:369-380.
Sahagún, Bernardino de
1950-82
Florentine Codex, General History of the Things of New Spain. 12
books. Translated and Edited by Arthur J.O. Anderson and Charles E. Dibble.
School of American Research, Santa Fe NM, and the University of Utah Press,
Salt Lake City, UT.
Sanders, William T.
1956 The central Mexican symbiotic region: A study in prehistoric settlement
patterns. In Prehistoric Settlement Patterns in the New World, pp. 115-127.
Viking Fund Publications in Anthropology, New York.
1965 The cultural ecology of the Teotihuacan Valley: A preliminary report of
the results of the Teotihuacan Valley Project. Department of Sociology & Anthropology,
Pennsylvania State University.
Sanders, William T., Jeffery R. Parsons and Robert S. Santley
1979 The Basin of Mexico: Ecological Processes in the Evolution of a
Civilization. Academic Press, New York.
Sanders, William T. and Robert. S. Santley
1983 A Tale of Three Cities: Energetics and Urbanization in Pre-Hispanic
Central Mexico. In Prehistoric Settlement Patterns: Essays in Honor of Gordon
478
R. Willey, edited by E. Z. Vogt and R. M. Leventhal, pp. 243-291. University of
New Mexico Press and Peabody Museum of Archaeology and Ethnology,
Harvard University, Cambridge, MA.
Sanders, William T. and David Webster
1988 The Mesoamerican Urban Tradition. American Anthropologist 90(3):521546.
Sandstrom, Alan R.
2008 Blood Sacrifice, Curing, and Ethnic Identity Among Contemporary Nahua
of Northern Veracruz, Mexico. In Ethnic Identity in Nahua Mesoamerica: the
View from Archaeology, Art History, Ethnohistory, and Contemporary
Ethnography, edited by F. F. Berdan, J. K. Chance, A. R. Sandstrom, B. J. Stark,
J. Taggart and E. Umberger, pp. 150-182. University of Utah Press, Salt Lake
City.
2009 The Weeping Baby and the Nahua Corn Spirit: The Human Body as Key
Symbol in the Huasteca Veracruzana, Mexico. In Mesoamerican Figurines: Small-Scale
Indices of Large-Scale Social Phemonena, edited by C. T. Halperin, K. A. Faust, R.
Taube and A. Giguet, pp. 261-296. University Press of Florida, Gainsville, FL.
Santamarina Novillo, Carlos
2006 El sistema de dominación azteca: el imperio tepaneca, Departemento de
Historía de América II (Antropología de América), Universidad Complutense de
Madrid.
Santley, Robert. S., Philip J. Arnold III and Christopher A. Pool
1989 The Ceramic Production System at Matacapan, Veracruz, Mexico. Journal
of Field Archaeology 16:107-132.
Scheidel, Walter
2006 Republics between hegemony and empire: How ancient city-states built
empires and the USA doesn't (anymore). Version 1.0. Princeton/Stanford Working
Papers in Classics. Stanford University, Stanford, CA.
2010 Real Wages in Early Economies: Evidence for Living Standards from
1800 BCE to 1300 CE. Journal of the Economic & Social History of the Orient
53(3):425-462.
2015 Rome, Tenochtitlan, and beyond: comparing empires across space and
time, version 1.0. Princeton/Stanford Working Papers in Classics. Stanford University.
Schneider, Jane
1977 Was There a Pre-capitalist World-system? Peasant Studies 6:20-29.
Schortman, Edward M. and Patricia A. Urban
479
1992 Current Trends in Interaction Research. In Resources, Power, and
Interregional Interaction, edited by E. M. Schortman and P. A. Urban, pp. 235225. Plenum, New York.
1994 Living on the Edge: Core/Periphery Relations in Ancient Southeastern
Mesoamerica. Current Anthropology 35(4):401-430.
Scott, James C.
1990 Domination and the Arts of Resistance: Hidden Transcripts. Yale
University Press, New Haven.
Séjourné, Laurette
1970 Arqueología del Valle de México, I: Culhuacan. Instituto Nacional de
Antropología e Historia, Mexico City.
1983 Arqueologia e Historia del Valle de Mexico de Xochimilco a Amecameca.
Siglo Veintiuno Editores, Mexico City.
Sellet, Frederic
1993 Chaine Operatoire: the concept and its applications. Lithic Technology
18:106-112.
Sen, Amartya K.
1993 Capability and Well-Being. In The Quality of Life, edited by M. C.
Nussbaum and A. K. Sen, pp. 30-53. Clarendon Press, Oxford.
Sergheraert, Maëlle
2009 L'expansion mexica (1430-1520 après J.-C.): La question du contrôle
impérial dans les provinces extérieures de l'Empire, Department of Archaeology,
Université Paris 1, Panthéon-Sorbonne, Paris.
2011 Architectural style evolution: Calixtlahuaca's monumental structures
through ages. Paper presented at the 76th Annual Meeting, Society for American
Archaeology, Sacramento, CA.
Shackel, Paul A.
1992 Probate Inventories in Historical Archaeology: A Review and
Alternatives. In Text-Aided Archaeology, edited by B. J. Little, pp. 205-216. CRC
Press, Boca Raton.
Sheets, Payson D.
1992 The Ceren Site: A Prehistoric Village Buried by Ash in Central America.
Case Studies in Archaeology Series. Harcourt Brace Jovanovich College
Publishers, Fort Worth, TX.
480
Sheets, Payson D., David Lentz, Dolores Piperno, John Jones, Christine Dixon, George
Maloof and Angela Hood
2012 Ancient Manioc Agriculture South of the Ceren Village, El Salvador.
Latin American Antiquity 23(3):259-281.
Shimada, Izumi
2007 Craft Production in Complex Societies: Multicraft and Producer
Perspectives. University of Utah Press, Salt Lake City.
Silliman, Stephen W.
2001 Theoretical Perspectives on Labor and Colonialism: Reconsidering the
California Missions. Journal of Anthropological Archaeology 20(4):379-407.
Silverstein, Jay
2000 A Study of the Late Postclassic Aztec-Tarascan Frontier in Northern
Guerrero, Mexico: The Oztuma-Cutzamala Project, Department of Anthropology,
Pennsylvania State University, University Park.
2001
Aztec Imperialism at Oztuma, Guerrero. Ancient Mesoamerica 12(1):31-
48.
Sinopoli, Carla M.
1994 The Archaeology of Empires. Annual Review of Anthropology 23:159180.
2001 Imperial Integration and Imperial Subjects. In Empires: Perspectives from
Archaeology and History, edited by S. E. Alcock, T. N. D'Altroy, K. D. Morrison and C.
M. Sinopoli, pp. 195-200. Cambridge University Press, New York.
Siroky, David S., Valeriy Dzutsev and Michael Hechter
2013 The differential demand for indirect rule: evidence from the North
Caucasus. Post-Soviet Affairs 29(3):268-286.
Skinner, George W.
1977 Cities and the Hierarchy of Local Systems. In The City in Late Imperial
China, edited by G. W. Skinner, pp. 275-351. Stanford University Press, Stanford,
CA.
Skoglund, Thanet, Barbara L. Stark, Hector Neff and Michael D. Glascock
2006 Compositional and Stylistic Analysis of Aztec-Era Ceramics: Provincial
Strategies at the Edge of Empire, South-Central Veracruz, Mexico. Latin
American Antiquity 17:541-560.
Smith, Carol A.
481
1976 Regional Economic Systems: Linking Geographic Models and
Socioeconomic Problems. In Regional Analysis, volume 1, Economic Systems,
edited by C. A. Smith, pp. 3-63. Academic Press, New York.
Smith, Michael E.
1984 The Aztlan Migrations of the Nahuatl Chronicles: Myth or History?
Ethnohistory 31:153-186.
1987a The Expansion of the Aztec Empire: A Case Study in the Correlation of
Diachronic Archaeological and Ethnohistorical Data. American Antiquity 52(1):37-54.
1987b Household Possessions and Wealth in Agrarian States: Implications for
Archaeology. Journal of Anthropological Archaeology 6:297-335.
1990 Long-Distance Trade under the Aztec Empire: The Archaeological
Evidence. Ancient Mesoamerica 1(2):135-169.
1992a Archaeological Research at Aztec-Period Rural Sites in Morelos, Mexico.
Volume 1, Excavations and Architecture / Investigaciones Arqueológicas en Sitios
Rurales de la Época Azteca en Morelos, Tomo 1, Excavaciones y Arquitectura.
University of Pittsburgh Memoirs in Latin American Archaeology no. 4. University of
Pittsburgh, Pittsburgh.
1992b Braudel's temporal rhythms and chronology theory in archaeology. In
Archaeology, Annales, and Ethnohistory, edited by A. B. Knapp, pp. 23-34. Cambridge
University Press, New York.
1993 Economies and Polities in Aztec-Period Morelos: Ethnohistoric Overview.
In Economies and Polities in the Aztec Realm, edited by M. G. Hodge and M. E. Smith,
pp. 313-348. Institute for Mesoamerican Studies, Albany.
1999
40:528-530.
Comment on Hirth's "Distribution Approach". Current Anthropology
2002 Domestic Ritual at Aztec Provincial Sites in Morelos. In Domestic Ritual
in Ancient Mesoamerica, edited by P. Plunket, pp. 93-114. Monograph. vol. 46. Cotsen
Institute of Archaeology, UCLA, Los Angeles.
2003a The Aztecs. 2nd ed. The Peoples of America. Blackwell, Malden.
2003b Comercio postclásico en la cerámica decorada: Malinalco, Toluca,
Guerrero y Morelos. Arqueología (INAH) 29:63-84.
2003c Economic Change in Morelos Households. In The Postclassic
Mesoamerican World, edited by M. E. Smith and F. F. Berdan, pp. 249-258. University
of Utah Press, Salt Lake City.
482
2003d Information Networks in Postclassic Mesoamerica. In The Postclassic
Mesoamerican World, edited by M. E. Smith and F. F. Berdan, pp. 181-185. University
of Utah Press, Salt Lake City.
2003e Postclassic Urbanism at Calixtlahuaca: Reconstructing the Unpublished
Excavations of José García Payón. Report to the Foundation for the Advancement of
Mesoamerican Studies, Inc. Electronic document, http://www.famsi.org/reports/01024/.
2004a The Archaeology of Ancient State Economies. Annual Review of
Anthropology 33:73-102.
2004b Los Hogares de Morelos en el Sistema Mundial Mesoamericana
Postclásico. Relaciones 25(99):79-113.
2005 Los materiales arqueológicos de Calixtlahuaca y la sociedad Posclásica de
Matlatzinco. Paper presented at the VII Coloquio Internacional sobre Otopames, El
Colegio Mexiquense, Toluca, Mexico.
2006a Artefactos domésticos de casas postclásicas en Cuexcomate y Capilco,
Morelos. Report submitted to the Consejo de Arqueología, Instituto Nacional de
Antropología e Historia.
2006b Calixtlahuaca, organización de un centro urbano posclásico: Informe
Técnico Parcial, temporada de 2006. Consejo de Arqueología, Instituto Nacional de
Antropología e Historia.
2006c Excavaciones de casas postclásicas en la zona urbana de Yautepec,
Morelos: informe final. 2 vols. Report submitted to the Consejo de Arqueología, Instituto
Nacional de Antropología e Historia.
2008 Aztec City-State Capitals. Ancient Cities of the New World. University
Press of Florida, Gainsville.
2010 Regional and Local Market Systems in Aztec-Period Morelos. In
Archaeological Approaches to Market Exchange in Ancient Societies, edited by C. P.
Garraty and B. L. Stark, pp. 161-183. Boulder, University Press of Colorado.
2011 Proyecto: Calixtlahuaca: Organización de un Centro Urbano Posclásico:
Informe Técnico Parcial, Temporada de 2007. Report submitted to the Consejo de
Archaeologia, Instituto Nacional de Antropologia e Historia.
2014 Peasant Mobility, Local Migration and Premodern Urbanization. World
Archaeology 24(4):516-533.
483
2015a The Aztec Empire. In Fiscal Regimes and the Political Economy of
Premodern States, edited by A. Monson and W. Scheidel, pp. 71-114. Cambridge
University Press, Cambridge.
2015b Aztec Taxation at the City-State and Imperial Levels. In Fiscal Regimes
and the Political Economy of Early States, edited by A. Monson and W. Scheidel.
Cambridge University Press.
2015c How Can Archaeologists Make Better Arguments? SAA Archaeological
Record 2015(September):18-23.
2015d Quality of Life and Prosperity in Ancient Households and Communities.
In The Oxford Handbook of Historical Ecology and Applied Archaeology, edited by C.
Isendahl and D. Stump, pp. 1-15. Oxford University Press, Oxford.
Smith, Michael E., Patricia Aguirre, Cynthia Heath-Smith, Kathryn Hirst, Scott O'Mack
and Jeffrey Price
1989 Architectural Patterns at Three Aztec-Period Sites in Morelos, Mexico.
Journal of Field Archaeology 16(2):185-203.
Smith, Michael E. and Frances F. Berdan
1996a Appendix 4: Province Descriptions. In Aztec Imperial Strategies, edited by
F. F. Berdan, R. E. Blanton, E. H. Boone, M. G. Hodge, M. E. Smith and E.
Umberger, pp. 265-293. Dumbarton Oaks, Washington.
1996b Introduction. In Aztec Imperial Strategies, edited by F. F. Berdan, et al.,
pp. 151-180. Dumbarton Oaks, Washington, DC.
2003a Postclassic Mesoamerica. In The Postclassic Mesoamerican World, edited
by M. E. Smith and F. F. Berdan, pp. 3-13. University of Utah Press, Salt Lake City.
2003b The Postclassic Mesoamerican World. University of Utah Press, Salt Lake
City.
Smith, Michael E., Aleksander Borejsza, Angela C. Huster, Charles D. Frederick, Isabel
Rodríguez López and Cynthia Heath-Smith
2013 Aztec period houses and terraces at Calixtlahuaca: The changing
morphology of a Mesoamerican hilltop urban center. Journal of Field
Archaeology 38(3):225-241.
Smith, Michael E., Timothy Dennehy, April Kamp-Whittaker, Emily Colon and Rebecca
Harkness
2014 Quantitative Measures of Wealth Inequality in Ancient Central Mexican
Communities. Advances in Archaeological Practice 2(4):311-323(313).
Smith, Michael E. and Cynthia Heath-Smith
484
1980 Waves of Influence in Postclassic Mesoamerica? A Critique of the
Mixteca - Puebla Concept. Anthropology 4:15-50.
1993 Rural Economy in Late Postclassic Morelos: An Archaeological Study. In
Economies and Polities in the Aztec Realm, edited by M. G. Hodge and M. E. Smith, pp.
349-376. Institute for Mesoamerican Studies, Albany.
Smith, Michael E. and Kenneth G. Hirth
1988 The Development of Prehispanic Cotton-Spinning Technology in Western
Morelos, Mexico. Journal of Field Archaeology 15:249-358.
Smith, Michael E., Juliana Novic, Angela C. Huster and Peter C. Kroefges
2009 Reconocimiento superficial y mapeo en Calixtlahuaca en 2006. Expresión
Antropológica (Instituto Mexiquense de Cultura) 36:39-55.
Smith, Michael E. and Katharina J. Schreiber
2006 New World States and Empires: Politics, Religion, and Urbanism. Journal
of Archaeological Research 14:1-52.
Smith, Michael E., Jennifer B. Wharton and Jan Marie Olson
2003 Aztec Feasts, Rituals, and Markets: Political Uses of Ceramic Vessels in a
Commercial Economy. In The Archaeology and Politics of Food and Feasting in
Early States and Empires, edited by T. L. Bray, pp. 235-268. Kluwer Publishers,
New York.
Smith, Stuart Tyson
2003f Pharaohs, Feasts, and Foreigners: Cooking, Foodways, and Agency on
Egypt's Southern Frontier. In The Archaeology and Politics of Food and Feasting
in Early States and Empires, edited by T. L. Bray, pp. 39-64. Kluwer Publshers,
New York.
Somerville, Andrew D., Mikael Fauvelle and Andrew W. Froehle
2013 Applying new approaches to modeling diet and status: isotopic evidence
for commoner resiliency and elite variability in the Classic Maya lowlands.
Journal of Archaeological Science 40:1539-1553.
Spence, Michael W.
1996 Commodity or Gift: Teotihuacan Obsidian in the Maya Region. Latin
American Antiquity 7(1):21-39.
Spielmann, Katherine A., Tiffany Clark, Diane Hawkley, Katherine Rainey and Suzanne
K. Fish
2009 “…being weary, they had rebelled”: Pueblo subsistence and labor under
Spanish colonialism. Journal of Anthropological Archaeology 28:102-125.
Spielmann, Katherine A., Jeanette L. Mobley-Tanaka and James M. Potter
485
2006 Style and Resistance in the Seventeenth Century Salinas Province.
American Antiquity 71(4):621-647.
Stark, Barbara L., Matthew A. Boxt, Janine Gasco, Rebecca B. González Lauck, Jessica
D. Hedgepeth Balkin, Arthur A. Joyce, Stacie M. King, Charles L. F. Knight, Robert
Kruger, Marc N. Levine, Richard G. Lesure, Rebecca Mendelsohn, Marx NavarroCastillo, Hector Neff, Michael Ohnersorgen, Christopher A. Pool, L. Mark Raab, Robert
M. Rosenswig, Marcie Venter, Barbara Voorhies, David T. Williams and Andrew
Workinger
2016 Economic growth in Mesoamerica: Obsidian consumption in the coastal
lowlands. Journal of Anthropological Archaeology 41:263-282.
Stark, Barbara L. and John K. Chance
2008 Diachronic and Multidisciplinary Perspectives on Mesoamerican
Ethnicity. In Ethnic Identity in Nahua Mesoamerica: the View from Archaeology,
Art History, Ethnohistory, and Contemporary Ethnography, edited by F. F.
Berdan, J. K. Chance, A. R. Sandstrom, B. L. Stark, J. Taggart and E. Umberger,
pp. 1-37. University of Utah Press, Salt Lake City.
2012 The Strategies of Provincials in Empires. In The Comparative
Archaeology of Complex Societies, edited by M. E. Smith, pp. 192-237. Cambridge
University Press, Cambridge.
Stark, Barbara L. and Christopher P. Garraty
2004 Evaluation of Systematic Surface Evidence for Pottery Production in
Veracruz, Mexico. Latin American Antiquity 15(2):123-143.
Stark, Barbara L. and Alanna Ossa
2010 Origins and Development of Mesoamerican Marketplaces: Evidence from
South-Central Veracruz, Mexico. In Archaeological Approaches to Market
Exchange in Ancient Societies, edited by C. P. Garraty and B. L. Stark, pp. 99126. Boulder, University Press of Colorado.
Stark, Miriam T.
1991 Ceramic production and community specialization: A Kalinga
ethnoarchaeological study. World Archaeology 23(1):64-78.
Steckel, Richard H.
2008 Biological Measures of the Standard of Living. Journal of Economic
Perspectives 22(1):129-152.
Steckel, Richard H. and Jerome C. Rose (editors)
2002 The Backbone of History: Health and Nutrition in the Western
Hemisphere. Cambridge University Press, Cambridge.
Steere, Benjamin A. and Stephen A. Kowalewski
486
2012 Wealth Stratification in Ancient Mesoamerica. Social Evolution and
History 11(1):20-48.
Stein, Gil J.
2002a Colonies Without Colonialism: A Trade Diaspora Model of Fourth
Millennium B.C. Mesopotamian Enclaves in Anatolia. In The Archaeology of
Colonialism, edited by C. L. Lyons and J. K. Papadopoulos, pp. 27-64. Getty
Research Institute, Los Angeles.
2002b From Passive Periphery to Active Agents: Emerging Perspectives in the
Archaeology of Interregional Interaction. American Anthropologist 104:903-916.
2002c Introduction: The Comparative Archaeology of Colonial Encounters. In
The Archaeology of Colonial Encounters, pp. 3-31.
2012 Food Preparation, Social Context, and Ethnicity in a Prehistoric
Mesopotamian Colony. In The Menial Art of Cooking: Archaeological Studies of
Cooking and Food Preparation, edited by S. R. Graff and E. Rodríguez-Alegría, pp. 4764. University Press of Colorado, Boulder.
Steward, Julian H.
1938 Basin-Plateau Aboriginal Sociopolitical Groups. University of Utah Press
1997 Reprint ed. Smithsonian Institution, Bureau of American Ethnology
Stockett, Miranda K.
2005 Approaching Social Practice Through Access Analysis at Las Canoas,
Honduras. Latin American Antiquity 16(4):385-407.
Stoner, Wesley D. and Michael D. Glascock
2013 Neutron Activation Analysis of Pottery from the Toluca Valley. Missouri
University Research Reactor Archaeometry Laboratory.
Stoner, Wesley D., John K. Millhauser, Enrique Rodríguez-Alegría, Lisa Overholtzer and
Michael D. Glascock
2013 Taken with a Grain of Salt: Experimentation and the Chemistry of
Archaeological Ceramics from Xaltocan, Mexico. Journal of Archaeological
Method and Theory:1-37.
Stross, Brian
2010 Maize in Word and Image in Southeastern Mesoamerica. In Histories of
Maize in Mesoamerica, edited by J. E. Staller, R. H. Tykot and B. F. Benz, pp.
201-222. Left Coast Press, Walnut Creek.
Stuart, David
2000 "The Arrival of Strangers": Teotihuacan and Tollan in Classic Maya
History. In Mesoamerica's Classic Heritage: From Teotihuacan to the Aztecs,
487
edited by D. Carrasco, L. Jones and S. Sessions, pp. 465-513. University Press of
Colorado, Boulder.
Sugiura Yamamoto, Yoko
1998 El Valle de Toluca después del ocaso del estado Teotihuacano: el
Epiclásico y el Posclásico. In Historia general del Estado de México, vol. 1:
geografía y arqueología, edited by Y. Sugiura Yamamoto, pp. 199-259. Gobierno
del Estado de México and El Colegio Mexiquense, Toluca.
2000 Proyecto Arqueológico del Valle de Toluca: informe final. Informe
entragado al Consejo de Arqueología, Instituto Nacional de Antropología e Historia.
2005 Y atrás quedó la Ciudad de los Dioses: historia de los asentamientos en el
Valle de Toluca. Instituto de Investigaciones Antropológicas, Universidad Nacional
Autónoma de México, Mexico City.
2009 La gente de la ciénaga en tiempos antiguos: la historia de Santa Cruz
Atizapán. Universidad Nacional Autónoma de México, Instituto de Investigaciones
Antropológicas, DGAPA.
2011 El Valle de Toluca después del ocaso del estado Teotihuacano: el
Epiclásico y el Posclásico. In Historia general del Estado de México, vol. 1: geografía y
arqueología, edited by Y. Sugiura Yamamato, pp. 217-269. Gobierno del Estado de
México and El Colegio Mexiquense, Toluca.
Sullivan, Kristin S.
2006 Specialized Production of San Martín Orange Ware at Teotihuacan,
Mexico. Latin American Antiquity 17(1):23-53.
Tax, Sol
1953 Penny Capitalism: A Guatemalan Indian Economy. Publication 16.
Smithsonian Institution, Institute of Social Anthropology, Washington, DC.
Tilly, Charles
1978 Migration in Modern European History. In Human Migration: Patterns
and Policies, edited by W. McNeill and R. Adams. Indiana University Press,
Bloomington.
Tomaszewski, Brian M.
2006 A Cost-Effective Approach to GPS/ArcGIS Integration for Archaeological
Surveying. ArcUser Fall:http://www.esri.com/news/arcnews/fall06articles/a-costeffective.html.
Tomaszewski, Brian M. and Michael E. Smith
2011 Politics, Territory, and Historical Change in Postclassic Matlatzinco
(Toluca Valley, central Mexico). Journal of Historical Geography 37:22-39.
488
Tommasi de Magrelli, Wanda
1978 La cerámica funeraria de Teotenango. Biblioteca Enciclopédica del
Estado de México 61. Estado de México, Toluca.
Torquemada, Fray Juan de
1975-83
Monarquía indiana. 7 vols. Universidad Nacional Autónoma de
México, Mexico City.
Tozzer, Alfred M.
1907 A Comparative Study of the Mayas and the Lacandones. Macmillan, New
York.
Tronchetti, Carlo and Peter Van Dommelen
2005 Entangled Objects and Hybrid Practices: Colonial Contacts and Elite
Connections at Monte Prama Sardinia. Journal of Mediterranean Archaeology
18(2):183-208.
Tucker, Bram
2014 Rationality and the Green Revolution. In Applied Evolutionary
Anthropology: Darwinian Approaches to Contemporary World Issues, edited by
A. M. Gibson and W. D. Lawson, pp. 15-38. Springer New York, New York, NY.
Turkon, Paula
2004 Food and status in the prehispanic Malpaso Valley, Zacatecas, Mexico.
Journal of Anthropological Archaeology 23(2):225–251.
Twiss, Katheryn
2012 The Archaeology of Food and Social Diversity. Journal of Archaeological
Research 20:357-395.
Umberger, Emily
1996 Aztec Presence and Material Remains in the Outer Provinces. In Aztec
Imperial Strategies, edited by F. F. Berdan, et al., pp. 151-180. Dumbarton Oaks,
Washington, DC.
2008 Ethnicity and Other Identities in the Sculptures of Tenochtitlan. In Ethnic
Identity in Nahua Mesoamerica: the View from Archaeology, Art History, Ethnohistory,
and Contemporary Ethnography, edited by F. F. Berdan, J. K. Chance, A. R. Sandstrom,
B. J. Stark, J. Taggart and E. Umberger, pp. 64-104. University of Utah Press, Salt Lake
City.
Umberger, Emily and Casandra Hernández Fahan
submitted 2014
Matlatzinco Before the Aztecs: José García Payón and the
Sculptural Corpus of Calixtlahuaca. Ancient Mesoamerica:70p.
489
van der Veen, Marijke
2003 When is Food a Luxury? World Archaeology 34(3):405-427.
Van Dommelen, Peter
1997 Colonial Constructs: Colonialism and Archaeology in the Mediterranean.
World Archaeology 28(3):305-323.
2012 Colonialism and Migration in the Ancient Mediterranean. Annual Review
of Anthropology 41:393-409.
Vargas Pacheco, Ernesto
1975 La cerámica. In Teotenango: el antiguo lugar de la muralla, edited by R.
Piña Chán, pp. 189-264. vol. 1. Gobierno del Estado de México, Mexico City.
Velázquez V., Ricardo
1975 Miscelanea de Barro. In Teotenango: el antiguo lugar de la muralla,
edited by R. Piña Chán, pp. 319-331. vol. 2. Gobierno del Estado de México,
Mexico City.
Venter, Marcie L.
2008 Community Strategies in the Aztec Imperial Frontier: Perspectives from
Totogal, Veracruz, Mexico, Department of Anthropology, University of
Kentucky, Lexington.
2012 A Reassessment of the Extent of the Eastern Aztec Empire in the
Mesoamerican Gulf Lowlands. Ancient Mesoamerica 23(2):235-250.
Villanueva Villalpando, Jorge
1999 Informe técnico final: Restauración de la alfarda sur, fachada oriente, del
edificio no. 3 de la zona arqueológica de Calixtlahuaca. Instituto Nacional de
Antropología e Historia.
Vogt, Evon Z.
1969 Zinacantán: A Maya Community in the Highlands of Chiapas. Harvard
University Press, Cambridge.
Voorhies, Barbara and Janine Gasco
2004 Postclassic Soconusco Society: The Late Prehistory of the Coast of
Chiapas, Mexico. IMS Monograph 14. Institute for Mesoamerican Studies,
University at Albany, Albany, NY.
Voss, Barbara L.
2012 Status and Ceramics in Spanish Colonial Archaeology. Historical
Archaeology 46(2):39-52.
Wallerstein, Immanuel
490
1976 The Modern World-System: Capitalist Agriculture and the Origins of the
European World-Economy in the Sixteenth Century. Studies in Social
Discontinuity. Academic Press, New York.
Warren, Kea M.
2015 3,065 Sherd Disks and their Potential Uses at Calixtlahuaca, in the Toluca
Valley, Mexico (poster). Paper presented at the 80th Annual Meeting, Society for
American Archaeology, San Francisco, CA.
Wells, Peter S.
2005 Creating an Imperial Frontier: Archaeology of the Formation of Rome’s
Danube Borderland. Journal of Archaeological Research 13(1):49-89.
White, Christine D.
2005 Gendered food behaviour among the Maya : Time, place, status and ritual.
Journal of Social Archaeology 5(3):356–382.
White, Christine D., Fred J. Longstaffe and Henry P. Schwarcz
2010 Social Directions in the Isotopic Anthropology of Maize in the Maya
Region. In Histories of Maize in Mesoamerica, edited by J. E. Staller, R. H. Tykot
and B. F. Benz, pp. 142-158. Left Coast Press, Walnut Creek.
White, Christine D., Rebecca Storey, Fred J. Longstaffe and Michael W. Spence
2004 Immigration, Assimilation, and Status in the Ancient City of Teotihuacan:
Stable Isotopic Evidence from Tlajinga 33. Latin American Antiquity 15(2):176198.
White, Richard
1991 The Middle Ground: Indians, empires, and republics in the Great Lakes
region, 1650-1815. Cambridge University Press, Cambridge.
Whittington, Stephen and Andrew Workinger
2015 The Archaeology and History of Colonialism, Culture Contact, and
Indigenous Cultural Development at Teozocoalco, Mixteca Alta. In Bridging the
Gaps: Integrating Archaeology and History in Oaxaca, Mexico, edited by D.
Zborover and P. C. Kroefges, pp. 209-229. University Press of Colorado, Boulder,
CO.
Wiessner, Polly
1983 Style and Social Information in Kalahari San Projectile Points. American
Antiquity 48(2):253-276.
2001 Of Feasting and Value: Enga Feasts in a Historical Perspective (Papua
New Guinea). In Feasts: Archaeological and Ethnographic Perspectives on Food,
Politics and Power, edited by M. Dietler and B. Hayden, pp. 115-143. Smithsonian
Institute Press, Washington, DC.
491
Wilk, Richard R.
1983 Little House in the Jungle: The Causes of Variation in House Size Among
Kekchi Maya. Journal of Anthropological Archaeology 2:99-116.
1989 The Household Economy: Reconsidering The Domestic Mode of
Production. Westview Press, Boulder, CO.
Wilk, Richard R. and Lisa C. Cliggett
2007 Economies and Cultures: Foundations of Economic Anthropology. 2nd ed.
Westview Press, Boulder, CO.
Winterhalder, Bruce, Flora Lu and Bram Tucker
1999 Risk-senstive adaptive tactics: Models and evidence from subsistence
studies in biology and anthropology. Journal of Archaeological Research
7(4):301-348.
Wittfogel, K.
1957 Oriental Despotism. Yale University Press, New Haven, MA.
Wolf, Eric R.
1982 Europe and the People Without History. University of California Press,
Los Angeles.
Wood, Stephanie
1997 Matters of Life at Death: Nahuatl Testaments of Rural Women, 15891801. In Indian Women of Early Mexico, edited by S. Schroeder, S. Wood and R.
Haskett, pp. 165-185. University of Oklahoma City Press, Oklahoma City.
Xiuhtecutli, Nezahualcoyotl
2014 Surrounded?: Assessing the Aztec Blockade on Tlaxcallan from Surface
Evidence from Tepetipac. Paper presented at the 79th Annual Meeting of the
Society for American Archaeology, Austin, TX.
Yang, Martin C.
1945 A Chinese Village: Taitou, Shantung Province. Columbia University
Press, New York.
Yoffee, Norman
2005 Myths of the Archaic State: Evolution of the Earliest Cities, States, and
Civilizations. Cambridge University Press, Cambridge.
Zacarías B., María Patricia
1975 Los enterramientos. In Teotenango: el antiguo lugar de la muralla, edited
by R. Piña Chán, pp. 365-409. vol. 2. Gobierno del Estado de México, Dirección
de Turismo, Mexico City.
492
Zorita, Alonso de
1963 [1566-1585]
Breve y Sumaria Relación de los Señores de la Nueva
España. UNAM, Mexico City.
Zúñiga Bárcenas, Beatriz
1992 Informe de los trabajos de conservación y mantenamiento llevados a cabo
en la zona arqueológica de Calixtlahuaca, México. Report submitted to the
Instituto Nacional de Antropología e Historia.
493
APPENDIX A
CERAMIC TYPE LIST AND CODING SHEET
494
This section provides basic classification coding information for the Calixtlahuaca
Archaeological Project’s general ceramic typology. It includes a master list of type codes
and descriptions, with vessel forms, source regions, and decorative families used for
analysis, and a copy of the project ceramic classification form. A copy of the project
ceramic database and a more detailed project ceramic typebook will be available on
tDAR as part of the general CAP dataset (tDAR ID: 401248).
Table A.1 Source region codes used for regional ceramic provenience assignments
Source
Region Code
(blank)
1
2
3
4
5
6
Source Region
Not Postclassic or Not a Vessel
Local Toluca Valley
Basin of Of Mexico
Morelos
South or Southwest State of Mexico
Mixed Local/Basin of Mexico
Other Non-Local Region
Table A.2 Ceramic type codes used by the Calixtlahuaca archaeological project. For a
listing of primary vessel form code meanings, see Table B.1. For source regions, see
Table A.1. For decorative family descriptions, see Table 8.13. Total Sherds Classified
include all classified excavated material (DS-5 Sample).
495
Primary
Vessel
Form
Code Type Description
Unknown
0
Eroded uncertain
Bowls
10
Eroded uncertain
11
Plain
200 Red rim - fragment
201 Red-rim bowl-plato - Vessel
172 A-3: R rim w/cross
278 Plain with incision
202 B-0: Red, frag / uncert
203 B-1: Plain red
204 B-2: BW/R w/ rows of motifs
205 B-3: BW/R, white zone (#64)
206 B-4: BW/R, cont. band w/ motif #32
207 B-5: Incised red (V. Bravo)
192 Guinda, variant B
208 B-6: Other B Mex looking red
209 B-7: B/R, 3 curving panels w/ scroll
210 B-8: B/R, contin. band w/ vert. lines
211 B-9: Wide-band graphite red
270 B-10: Complex dec w/orange
137 B-11: Red-or incised
212 B-xx: Red vessel, untyped
265 C-0: frag / uncert
213 C-1
214 C-2
272 B / C: fragment, bowl
215 D-0: Red, frag / uncert
216 D-1: Malinalco polychrome
188 D-3 or 4
219 D-5: Cmplx geom thin-line R/white
135 D-6: Horiz bands, sloppy dec
221 E-0: R/buff, frag / uncert
269 E-0: Black/buff, bowl
222 E-1: Wide-band red
223 E-3: Mostly red, concentr designs
224 E-4: Jagged triangles designs
225 E-5: 3 panels w/ frames
496
Total
Source Decorative
Sherds
Region
Family Classified
21
1
14,296
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1
1
1
1
1
1
1
1
1
1
1
4
2
2
4
1
2
1
4
4
1
1
4
1
4
4
4
4
1
1
1
1
1
1
1
10,398
10,674
799
8
80
3
4,646
191
142
1
23
86
83
406
4
4
3
16
70
12
56
545
37
11
86
6
9
6
7
3,300
4
202
77
94
714
A
A
E
BINCS
B
B
B
B
B
BINCS
AZRED
AZRED
B
B
B
B
BINCS
B
C
C
C
C
D
D
D
D
D
E
E
E
E
E
E
Primary
Vessel
Form
Code Type Description
Bowls (Continued)
226 E-6: red stripes around band of R/B
227 E-7: 2 panels, hourglass motif
228 E-8: wavy-line R/B
277 E-9: Wide horizontal bands
132 E-10: Thin par. Lines
133 E-11: Par lines, dif. Direcions
173 E-12: R/bf, int & ext
174 E-13: sun motif
185 E-14: large circles
184 E-16: Red int, R/B ext
229 E-xx: R/buff vessel, untyped
264 B/E: Red fragment
230 G-0: Neg R/B, frag / uncert
231 G-1: 3 patches, busy spiral neg.
232 G-2: Geom neg, red bands (SM Ix)
234 G-xx: Neg R/B vessel, untyped
189 D or H, uncertain
235 H-0: Neg R/white, frag / uncert
236 H-1: 3 patches, busy negative
237 H-xx: Neg R/wh vessel, untyped
25
Painted, eroded
109 Stamped base
122 Other Decorated Types
1
Aztec orange, untyped
24
Aztec I black/orange
108 Aztec II black/orange
21
Aztec III black/orange
114 Aztec III/IV black/orange
271 Aztec III, Local A
274 Aztec III/IV, Local A
181 Aztec III, Local, B
182 Aztec III/IV, Local B
177 Aztec IV
22
Chalco-Cholula Polychrome
12
Mor: C-2
26
Mor: I
17
Mor: Tla poly, fragments
166 Coyotlatelco?
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
497
Total
Source Decorative
Sherds
Region
Family Classified
1
1
1
1
1
1
1
1
1
1
1
1
4
4
4
4
4
4
4
4
1
1
6
2
2
2
2
2
5
5
2
2
2
2
3
3
3
4
E
E
E
E
E
E
E
E
E
E
E
G
G
G
G
H
H
H
H
BoM
OI
AZOR
AZOR
AZOR
AZOR
AZOR
AZOR
AZOR
AZOR
AZOR
AZOR
BoM
OI
OI
OI
283
10
41
507
3,397
841
96
90
39
33
5
2,939
81
4
9
4
6
1
5
1
25
4
97
9
6
1
295
44
445
31
405
65
2
13
1
2
1
8
Primary
Vessel
Form
Code Type Description
Jars
30
Eroded
31
Plain
32
Painted, eroded
38
Other decorated
253 B-0: Red, frag / uncert
254 B-2: BW/R w/ rows of motifs
255 E-0: R/buff, frag / uncert
268 E-0: Black/buff
256 E-2: R/B jar w/ parallel lines
266 B / E: fragment, olla
263 D-0: Red, frag / uncert
258 G-4: Neg, wide red band # 45
257 F-1: Vertical stripes & dots
123 Aztec III black/or jar
33
Tlahuica polychrome
34
Xochimilco polychrome
39
Black-on-white
275 Incised olla
145 Coarse orange
146 Coarse orange, cream slip
Basins
40
Eroded
41
Plain
42
Thin-walled Basin
Comals
45
Eroded
46
Plain
Other Vessel Forms
60
Fragment / uncertain
262 Sahum, Tol Valley form
64
Sahum, ValMex/Morelos form
267 Basin censer
65
Large brazier
101 Deeply incised censer
44
Hanging censer
170 Other censer
498
Total
Source Decorative
Sherds
Region
Family Classified
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
4
5
14
5
5
1
1
1
4
1
1
1
1
1
1
1
4
4
2
3
2
6
1
2
2
6
6
6
1
1
2
41
45
69
7
7
5
5
600
560
21
12
12
13
13
13
13
13
1
1
2
1
1
1
2
1
97
683
117
4,050
89
423
1
313
OI
B
B
E
E
E
D
G
F
AZOR
OI
BoM
OI
BINCS
119,128
103,168
78
84
1,010
18
2,114
16
3,980
1,418
118
5
2
10
1
2
4
6
29
70
Primary
Vessel
Form
Code Type Description
Other Vessel Forms (Continued)
170 Other censer
69
Pitcher, guinda
259 D-2: Pitcher, "Techialoyan"
72
Pitcher, other dec
70
Copa, fragment
112 Copa, flaring
110 Copa, Cholula
113 Guinda vase
273 Pyriform incised vase
131 Tlaloc jar
179 Tlaloc vessel
66
Spinning bowl, local
67
Spinning bowl, Aztec
186 Spin bowl, local Az-B
68
Texcoco fabric marked
71
Miniature vessel
115 Stirrup vessel
168 Asymmetrical bowl
167 Thick-rim vessel
180 Odd composite object
134 Crude unfinished
178 Ring base bowl
100 Ladle
74
Other form
169 Hollow support/handle
Non-Vessel Ceramics
80
Fragment / uncertain
81
Spindle whorl, small
82
Spindle whorl, large
84
Figurine
83
Sherd disk
91
Sherd disk, perforated
92
Worked sherd
86
Water-worn sherd
85
Stamped base
87
Pipe
13
10
10
10
8
8
8
8
14
16
16
18
18
18
20
11
5
20
5
12
11
20
17
20
Total
Source Decorative
Sherds
Region
Family Classified
1
2
4
1
1
1
2
2
4
1
1
1
2
2
2
1
1
1
1
1
1
1
5
1
1
AZRED
D
E
E
AZRED
BOM
AZRED
BINCS
313
32
1
87
14
3
10
82
27
1
8
4
39
21
612
170
7
12
1,377
164
1,037
43
54
15
1,014
164
35
131
355
2,830
3
31
52
4
32
499
Primary
Vessel
Form
Code Type Description
Non-Vessel Ceramics (Continued)
90
Ball
98
Cylinder (solid)
95
Bead
175 lip plug
97
Tube
102 Relief
130 Sculpture
183 Adorno
171 Ceramic cone
99
Mold for figurines
105 Mold, other
96
Lump
138 Cer production-misc
49
Net weight
187 Spike
280 Music-Wind
281 Music-Rattle
Non-Postclassic Ceramics
501 Formative
502 Classic
504 Epiclassic
505 Colonial
116 Colonial glazed earth.
506 Glazed earthenware, unph
507 Other modern/colonial
508 Modern vessels
Total
Sherds
Source Decorative
Region
Family Classified
36
81
8
2
7
13
3
15
7
5
4
28
2
1
20
82
37
21
6
2
37
3
1
2
270
2
4
Figure A.1 Calixtlahuaca Archaeological Project Ceramic Lot Classification Form.
500
CERAMICS CLASSIFICATION
Date:
Initials:
Ent:
code
10
11
202
213
215
221
264
Total lavado:
Locus
Type
N
rm
bd
BOWLS / TRIPODS
Erod. Bowls
Plain
B-0
C-1
D-0
E-0
B / E red fragment
apn molc
code
Type
Lot
N
BASINS
rm
Ltyp
bd
apn
40 Eroded
41 Plain
COMALS
45 Eroded
46 Plain
X: OTHER VESSELS & MISC OBJECTS
82 Large whorl
83 Sherd disk
260 Pie-crust flange
262 Sahum, TV style
OLLAS
30
31
253
254
255
256
Erod Ollas
Plain
B-0
B-2
E-0
E-2
Z: ERODED UNCERTAIN
0
TOTALS:
Total Eroded:
Grand Total:
REMOVED
Type cat. No.
506: Glazed Earth.
Typ cat. No.
DISTINCTIVE ATTRIBUTES (forms, paste, etc
Code Descr
N
Comments:
Peso:
501
Sherd
Size:
APPENDIX B
ATTRIBUTE CODES, CODING SHEETS, AND CONTEXTS SAMPLED
502
This appendix provides coding and sample information for the ceramic attribute analysis.
It includes a list of codes, sample coding sheets, and a list of specific proveniences
sampled. Copies of the database of attribute results will be provided upon request to the
author.
Table B.1 List of ceramic attributes recorded, with a full list of attribute codes.
503
Excavated Ceramic Attribute Codes
Provenience
Anota el numero de bolsa
Huster -19/8/2011
5 Concavo
6 Ovulado
7 Recto-divergente
8 Azteca
9 Globular
11 Asa Hueco
12 Fragmento de Asa
48 Fragmento de Soporte
33 Jarra, forma importada
36 Jarra, forma local
14 Olla cuello alto, curvado
15 Olla cuello baja, curva profunda
17 Olla cuello recto, hacia afuera
49 Olla cuello hacia adentro
58 Olla cuello recto, vertical
59 Olla cuello outflared
60 Olla cuello forma de S (BA)
46 Cuenco delgado
61 Comal, plano
62 Comal, borde elevado
63 Comal, borde grosor variable
18 Mini olla
19 Mini cajete
20 Mini sahumador
21 Mini jarra
22 Mini tripode
23 Mini molcajete
24 Mini plato tripode
25 Mini incensario
26 Mini, otro
27 Sahum Estilo Valle de Toluca
28 Sahum Estilo Valle de Mexico/Morelos
53 Sahum con cortados
47 Incensario forma vaso
29 Incensario Biconico
30 Anafre
31 Incensario Rayado
32 Incensario Colgada
54 Incensario tripode
55 Brasero Grande
38 Vasija de sal
39 Asa estribillo
40 Vasija Asimetrico
55 Copa, biconico
56 Copa, vaso de vino
General
Numero de A - anota el numero de A, con
cuatro digitos numericos (A-0001).
Tipo - Codigo de tipo de cerámica, de la lista
que usan para la clasificación general.
Parte - Parte de vasija (borde, cuerpo,
soporte/asa, molcajete). Si el tepalcate
incluye más que un parte, anótalas todas.
1 Borde
2 Cuerpo
3 Asa/Soporte
4 Molcajete
Forma
Categoria de Vasija (Cat de V)
1 Tripode (o tripode/molcajete incierto)
2 Molcajete
3 Cajete sencillo
4 Cajete, general (si no se sabe si es
cajete o tripode)
5 Olla
6 Cuenca/Cazuela
7 Comal
8 Copa
10 Jarra
11 Vasija Miniatura
12 Sahumidor
13 Incensario
14 Vasija Piriforma
16 Vasija Tlaloc
17 Cucharon
18 Cajete de hilar
20 Otra forma
21 Quien sabe
Forma Especifica de Vasija
1 Plato
2 Conico
3 Hemisferico
4 Recto
504
Forma Especifica de Vasija, cont.
57 Copa, globo
42 Otro
44 Quien sabe
Formas de Bordes (F de B)
1 Curvado/Curved
2 Ahusado/Tapering
3 Cuadrado/Square
4 Triangular/Triangular
5 Bolito interior/Ball on interior
6 Bolito exterior/Ball on exterior
7 Curvado al interior/Curved interior only
8 Curvado al exterior/Curved exterior only
9 Cincel interior/Chisel in-facing
10 Cincel exterior/Chisel out-facing
11 Enlargado interior/Bulge to interior
12 Enlargado exterior/Bulge to exterior
13 S pequena/Small S curve
14 Cuadrado interior, redondo
15 exterior/Square interior, rounded
16 exterior (comals)
17 Indentada/Indented (comals)
18 BA, top > bottom, shallow, pointed
19 BA, bottom longer than top, rounded
20 BA, top/bottom ~equal
21 Other
Angulo de Borde (A de B) - Ollas
1 Continuous from neck angle
2 Mildly outflaring relative to neck,
3 smooth angle
4 Mildly outflaring relative to neck, sharp
5 angle
6 Strongly outflaring relative to neck,
smooth angle
7 Strongly outflaring relative to neck,
sharp angle
8 Other
% de borde que está presente: usa las líneas
rayando en cartón con tamaños de círculos
Ancho del borde (A de B) - mide el ancho
máximo de la borde con los calibradores
Decoracion
Engobe
1 Erosionado
2 Sin engobe
3 Negro
4 Blanco
5 Crema
6 Café clara
7 Café oscura
8 Naranja clara
9 Naranja fuerte
10 Naranja en molcajete
11 Rosa/Rojiza
12 Otro
13 Quien sabe
Motivos: Cuales de los motivos de la lista de
motivos aparecen en el tepalcate? Puede
marcar más que una.
Barro
Pastas: Ve la lista de tipos de pastas de Julie
Diámetro de vasija (D de V) - usa el cartón con
tamaños de círculos. Mide al centimetro mas
cerca
505
Support and Molcajete Codes
Forma de Soporte/Asa (F de S)
Tipo de Soporte/Asa
1 Soporte solido
1 Cilindrico
2 Soporte hueco
2 Ahusado
3 Soporte hueco con sonaja
3 Aplanizado
4 Soporte Annular
4 Almena
5 Asa lazada vertical
5 Almena cortada
6 Asa lazada horizontal
6 Hueco, compuesto (angulos rectas)
7 Asa de ceja, horozontal
7 Forma de patita
8 Asa vertical en borde, pequena
8 Soporte, estilo arana (curvado y
ahusado)
9 Asa vertical en borde, larga
10 Asa vertical en borde y otra asa
9 Bolito
11 Asa de estribillo
10 Soporte, forma de cincel
12 Asa de jarra (larga, vertical)
11 Ahusado, curva hacia afuera
13 Asa de jarra, hueca
12 Bulboso
14 otro soporte
13 Cylindrico, curva hacia afuera
15 Soporte de pedestal
15 Filamentos retorcidos
16 varias diversas asas
17 Termino de Espiral
17 Soporte lazada
18 Termino bifurcado
18 Asa lazada vertical, vas. Desconcida
19 Asa Almena
19 Asa lazada en borde, vas. Desconocida
14 Otro
20 Asa de ceja, horozontal, vas. Desconocida
16 Quien sabe
21 Asa lazada horizontal, vas. Desconcida
Diseno de Soporte/Asa (D de S)
22 incierto
23 asa lazada, incierto
1 Decoracion muy erosionado
24 Asa hueco
2 Sencillo
3 Todo de un color
Altura de Soporte/Asa (Alt): Si es completo,
4 Una linea vertical de color
mide el soporte o asa en su dimensión
5 Circulo
máxima al milimetro
6 Mancha
7 Complejo
8 Fronte de soporte todo pintado
Diseno de Molcajete (D de M) - Usa la lista de
Fronte y atras de soporte con disenos
imagenes de tipos de rayadura
9 differentes
10 Mitad de abajo de soporte todo pintado
11 Rayas horozontales
12 Erosionado
13 No aplica
506
Provenience
Unit Locus Lot
General
Tipo
A-#
Borde
Forma
Parte C de V F de V F de B A de B Diam
%
Decoracion
Ancho Engobe Motivos
Barro INAA
Pasta #
Figure B.1 Ceramic attribute recording forms for rim sherds (p.1) and appendage or
molcajete sherds (p.2)
507
Provenience
Unit Locus Lot
General
Tipo
A-#
Detalles
Forma
Parte C de V F de V A de B T de S F de S D de S Alt
Barro
Decoracion
D de M EngobeMotivos Pasta
508
Table B.2 List of household context lots sampled for ceramic attribute analysis, with total
sherd counts and number of sherds included in attribute sample. Arranged by household
component.
Phase Unit
2
307
2
307
2
307
2
307
2
307
2
307
Total
Locus
2
2
5
5
12
12
Lot
3
4
2
3
3
4
Total
N.
Sherd Attribute
Count Samples
691
64
900
57
1275
54
318
20
274
8
1312
76
4770
279
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Total
315
315
315
315
315
315
315
315
315
315
315
315
315
315
12
13
13
13
14
14
15
17
17
21
22
22
22
28
10
9
12
13
5
8
2
6
7
10
10
11
13
1
487
118
120
178
147
844
23
69
205
760
1566
1165
389
761
6832
27
4
39
9
10
27
1
1
13
32
61
16
15
26
281
2
2
2
2
2
2
2
2
2
2
Total
316
316
316
316
316
316
316
316
316
316
4
4
5
5
5
5
6
7
7
15
4
5
4
5
6
7
5
4
7
4
495
211
216
215
143
22
235
238
549
618
2942
22
20
24
23
12
3
33
21
58
59
275
509
Total
N.
Sherd Attribute
Count Samples
55
5
687
53
197
1
518
68
196
15
524
38
180
15
424
30
662
54
3443
279
Phase
2
2
2
2
2
2
2
2
2
Total
Unit Locus
320
5
320
10
320
10
320
10
320
14
320
16
320
17
320
17
320
22
Lot
5
3
4
5
3
4
4
6
3
2
2
2
2
2
2
Total
323
323
323
323
323
323
2
2
2
2
13
13
8
9
12
13
2
3
267
310
1434
848
2373
901
6133
11
24
79
28
101
37
280
2
2
2
Total
324
324
324
16
16
16
3
4
5
408
454
52
914
40
61
7
108
4
4
4
4
4
Total
303
303
303
303
303
1
1
1
6
6
6
7
8
7
9
692
1255
1399
2219
1097
6662
33
57
70
119
144
423
4
4
4
Total
307
307
307
8
20
20
7
8
9
2094
10182
1610
13886
40
212
24
276
Phase Unit
4
308
4
308
4
308
4
308
4
308
Total
Locus
2
2
2
3
3
Lot
3
4
5
2
3
Total
N.
Sherd Attribute
Count Samples
1444
110
927
73
126
10
824
59
408
32
3729
284
4
4
4
4
44
44
4
4
4
4
4
4
Total
311
311
311
311
311
311
311
311
311
311
311
311
1
1
1
2
5
5
5
11
11
18
20
21
4
6
7
8
5
6
7
5
6
1
1
3
489
745
420
352
444
407
170
127
212
1068
137
417
4988
35
111
48
22
35
59
29
15
38
106
8
41
547
4
4
4
4
4
4
4
4
4
4
Total
316
316
316
316
316
316
316
316
316
316
4
5
6
8
8
15
16
17
17
25
9
3
4
5
8
3
2
2
3
2
639
521
412
844
413
674
705
936
520
912
6576
17
23
1
51
15
25
33
46
20
50
281
4
4
4
Total
322
322
322
1
2
2
2
2
3
316
713
484
1513
59
120
92
271
6
6
6
6
6
Total
307
307
307
307
307
8
9
10
20
20
3
2
2
2
3
94
904
518
3081
1811
6408
5
29
23
125
95
277
510
Lot
2
2
3
2
3
4
2
3
3
Total
N.
Sherd Attribute
Count Samples
149
15
119
17
74
7
696
80
38
6
66
7
1252
83
233
20
467
37
3094
272
Phase
6
6
6
6
6
6
6
6
6
Total
Unit Locus
309
7
309
10
309
11
309
14
309
14
309
14
309
18
309
18
309
25
6
6
6
6
6
Total
316
316
316
316
316
5
6
7
15
15
2
1
2
1
2
403
1421
1055
996
1017
4892
19
88
54
49
64
274
6
6
6
6
6
6
6
6
6
6
Total
317
317
317
317
317
317
317
317
317
317
7
12
13
14
25
25
28
34
35
35
3
8
4
6
6
8
6
4
4
5
556
401
428
533
632
581
818
420
695
638
5702
40
21
25
30
24
1
51
20
35
32
279
6
6
6
6
6
6
Total
324
324
324
324
324
324
1
1
1
4
4
6
5
6
8
4
5
4
468
117
231
869
441
1312
3438
40
10
19
63
30
115
277
APPENDIX C
INSTRUMENTAL NEUTRON ACTIVATION ANAYLIS (INAA) SAMPLE LISTS
AND LAB REPORT
511
This appendix provides additional information on the ceramic samples submitted
to MURR for INAA analysis. It includes catalogs of samples submitted by Huster and
Smith, with excavation context, phase, and ceramic type information. It also includes a
copy of the analysis report provided by MURR. A full copy of the INAA elemental
concentration results will be posted to tDAR and on the MURR Archaeometry website.
They are also available upon request to the author.
Table C.1 Catalog of random household sample ceramic sherds submitted for INAA, with
project chemical source group assignments and macroregional matches. For a full list of
ceramic type descriptions, see Table A.2.
ACH
Context
Samp. # Unit Locus Lot Phase
Type
Code Ceramic Type
Vessel
Form
512
Chem
Group
Macroreg. Chem
Grp
1
303
6
9
4
11
Plain Bowl
Simple
bowl
Simple
bowl
Simple
bowl
Bowl, legs
unknown
Simple
bowl
Simple
bowl
Bowl, legs
unknown
Bowl, legs
unknown
Tripod
bowl
Molcajete
CAL_F
2
303
6
9
4
11
Plain Bowl
3
303
6
9
4
11
Plain Bowl
4
303
6
9
4
11
Plain Bowl
5
303
6
9
4
10
Eroded Bowl
6
303
6
9
4
10
Eroded Bowl
7
303
6
9
4
221
8
303
6
9
4
185
9
303
6
9
4
277
10
303
6
9
4
277
11
303
6
9
4
277
12
303
6
9
4
185
13
303
6
9
4
207
14
303
6
9
4
11
Matlatzinca Red-on-Buff (variant
unknown)
Matlatzinca Red-on-Buff (red
circles variant)
Matlatzinca Red-on-Buff (wide
band variant)
Matlatzinca Red-on-Buff (wide
band variant)
Matlatzinca Red-on-Buff (wide
band variant)
Matlatzinca Red-on-Buff (red
circles variant)
Matlatzinca Incised Redware
(interior)
Plain Bowl
CAL_F
30
Matlatzinca Red or Red-on-Buff
fragment
Eroded Jar
Tripod
bowl
Tripod
bowl
Tripod
bowl
Bowl, legs
unknown
Tripod
bowl
Jar
15
303
6
9
4
264
16
303
6
9
4
17
303
6
9
4
30
Eroded Jar
Jar
CAL_F
18
303
6
9
4
30
Eroded Jar
Jar
CAL_F
19
303
6
9
4
30
Eroded Jar
Jar
CAL_F
20
303
6
9
4
30
Eroded Jar
Jar
CAL_F
21
303
6
9
4
30
Eroded Jar
Jar
CAL_F
22
303
6
9
4
30
Eroded Jar
Jar
CAL_F
23
303
6
9
4
30
Eroded Jar
Jar
CAL_F
513
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
CAL_F
ACH
Context
Samp. # Unit Locus Lot Phase
24
303
6
9
4
Type
Code Ceramic Type
31
Plain Jar
Vessel
Form
Jar
Chem
Group
CAL_F
25
303
6
9
4
31
Plain Jar
Jar
CAL_F
26
303
6
9
4
72
Matlatzinca Pitcher
Pitcher
CAL_F
27
303
6
9
4
170
Other Censer
28
303
6
9
4
134
29
303
6
9
4
134
30
303
6
9
4
134
31
303
6
7
4
202
32
303
6
7
4
132
33
303
6
7
4
10
34
303
6
7
4
10
35
303
1
8
4
10
36
303
1
8
4
202
37
303
1
8
4
166
38
303
1
8
4
277
39
303
1
6
4
202
40
303
1
6
4
10
41
303
1
6
4
11
42
303
1
7
4
264
43
303
1
7
4
133
44
303
1
7
4
11
45
303
1
7
4
11
46
303
6
7
4
30
Frying Pan
Censer
Crude Unfinished
Bowl, legs
unknown
Crude Unfinished
Bowl, legs
unknown
Crude Unfinished
Simple
bowl
Matlatzinca Red (variant
Bowl, legs
unknown)
unknown
Matlatzinca Red-on-Buff (parallel Bowl, legs
line variant)
unknown
Eroded Bowl
Simple
bowl
Eroded Bowl
Simple
bowl
Eroded Bowl
Bowl, legs
unknown
Matlatzinca Red (variant
Tripod
unknown)
bowl
Coyotlatelco
Bowl, legs
unknown
Matlatzinca Red-on-Buff (wide
Tripod
band variant)
bowl
Matlatzinca Red (variant
Tripod
unknown)
bowl
Eroded Bowl
Tripod
bowl
Plain Bowl
Bowl, legs
unknown
Matlatzinca Red or Red-on-Buff
Tripod
fragment
bowl
Matlatzinca Red-on-Buff (multiple Bowl, legs
sets of parallel lines)
unknown
Plain Bowl
Bowl, legs
unknown
Plain Bowl
Simple
bowl
Eroded Jar
Jar
514
CAL_F
CAL_F
CAL_F
CAL_F
CAL_A
CAL_A
CAL_F
CAL_A
CAL_A
CAL_A
CAL_E
CAL_F
CAL_A
CAL_U
CAL_A
CAL_A
CAL_A
CAL_A
CAL_D
CAL_A
Macroreg. Chem
Grp
ACH
Context
Samp. # Unit Locus Lot Phase
47
303
6 7
4
Type
Code Ceramic Type
30
Eroded Jar
48
303
6
7
4
30
Eroded Jar
Jar
CAL_A
49
303
6
7
4
30
Eroded Jar
Jar
CAL_U
50
303
6
7
4
30
Eroded Jar
Jar
CAL_F
51
303
1
8
4
30
Eroded Jar
Jar
CAL_C
52
303
1
6
4
31
Plain Jar
Jar
CAL_U
53
303
1
6
4
30
Eroded Jar
Jar
CAL_A
54
303
1
7
4
167
Wide rim jar
Jar
CAL_A
55
303
1
7
4
31
Plain Jar
Jar
CAL_F
56
303
6
7
4
267
Biconical Spiked Censer
Censer
CAL_U
57
303
6
7
4
168
Asymetrical bowl
CAL_A
58
303
1
6
4
42
Thin walled basin
Asymetrical
Bowl
Basin
59
303
1
7
4
134
Crude Unfinished
CAL_A
60
303
1
7
4
267
Biconical Spiked Censer
Minature
Vessel
Censer
61
307
2
4
2
11
Plain Bowl
CAL_A
62
307
2
4
2
221
63
307
2
3
2
10
Matlatzinca Red-on-Buff (variant
unknown)
Eroded Bowl
64
307
2
3
2
277
65
307
2
3
2
277
66
307
5
2
2
202
67
307
5
2
2
277
68
307
12
4
2
11
Matlatzinca Red-on-Buff (wide
band variant)
Matlatzinca Red-on-Buff (wide
band variant)
Matlatzinca Red (variant
unknown)
Matlatzinca Red-on-Buff (wide
band variant)
Plain Bowl
69
307
12
4
2
11
Plain Bowl
Simple
bowl Bowl,
legs
unknown
Simple
bowl
Tripod
bowl
Tripod
bowl
Tripod
bowl Bowl,
legs
unknown
Bowl, legs
unknown
Bowl, legs
unknown
Vessel
Form
Jar
515
Chem
Macroreg. Chem
Group Grp
CAL_A
CAL_B
CAL_A
CAL_A
CAL_A
CAL_A
CAL_U
CAL_U
CAL_A
CAL_U
CAL_U
ACH
Context
Samp. # Unit Locus Lot Phase
70
307
12 4
2
Type
Code Ceramic Type
10
Eroded Bowl
Chem
Macroreg. Chem
Group Grp
CAL_A
71
307
12
4
2
200
CAL_A
72
307
12
4
2
1
73
307
5
3
2
11
74
307
5
3
2
221
75
307
3
2
264
76
307
2
3
2
30
Vessel
Form
Simple
bowl
Tripod
Red rim bowl
bowl Bowl,
Aztec I, Black on Orange
legs
unknown
Plain Bowl
Simple
bowl Bowl,
Matlatzinca Red-on-Buff (variant legs
unknown)
unknown
Matlatzinca Red or Red-on-Buff
Bowl, legs
fragment
unknown
Eroded Jar
Jar
77
307
2
3
2
30
Eroded Jar
Jar
CAL_A
78
307
2
3
2
255
Jar
CAL_A
79
307
5
2
2
30
Matlatzinca Red-on-Buff Jar
(variant unknown)
Eroded Jar
Jar
CAL_A
80
307
5
2
2
31
Plain Jar
Jar
CAL_A
81
307
5
2
2
31
Plain Jar
Jar
CAL_A
82
307
12
4
2
30
Eroded Jar
Jar
CAL_A
83
307
12
4
2
30
Eroded Jar
Jar
CAL_U
84
307
12
4
2
30
Eroded Jar
Jar
CAL_U
85
307
5
3
2
30
Eroded Jar
Jar
CAL_A
86
307
2
3
2
134
Crude Unfinished
CAL_A
87
307
5
2
2
42
Thin walled basin
Simple
bowl
Basin
88
307
5
2
2
134
Crude Unfinished
CAL_U
89
307
12 4
2
134
Crude Unfinished
90
307
4
2
267
Biconical Spiked Censer
Simple
bowl
Simple
bowl
Censer
91
307
8
7
4
11
Plain Bowl
92
307
8
7
4
202
Matlatzinca Red (variant
unknown)
12
12
516
CAL_D
CAL_U
CAL_A
CAL_A
CAL_A
CAL_U
CAL_U
CAL_A
Bowl, legs CAL_F
unknown
Bowl, legs CAL_A
unknown
ACH
Context
Samp. # Unit Locus Lot Phase
93
307
20 9
4
Type
Code Ceramic Type
11
Plain Bowl
Chem
Macroreg. Chem
Group Grp
CAL_A
94
307
20
9
4
11
CAL_A
95
307
20
8
4
11
96
307
20
8
4
11
97
307
20
8
4
11
98
307
20
8
4
11
99
307
20
8
4
132
100
307
20
8
4
200
101
307
20
8
4
200
102
307
20
8
4
204
103
307
20
8
4
222
104
307
20
8
4
11
105
307
20
8
4
10
106
307
20
9
4
167
Vessel
Form
Bowl, legs
unknown
Bowl, legs
Plain Bowl
unknown
Plain Bowl
Simple
bowl
Plain Bowl
Simple
bowl
Plain Bowl
Bowl, legs
unknown
Plain Bowl
Bowl, legs
unknown
Matlatzinca Red-on-Buff (parallel Basin
line variant)
Tripod
Red rim bowl
bowl
Red rim bowl
Tripod
bowl
Matlatzinca Red (black and white Tripod
on red)
bowl
Matlatzinca Red-on-Buff (simple Tripod
red band variant)
bowl
Plain Bowl
Bowl, legs
unknown
Eroded Bowl
Bowl, legs
unknown
Wide rim jar
Jar
107
307
20
8
4
31
Plain Jar
Jar
CAL_A
108
307
20
8
4
31
Plain Jar
Jar
CAL_U
109
307
20
8
4
31
Plain Jar
Jar
CAL_A
110
307
20
8
4
167
Wide rim jar
Jar
CAL_A
111
307
20
8
4
255
Jar
CAL_A
112
307
20
8
4
256
Jar
CAL_A
113
307
20
8
4
256
Jar
CAL_U
114
307
20
8
4
167
Matlatzinca Red-on-Buff Jar
(variant unknown)
Matlatzinca Red-on-Buff Jar (sets
of parallel lines variant)
Matlatzinca Red-on-Buff Jar (sets
of parallel lines variant)
Wide rim jar
Jar
CAL_A
115
307
20
8
4
167
Wide rim jar
Jar
CAL_A
517
CAL_A
CAL_H
CAL_U
CAL_U
CAL_A
CAL_H
CAL_A
CAL_A
CAL_U
CAL_U
CAL_A
CAL_A
ACH
Context
Samp. # Unit Locus Lot Phase
116
307
20 8
4
Type
Code Ceramic Type
262 Toluca Valley Sahumador
Chem
Macroreg. Chem
Vessel
Form
Group Grp
Frying Pan CAL_A
Censer
Censer
CAL_U
117
307
20
8
4
267
Biconical Spiked Censer
118
307
20
8
4
267
Biconical Spiked Censer
Censer
CAL_A
119
307
20
8
4
46
Plain Comal
Comal
CAL_B
120
307
20 8
4
170
Other Censer
121
307
9
2
6
10
122
307
9
2
6
132
123
307
9
2
6
264
124
307
10
2
6
10
125
307
20
2
6
10
126
307
20
2
6
10
127
307
20
2
6
10
128
307
20
2
6
10
129
307
20
2
6
202
130
307
20
2
6
264
131
307
20
2
6
226
132
307
20
3
6
10
133
307
20
3
6
202
134
307
20
3
6
226
135
307
20
3
6
181
136
307
20
2
6
137
307
20
2
138
307
20
2
CAL_U
30
Frying Pan
Censer
Bowl, legs
Eroded Bowl
unknown
Matlatzinca Red-on-Buff (parallel Tripod
line variant)
bowl
Matlatzinca Red or Red-on-Buff
Tripod
fragment
bowl
Eroded Bowl
Bowl, legs
unknown
Eroded Bowl
Simple
bowl
Eroded Bowl
Simple
bowl
Eroded Bowl
Simple
bowl
Eroded Bowl
Bowl, legs
unknown
Matlatzinca Red (variant
Tripod
unknown)
bowl
Matlatzinca Red or Red-on-Buff
Bowl, legs
fragment
unknown
Matlatzinca Red-on-Buff (simple Bowl, legs
interior and exterior)
unknown
Eroded Bowl
Bowl, legs
unknown
Matlatzinca Red (variant
Tripod
unknown)
bowl
Matlatzinca Red-on-Buff (simple Tripod
interior and exterior)
bowl
Aztec III Black-on-Orange, local
Tripod
variant B
bowl
Eroded Jar
Jar
6
30
Eroded Jar
Jar
CAL_A
6
30
Eroded Jar
Jar
CAL_U
518
CAL_U
CAL_C
CAL_A
CAL_A
CAL_A
CAL_A
CAL_H
CAL_D
CAL_A
CAL_U
CAL_C
CAL_U
CAL_E
CAL_A
CAL_A
CAL_G
East
Morelos
ACH
Context
Samp. # Unit Locus Lot Phase
139
307
20 2
6
Type
Code Ceramic Type
167 Wide rim jar
Vessel
Form
Jar
140
307
20
2
6
167
Wide rim jar
Jar
CAL_A
141
307
20
2
6
167
Wide rim jar
Jar
CAL_U
142
307
20
3
6
30
Eroded Jar
Jar
CAL_A
143
307
20
3
6
30
Eroded Jar
Jar
CAL_A
144
307
20
3
6
30
Eroded Jar
Jar
CAL_A
145
307
20
3
6
31
Plain Jar
Jar
CAL_A
146
307
2
6
45
Eroded Comal
Comal
CAL_U
147
307
20
2
6
267
Biconical Spiked Censer
Censer
CAL_A
148
307
20
2
6
168
Asymetrical bowl
CAL_B
149
307
20
3
6
42
Thin walled basin
Asymetrical
Bowl
Basin
150
307
20
3
6
134
Crude Unfinished
Jar
CAL_U
151
308
2
3
4
10
Eroded Bowl
CAL_A
152
308
2
3
4
10
Eroded Bowl
153
308
2
3
4
10
Eroded Bowl
154
308
2
3
4
10
Eroded Bowl
155
308
2
3
4
10
Eroded Bowl
156
308
2
3
4
11
Plain Bowl
157
308
2
3
4
264
158
308
2
3
4
200
Matlatzinca Red or Red-on-Buff
fragment Red
rim bowl
159
308
2
3
4
200
Red rim bowl
160
308
2
4
4
11
Plain Bowl
161
308
2
4
4
11
Plain Bowl
Simple
bowl Bowl,
legs
unknown
Tripod
bowl Bowl,
legs
unknown
Bowl, legs
unknown
Simple
bowl
Tripod
bowl
Tripod
bowl Bowl,
legs
unknown
Simple
bowl
Simple
bowl
9
519
Chem
Macroreg. Chem
Group Grp
CAL_B
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_B
CAL_U
ACH
Context
Samp. # Unit Locus Lot Phase
162
308
2 4
4
Type
Code Ceramic Type
11
Plain Bowl
Chem
Macroreg. Chem
Group Grp
CAL_H
163
308
3
2
4
11
CAL_A
164
308
3
2
4
213
165
308
3
2
4
215
166
308
2
3
4
31
Vessel
Form
Simple
bowl
Simple
Plain Bowl
bowl
Matlatzinca Simple polychrome on Tripod
white
bowl
Matlatzinca red on white
Tripod
bowl
Plain Jar
Jar
167
308
2
3
4
31
Plain Jar
Jar
CAL_U
168
308
2
4
4
31
Plain Jar
Jar
CAL_A
169
308
2
4
4
167
Wide rim jar
Jar
CAL_A
170
308
2
5
4
167
Wide rim jar
Jar
CAL_A
171
308
3
2
4
31
Plain Jar
Jar
CAL_A
172
308
3
2
4
31
Plain Jar
Jar
CAL_A
173
308
3
2
4
31
Plain Jar
Jar
CAL_A
174
308
3
2
4
31
Plain Jar
Jar
CAL_H
175
308
3
2
4
167
Wide rim jar
Jar
CAL_U
176
308
2
3
4
45
Eroded Comal
Comal
CAL_U
177
308
2
4
4
67
308
2
4
4
267
Spinning
Bowl
Censer
CAL_D
178
Aztec III, Black-on-Orange
Spinning Bowl
Biconical Spiked Censer
179
308
3
2
4
132
Basin
CAL_A
180
308
3
2
4
100
Matlatzinca Red-on-Buff (parallel
line variant)
Spoon
Spoon
CAL_A
181
309
7
2
6
271
309
14 2
6
11
183
309
14
2
6
11
Plain Bowl
184
309
14
2
6
11
Plain Bowl
Tripod
bowl
Simple
bowl
Bowl, legs
unknown
Bowl, legs
unknown
CAL_A
182
Aztec III/IV, Black-on-Orange,
Local variant A
Plain Bowl
520
CAL_A
CAL_U
CAL_B
CAL_U
CAL_H
CAL_H
CAL_A
ACH
Context
Samp. # Unit Locus Lot Phase
185
309
14 2
6
186
309
14
2
6
Type
Code Ceramic Type
202 Matlatzinca Red (variant
unknown)
208 Guinda Black on Red
187
309
14
2
6
132
188
309
14
2
6
181
189
309
18
2
6
10
190
309
18
2
6
10
191
309
18
2
6
202
192
309
18
3
6
181
193
309
18
3
6
132
194
309
25
3
6
264
195
309
10
2
6
30
Vessel
Form
Tripod
bowl
Simple
bowl
Matlatzinca Red-on-Buff (parallel Simple
line variant)
bowl
Aztec III Black-on-Orange, local
Tripod
variant B
bowl
Eroded Bowl
Tripod
bowl
Eroded Bowl
Simple
bowl
Matlatzinca Red (variant
Tripod
unknown)
bowl
Aztec III Black-on-Orange, local
Tripod
variant B
bowl
Matlatzinca Red-on-Buff (parallel Tripod
line variant)
bowl
Matlatzinca Red or Red-on-Buff
Tripod
fragment
bowl
Eroded Jar
Jar
196
309
10
2
6
30
Eroded Jar
Jar
CAL_A
197
309
14
2
6
167
Wide rim jar
Jar
CAL_A
198
309
18
2
6
30
Eroded Jar
Jar
CAL_A
199
309
18
2
6
30
Eroded Jar
Jar
CAL_A
200
309
18
2
6
256
Jar
CAL_A
201
309
18
2
6
167
Matlatzinca Red-on-Buff Jar (sets
of parallel lines variant)
Wide rim jar
Jar
CAL_A
202
309
25
3
6
30
Eroded Jar
Jar
CAL_A
203
309
25
3
6
167
Wide rim jar
Jar
CAL_A
204
309
25
3
6
30
Eroded Jar
Jar
CAL_A
205
309
10
2
6
132
Basin
CAL_A
206
309
14
2
6
67
309
14
2
6
267
Spinning
Bowl
Censer
CAL_U
207
Matlatzinca Red-on-Buff (parallel
line variant)
Aztec III, Black-on-Orange
Spinning Bowl Biconical
Spiked Censer
521
Chem
Group
CAL_A
CAL_Gu
inda
CAL_A
CAL_B
CAL_A
CAL_A
CAL_A
CAL_U
CAL_A
CAL_U
CAL_U
CAL_A
Macroreg. Chem
Grp
ACH
Context
Samp. # Unit Locus Lot Phase
208
309
18 2
6
Type
Code Ceramic Type
46
Plain Comal
209
309
210
18
Vessel
Form
Comal
2
6
267
Biconical Spiked Censer
309
14 2
6
21
Aztec III, Black-on-Orange
211
311
1
6
4
10
212
311
1
6
4
11
213
311
1
6
4
11
214
311
1
6
4
11
215
311
1
6
4
11
216
311
1
6
4
11
217
311
1
6
4
132
218
311
1
6
4
184
219
311
1
7
4
11
220
311
1
7
4
10
221
311
1
7
4
277
222
311
5
7
4
11
223
311
11
6
4
11
224
311
11
6
4
213
225
311
21
3
4
200
226
311
1
6
4
31
227
311
1
6
4
255
228
311
1
7
4
229
311
1
7
230
311
1
7
Censer
Chem
Macroreg. Chem
Group Grp
CAL_A
CAL_A
Tripod
bowl
Simple
Eroded Bowl
bowl
Plain Bowl
Simple
bowl
Plain Bowl
Simple
bowl
Plain Bowl
Simple
bowl
Plain Bowl
Simple
bowl
Plain Bowl
Simple
bowl
Matlatzinca Red-on-Buff (parallel Tripod
line variant)
bowl
Matlatzinca Red-on-Buff (red
Tripod
interior, red-on-buff exterior)
bowl
Plain Bowl
Bowl, legs
unknown
Eroded Bowl
Bowl, legs
unknown
Matlatzinca Red-on-Buff (wide
Tripod
band variant)
bowl
Plain Bowl
Bowl, legs
unknown
Plain Bowl
Simple
bowl
Matlatzinca Simple polychrome on Molcajete
white
Red rim bowl
Simple
bowl
Plain Jar
Jar
CAL_D
Jar
CAL_A
31
Matlatzinca Red-on-Buff Jar
(variant unknown)
Plain Jar
Jar
CAL_A
4
31
Plain Jar
Jar
CAL_A
4
31
Plain Jar
Jar
CAL_A
522
CAL_C
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_U
CAL_H
CAL_A
CAL_A
CAL_A
ACH
Context
Samp. # Unit Locus Lot Phase
231
311
1
7
4
Type
Code Ceramic Type
31
Plain Jar
232
311
233
311
234
5
Vessel
Form
Jar
Chem
Macroreg. Chem
Group Grp
CAL_A
7
4
31
Plain Jar
Jar
CAL_A
21
3
4
30
Eroded Jar
Jar
CAL_U
311
21
3
4
30
Eroded Jar
Jar
CAL_A
235
311
21
3
4
31
Plain Jar
Jar
CAL_A
236
311
1
6
4
134
Crude Unfinished
CAL_A
237
311
1
6
4
134
Crude Unfinished
238
311
5
7
4
134
Crude Unfinished
239
311
11
6
4
132
240
311
21
3
4
46
Matlatzinca Red-on-Buff (parallel
line variant)
Plain Comal
Simple
bowl
Simple
bowl
Simple
bowl
Basin
Comal
CAL_U
241
311
1
4
4
11
Plain Bowl
242
311
1
4
4
202
243
311
2
8
4
221
244
311
2
8
4
181
245
311
5
5 44
11
246
311
5
5 44
200
247
311
5
6 44
204
248
311
5
6 44
200
249
311
18
1
4
10
250
311
18
1
4
10
251
311
18
1
4
202
252
311
18
1
4
132
253
311
18
1
4
132
Simple
bowl
Simple
Matlatzinca Red (variant
unknown)
bowl
Matlatzinca Red-on-Buff (variant Bowl, legs
unknown)
unknown
Aztec III Black-on-Orange, local
Tripod
variant B
bowl
Plain Bowl
Bowl, legs
unknown
Red rim bowl
Simple
bowl
Matlatzinca Red (black and white Tripod
on red)
bowl
Red rim bowl
Simple
bowl
Eroded Bowl
Simple
bowl
Eroded Bowl
Tripod
bowl
Matlatzinca Red (variant
Bowl, legs
unknown)
unknown
Matlatzinca Red-on-Buff (parallel Bowl, legs
line variant)
unknown
Matlatzinca Red-on-Buff (parallel Tripod
line variant)
bowl
523
CAL_U
CAL_A
CAL_U
CAL_A
CAL_A
CAL_U
CAL_A
CAL_A
CAL_A
CAL_A
CAL_U
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
ACH
Context
Samp. # Unit Locus Lot Phase
254
311
18
1
4
1
4
1
4
4
Type
Code Ceramic Type
264 Matlatzinca Red or Red-on-Buff
fragment
204 Matlatzinca Red (black and white
on red)
30
Eroded Jar
311
1
4
4
253
258
311
5
5
44 255
259
311
5
6
260
311
5
6
44 256
261
311
18
1
4
262
311
18
1
263
311
18
264
311
265
311
266
311
5
6
44
132
267
311
5
6
44
268
311
11
5
269
311
18
270
311
271
255
311
256
311
257
18
Vessel
Form
Simple
bowl
Tripod
bowl
Jar
Chem
Macroreg. Chem
Group Grp
CAL_A
CAL_A
CAL_A
Matlatzinca Red Jar (variant
unknown)
Matlatzinca Red-on-Buff Jar
(variant unknown)
Plain Jar
Jar
CAL_A
Jar
CAL_A
Jar
CAL_A
Jar
CAL_A
30
Matlatzinca Red-on-Buff Jar (sets
of parallel lines variant)
Eroded Jar
Jar
CAL_A
4
30
Eroded Jar
Jar
CAL_A
1
4
30
Eroded Jar
Jar
CAL_A
18
1
4
266
Jar
CAL_A
20
1
4
31
Matlatzinca Red or Red-on-Buff
Jar fragment
Plain Jar
Jar
CAL_A
Basin
CAL_A
46
Matlatzinca Red-on-Buff (parallel
line variant)
Plain Comal
Comal
CAL_B
4
101
Scored Censer
Censer
CAL_F
1
4
70
Copa fragment
Goblet
CAL_A
18
1
4
134
Crude Unfinished
CAL_H
315
12
10
2
222
272
315
12
2
170
Matlatzinca Red-on-Buff
(interlocking triangle variant)
Other Censer
273
315
13
2
11
Plain Bowl
274
315
13
12
2
10
Eroded Bowl
275
315
13
12
2
11
Plain Bowl
276
315
22
11
2
10
Eroded Bowl
Minature
Vessel
Bowl, legs
unknown
Tripod
bowl
Simple
bowl
Bowl, legs
unknown
Tripod
bowl
Simple
bowl
10
9
44
31
524
CAL_H
CAL_H
CAL_H
CAL_A
CAL_U
CAL_A
ACH
Context
Samp. # Unit Locus Lot Phase
277
315
13
13
2
278
315
14
8
2
279
315
14
8
2
280
315
14
8
2
281
315
17
7
2
282
315
22
10
2
283
315
22
10
2
284
315
28
1
2
285
315
28
1
2
286
315
13
12
2
287
315
13
13
2
Type
Vessel
Code Ceramic Type
Form
132 Matlatzinca Red-on-Buff (parallel Bowl, legs
line variant)
unknown
Plain Bowl
Simple
11
bowl
132 Matlatzinca Red-on-Buff (parallel Bowl, legs
line variant)
unknown
225 Matlatzinca Red-on-Buff (interior Tripod
panels variant)
bowl
11
Plain Bowl
Simple
bowl
10
Eroded Bowl
Bowl, legs
unknown
226 Matlatzinca Red-on-Buff (simple Simple
interior and exterior)
bowl
11
Plain Bowl
Simple
bowl
11
Plain Bowl
Simple
bowl
256 Matlatzinca Red-on-Buff Jar (sets Jar
of parallel lines variant)
31
Jar
Plain Jar
288
315
14
2
31
Plain Jar
Jar
CAL_A
289
315
21
10
2
31
Plain Jar
Jar
CAL_H
290
315
22
10
2
31
Plain Jar
Jar
CAL_A
291
315
22
10
2
31
Plain Jar
Jar
CAL_A
292
315
22
10
2
31
Plain Jar
Jar
CAL_A
293
315
22
10
2
31
Plain Jar
Jar
CAL_A
294
315
28
1
2
30
Eroded Jar
Jar
CAL_A
295
315
28
1
2
31
Plain Jar
Jar
CAL_H
296
315
12
10
2
267
Biconical Spiked Censer
Censer
CAL_A
297
315
22
11
2
267
Biconical Spiked Censer
Censer
CAL_A
298
315
14
8
2
262
Toluca Valley Sahumador
CAL_H
299
315
17
7
2
101
Scored Censer
Frying Pan
Censer
Censer
8
525
Chem
Macroreg. Chem
Group Grp
CAL_A
CAL_A
CAL_H
CAL_H
CAL_A
CAL_A
CAL_A
CAL_H
CAL_H
CAL_A
CAL_A
CAL_H
ACH
Context
Samp. # Unit Locus Lot Phase
300
315
22 10
2
Type
Code Ceramic Type
134 Crude Unfinished
301
316
5
5
2
133
302
316
5
5
2
132
303
316
5
6
2
11
304
316
6
5
2
11
305
316
7
4
2
200
306
316
7
4
2
203
307
316
7
4
2
215
308
316
7
7
2
11
309
316
7
7
2
11
310
316
7
7
2
200
311
316
7
7
2
11
312
316
7
7
2
225
313
316
15
4
2
11
314
316
15
4
2
11
315
316
15
4
2
277
316
316
4
4
2
31
Chem
Macroreg. Chem
Vessel
Form
Group Grp
Simple
CAL_A
bowl
Matlatzinca Red-on-Buff (multiple Simple
CAL_A
sets of parallel lines) Matlatzinca bowl
CAL_A
Tripod
Red-on-Buff (parallel
line variant)
bowl
CAL_A
Plain Bowl
Simple
bowl
Plain Bowl
Bowl, legs CAL_A
unknown
Red rim bowl
CAL_A
Simple
bowl
Matlatzinca Red (simple red)
CAL_A
Tripod
bowl
Matlatzinca red on white
CAL_A
Tripod
bowl
Plain Bowl
CAL_U
Simple
bowl
Plain Bowl
CAL_A
Simple
bowl
Red rim bowl
CAL_H
Tripod
bowl
Plain Bowl
Bowl, legs CAL_A
unknown
Matlatzinca Red-on-Buff (interior Jar
CAL_A
panels variant)
Bowl, legs CAL_H
Plain Bowl
unknown
Plain Bowl
CAL_U
Simple
bowl
Matlatzinca Red-on-Buff (wide
CAL_A
Tripod
band variant)
bowl
CAL_A
Plain Jar
Jar
317
316
4
4
2
31
Plain Jar
Jar
CAL_A
318
316
4
5
2
31
Plain Jar
Jar
CAL_A
319
316
5
4
2
31
Plain Jar
Jar
CAL_A
320
316
5
4
2
253
Jar
CAL_A
321
316
7
7
2
31
Matlatzinca Red Jar (variant
unknown)
Plain Jar
Jar
CAL_A
322
316
4
2
30
Eroded Jar
Jar
CAL_A
15
526
ACH
Context
Samp. # Unit Locus Lot Phase
323
316
15 4
2
Type
Code Ceramic Type
31
Plain Jar
324
316
15
4
2
31
Plain Jar
Jar
CAL_U
325
316
15
4
2
31
Plain Jar
Jar
CAL_A
326
316
4
4
2
133
327
316
5
4
2
267
Matlatzinca Red-on-Buff (multiple Basin
sets of parallel lines)
Biconical Spiked Censer
Censer
328
316
6
5
2
71
Miniature Vessel
329
316
15 4
2
262
Toluca Valley Sahumador
330
316
7
7
2
101
Scored Censer
331
316
5
3
4
213
332
316
5
3
4
200
333
316
8
5
4
202
334
316
8
5
4
200
335
316
16
2
4
11
336
316
16
2
4
11
337
316
16
2
4
10
338
316
17
2
4
11
339
316
17
2
4
264
340
316
17
3
4
202
341
316
25
2
4
10
342
316
25
2
4
10
343
316
25
2
4
10
344
316
25
2
4
10
345
316
25
2
4
202
Matlatzinca Simple polychrome on Tripod
white
bowl
Red rim bowl
Simple
bowl
Matlatzinca Red (variant
Tripod
unknown)
bowl
Red rim bowl
Simple
bowl
Plain Bowl
Simple
bowl
Plain Bowl
Simple
bowl
Eroded Bowl
Tripod
bowl
Plain Bowl
Simple
bowl
Matlatzinca Red or Red-on-Buff
Simple
fragment
bowl
Matlatzinca Red (variant
Tripod
unknown)
bowl
Eroded Bowl
Simple
bowl
Eroded Bowl
Simple
bowl
Eroded Bowl
Bowl, legs
unknown
Eroded Bowl
Simple
bowl
Matlatzinca Red (variant
Tripod
unknown)
bowl
Vessel
Form
Jar
527
Chem
Macroreg. Chem
Group Grp
CAL_A
CAL_A
CAL_A
Minature CAL_A
Vessel
Frying Pan CAL_A
Censer
CAL_A
Censer
CAL_A
CAL_U
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_U
CAL_U
CAL_A
ACH
Context
Samp. # Unit Locus Lot Phase
346
316
5 3
4
Type
Code Ceramic Type
31
Plain Jar
Vessel
Form
Jar
Chem
Group
CAL_A
347
316
5
3
4
256
Jar
CAL_A
31
Matlatzinca Red-on-Buff Jar (sets
of parallel lines variant)
Plain Jar
348
316
8
5
4
Jar
CAL_A
349
316
8
5
4
31
Plain Jar
Jar
CAL_A
350
316
8
5
4
167
Wide rim jar
Jar
CAL_A
351
316
15
3
4
167
Wide rim jar
Jar
CAL_A
352
316
15
3
4
255
Jar
CAL_A
353
316
17
2
4
30
Matlatzinca Red-on-Buff Jar
(variant unknown)
Eroded Jar
Jar
CAL_A
354
316
17
2
4
31
Plain Jar
Jar
CAL_A
355
316
17
2
4
31
Plain Jar
Jar
CAL_U
356
316
8
5
4
132
CAL_A
357
316
8
8
4
180
358
316
16
2
4
134
Matlatzinca Red-on-Buff (parallel Basin
line variant)
Brazier
Spinning
Bowl
Crude Unfinished
Spoon
359
316
17
2
4
64
Texcoco Molded Sahumador
CAL_U
360
316
17
2
4
267
Biconical Spiked Censer
Frying Pan
Censer
Censer
361
316
6
1
6
11
Plain Bowl
CAL_A
362
316
6
1
6
11
Plain Bowl
363
316
6
1
6
114
Aztec III/IV, Black-on-Orange
364
316
7
2
6
208
Guinda Black on Red
365
316
7
2
6
222
366
316
15
1
6
271
316
15
1
6
11
368
316
15
2
6
10
Eroded Bowl
Tripod
bowl
Simple
bowl
Tripod
bowl
CAL_A
367
Matlatzinca Red-on-Buff (simple
red band variant)
Aztec III/IV, Black-on-Orange,
Local variant A
Plain Bowl
Simple
bowl
Bowl, legs
unknown
Tripod
bowl
Simple
bowl
Molcajete
528
CAL_A
CAL_U
CAL_A
CAL_A
CAL_B
CAL_Gu
inda
CAL_U
CAL_A
CAL_A
Macroreg. Chem
Grp
ACH
Context
Samp. # Unit Locus Lot Phase
369
316
15 2
6
Type
Code Ceramic Type
11
Plain Bowl
Chem
Macroreg. Chem
Group Grp
CAL_A
370
316
15
2
6
202
CAL_H
371
316
15
2
6
202
373
316
15
2
6
213
374
316
15
2
6
200
375
316
6
1
6
264
376
316
6
1
6
31
Vessel
Form
Simple
bowl
Bowl, legs
Matlatzinca Red (variant
unknown)
unknown
Matlatzinca Red (variant
Tripod
unknown)
bowl
Matlatzinca Simple polychrome on Bowl, legs
white
unknown
Red rim bowl
Tripod
bowl
Matlatzinca Red or Red-on-Buff
Simple
fragment
bowl
Plain Jar
Jar
377
316
6
1
6
31
Plain Jar
Jar
CAL_A
378
316
6
1
6
167
Wide rim jar
Jar
CAL_A
379
316
6
1
6
167
Wide rim jar
Jar
CAL_A
380
316
7
2
6
31
Plain Jar
Jar
CAL_A
381
316
7
2
6
31
Plain Jar
Jar
CAL_A
382
316
15
1
6
167
Wide rim jar
Jar
CAL_U
383
316
15
2
6
31
Plain Jar
Jar
CAL_A
384
316
15
2
6
167
Wide rim jar
Jar
CAL_A
385
316
2
6
31
Plain Jar
Jar
CAL_A
386
316
15
1
6
45
Eroded Comal
Comal
CAL_U
387
316
15
1
6
45
Eroded Comal
Comal
CAL_C
388
316
15
1
6
134
Crude Unfinished
CAL_U
389
316
15
2
6
267
Biconical Spiked Censer
Simple
bowl
Censer
390
316
15
2
6
100
Spoon
Spoon
CAL_A
391
317
7
3
6
200
Red rim bowl
CAL_A
392
317
12 8
6
11
Plain Bowl
Simple
bowl
Simple
bowl
5
529
CAL_A
CAL_U
CAL_A
CAL_A
CAL_A
CAL_A
CAL_B
ACH
Context
Samp. # Unit Locus Lot Phase
393
317
12 8
6
394
317
13
4
6
Type
Code Ceramic Type
202 Matlatzinca Red (variant
unknown)
10
Eroded Bowl
395
317
13
4
6
277
396
317
14
6
6
10
397
317
14
6
6
200
398
317
34
4
6
202
399
317
35
4
6
10
400
317
35
4
6
10
401
317
35
4
6
21
402
317
28
6
6
11
403
317
28
6
6
133
404
317
28
6
6
200
405
317
28
6
6
264
406
317
3
6
30
Chem
Macroreg. Chem
Vessel
Form
Group Grp
Tripod
CAL_A
bowl
Simple
CAL_A
bowl
Matlatzinca Red-on-Buff (wide
Bowl, legs CAL_A
band variant)
unknown
Eroded Bowl
Tripod
CAL_A
bowl
Red rim bowl
Simple
CAL_A
bowl
Matlatzinca Red (variant
Tripod
CAL_G
unknown)
bowl
Eroded Bowl
Simple
CAL_A
bowl
Eroded Bowl
Tripod
CAL_A
bowl
Aztec III, Black-on-Orange
Tripod
CAL_U
bowl
Plain Bowl
Simple
CAL_A
bowl
Matlatzinca Red-on-Buff (multiple Simple
CAL_A
sets of parallel lines)
bowl
Red rim bowl
Simple
CAL_U
bowl
Matlatzinca Red or Red-on-Buff
Tripod
CAL_A
fragment
bowl
Eroded Jar
Jar
CAL_A
407
317
12
8
6
31
Plain Jar
Jar
CAL_A
408
317
12
8
6
31
Plain Jar
Jar
CAL_A
409
317
34
4
6
167
Wide rim jar
Jar
CAL_A
410
317
35
5
6
30
Eroded Jar
Jar
CAL_A
411
317
35
5
6
38
Other decorated jar
Jar
CAL_U
412
317
28
6
6
30
Eroded Jar
Jar
CAL_A
413
317
28
6
6
167
Wide rim jar
Jar
CAL_C
414
317
28
6
6
167
Wide rim jar
Jar
CAL_U
415
317
28
6
6
167
Wide rim jar
Jar
CAL_U
7
530
ACH
Context
Samp. # Unit Locus Lot Phase
416
317
7 3
6
Type
Code Ceramic Type
100 Spoon
417
317
418
317
419
7
Vessel
Form
Spoon
Chem
Macroreg. Chem
Group Grp
CAL_U
3
6
45
Eroded Comal
Comal
CAL_A
13
4
6
64
Texcoco Molded Sahumador
CAL_A
317
35
5
6
262
Toluca Valley Sahumador
420
317
28
6
6
113
Guinda Goblet
Frying Pan
Censer
Frying Pan
Censer
Goblet
421
320
5
5
2
11
Plain Bowl
CAL_A
422
320
10 5
2
10
423
320
10
5
2
11
424
320
10
5
2
221
425
320
10
5
2
132
426
320
10
5
2
215
427
320
14
3
2
10
428
320
16
4
2
277
429
320
17
4
2
137
430
320
17
4
2
132
431
320
17
6
2
11
432
320
17
6
2
202
433
320
17
6
2
203
434
320
22
3
2
202
435
320
22
3
2
226
436
320
10
3
2
30
Simple
bowl
Simple
Eroded Bowl
bowl
Plain Bowl
Simple
bowl
Matlatzinca Red-on-Buff (variant Simple
unknown)
bowl
Matlatzinca Red-on-Buff (parallel Bowl, legs
line variant)
unknown
Matlatzinca red on white
Bowl, legs
unknown
Eroded Bowl
Simple
bowl
Matlatzinca Red-on-Buff (wide
Tripod
band variant)
bowl
Matlatzinca Incised Redware
Tripod
(exterior)
bowl
Matlatzinca Red-on-Buff (parallel Tripod
line variant)
bowl
Plain Bowl
Tripod
bowl
Matlatzinca Red (variant
Tripod
unknown)
bowl
Matlatzinca Red (simple red)
Tripod
bowl
Matlatzinca Red (variant
Tripod
unknown)
bowl
Matlatzinca Red-on-Buff (simple Tripod
interior and exterior decoration
bowl
Eroded Jar
Jar
437
320
10
3
2
30
Eroded Jar
Jar
CAL_A
438
320
10
5
2
30
Eroded Jar
Jar
CAL_A
531
CAL_A
CAL_A
CAL_U
CAL_U
CAL_A
CAL_U
CAL_C
CAL_A
CAL_A
CAL_E
CAL_A
CAL_U
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
East
Morelos
ACH
Context
Samp. # Unit Locus Lot Phase
439
320
10 5
2
Type
Code Ceramic Type
30
Eroded Jar
440
320
16
4
2
30
Eroded Jar
Jar
CAL_A
441
320
16
4
2
167
Wide rim jar
Jar
CAL_A
442
320
17
4
2
30
Eroded Jar
Jar
CAL_A
443
320
22
3
2
30
Eroded Jar
Jar
CAL_U
444
320
22
3
2
31
Plain Jar
Jar
CAL_A
445
320
22
3
2
256
Jar
CAL_A
446
320
16
4
2
267
Matlatzinca Red-on-Buff Jar (sets
of parallel lines variant) Biconical
Spiked Censer
Censer
CAL_U
447
320
16
4
2
267
Biconical Spiked Censer
Censer
CAL_A
448
320
17
4
2
134
Crude Unfinished
449
320
17
6
2
132
450
320
22
3
2
170
451
322
1
2
4
202
452
322
1
2
4
200
453
322
2
2
4
10
454
322
2
2
4
10
455
322
2
2
4
132
456
322
2
2
4
132
457
322
2
2
4
122
458
322
2
2
4
200
459
322
2
3
4
10
460
322
2
3
4
11
461
322
2
3
4
202
Vessel
Form
Jar
Simple
bowl
Matlatzinca Red-on-Buff (parallel Basin
line variant)
Frying Pan
Other Censer
Censer
Matlatzinca Red (variant
Tripod
unknown)
bowl
Red rim bowl
Bowl, legs
unknown
Eroded Bowl
Simple
bowl
Eroded Bowl
Simple
bowl
Matlatzinca Red-on-Buff (parallel Bowl, legs
line variant)
unknown
Matlatzinca Red-on-Buff (parallel Bowl, legs
line variant)
unknown
Other decorated bowl
Simple
bowl
Red rim bowl
Simple
bowl
Eroded Bowl
Simple
bowl
Plain Bowl
Simple
bowl
Matlatzinca Red (variant
Bowl, legs
unknown)
unknown
532
Chem
Macroreg. Chem
Group Grp
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_A
CAL_U
CAL_A
CAL_A
CAL_A
CAL_A
CAL_U
CAL_A
CAL_A
ACH
Context
Samp. # Unit Locus Lot Phase
462
322
2
3
4
463
322
2
3
4
Type
Code Ceramic Type
202 Matlatzinca Red (variant
unknown)
200 Red rim bowl
464
322
2
3
4
208
Guinda Black on Red
465
322
2
3
4
114
Aztec III/IV, Black-on-Orange
466
322
1
2
4
31
Plain Jar
467
322
1
2
4
31
Plain Jar
Jar
CAL_E
468
322
2
2
4
31
Plain Jar
Jar
CAL_A
469
322
2
2
4
31
Plain Jar
Jar
CAL_A
470
322
2
2
4
31
Plain Jar
Jar
CAL_A
471
322
2
2
4
167
Wide rim jar
Jar
CAL_A
472
322
2
3
4
31
Plain Jar
Jar
CAL_U
473
322
2
3
4
31
Plain Jar
Jar
CAL_A
474
322
2
3
4
31
Plain Jar
Jar
CAL_A
475
322
2
3
4
31
Plain Jar
Jar
CAL_B
476
322
1
2
4
221
Basin
CAL_A
477
322
2
2
4
267
Matlatzinca Red-on-Buff (variant
unknown)
Biconical Spiked Censer
Censer
CAL_A
478
322
2
2
4
267
Biconical Spiked Censer
Censer
CAL_E
479
322
2
2
4
267
Biconical Spiked Censer
Censer
CAL_A
480
322
2
2
4
72
Matlatzinca Pitcher
Pitcher
CAL_A
481
323
2
9
2
11
Plain Bowl
CAL_A
482
323
2
9
2
202
483
323
2
9
2
221
484
323
2 12
2
11
Matlatzinca Red (variant
unknown)
Matlatzinca Red-on-Buff (variant
unknown)
Plain Bowl
Simple
bowl
Tripod
bowl Bowl,
legs
unknown
Simple
bowl
533
Vessel
Form
Bowl, legs
unknown
Simple
bowl
Simple
bowl
Tripod
bowl
Jar
Chem
Group
CAL_A
CAL_A
CAL_Gu
inda
CAL_U
CAL_E
CAL_C
CAL_A
CAL_A
Macroreg. Chem
Grp
ACH
Context
Samp. # Unit Locus Lot Phase
485
323
2 12
2
Type
Code Ceramic Type
11
Plain Bowl
Chem
Group
CAL_B
486
323
2 12
2
11
CAL_U
487
323
2 12
2
11
488
323
2 12
2
277
489
323
13
2
2
11
490
323
13
2
2
133
491
323
13
2
2
214
492
323
13
2
2
225
493
323
13
2
2
230
494
323
13
3
2
11
495
323
13
3
2
11
496
323
2
8
2
31
Vessel
Form
Simple
bowl
Simple
Plain Bowl
bowl
Plain Bowl
Simple
bowl
Matlatzinca Red-on-Buff (wide
Tripod
band variant)
bowl
Plain Bowl
Simple
bowl
Matlatzinca Red-on-Buff (multiple Simple
sets of parallel lines)
bowl
Matlatzinca Complex polychrome Tripod
on white
bowl
Matlatzinca Red-on-Buff (int.
Tripod
panels)
bowl
Red-on-Buff with negative
Tripod
bowl
Plain Bowl
Simple
bowl
Plain Bowl
Simple
bowl
Plain Jar
Jar
497
323
2
9
2
255
Jar
CAL_U
498
323
2
12
2
31
Matlatzinca Red-on-Buff Jar
(variant unknown)
Plain Jar
Jar
CAL_A
499
323
13
2
2
31
Plain Jar
Jar
CAL_A
500
323
13
2
2
31
Plain Jar
Jar
CAL_U
501
323
13
2
2
31
Plain Jar
Jar
CAL_U
502
323
13
2
2
255
Jar
CAL_U
503
323
13
2
2
31
Matlatzinca Red-on-Buff Jar
(variant unknown)
Plain Jar
Jar
CAL_G
504
323
13
3
2
255
Jar
CAL_A
505
323
13
3
2
31
Matlatzinca Red-on-Buff Jar
(variant unknown)
Plain Jar
Jar
CAL_G
506
323
2 12
2
31
Crude Unfinished
CAL_U
507
323
2 12
2
31
Crude Unfinished
Simple
bowl
Minature
Vessel
534
Macroreg. Chem
Grp
CAL_A
CAL_A
CAL_C
CAL_C
CAL_E
CAL_U
CAL_U
CAL_G
CAL_A
CAL_A
CAL_A
East
Morelos
ACH
Context
Samp. # Unit Locus Lot Phase
508
323
13 2
2
Type
Code Ceramic Type
31
Toluca Valley Sahumador
509
323
13
2
2
31
Toluca Valley Sahumador
510
323
13
2
2
31
Other vessel form
511
324
16
3
2
31
Plain Bowl
512
324
16
3
2
11
Plain Bowl
513
324
16
3
2
11
Plain Bowl
514
324
16
3
2
11
Plain Bowl
515
324
16
3
2
225
516
324
16
4
2
10
Matlatzinca Red-on-Buff (interior
panels variant)
Eroded Bowl
517
324
16
4
2
200
Red rim bowl
518
324
16
4
2
11
Plain Bowl
519
324
16
4
2
11
Plain Bowl
520
324
16
4
2
11
Plain Bowl
521
324
16
4
2
225
522
324
16
4
2
225
523
324
16
4
2
230
Matlatzinca Red-on-Buff (interior
panels variant)
Matlatzinca Red-on-Buff (interior
panels variant)
Red-on-Buff with negative
524
324
16
4
2
173
525
324
16
5
2
215
Matlatzinca Red-on-Buff (interior
and exerior decoration in panels)
Matlatzinca red on white
526
324
16
3
2
31
Plain Jar
527
324
16
3
2
31
Plain Jar
Jar
CAL_C
528
324
16
3
2
31
Plain Jar
Jar
CAL_U
529
324
16
3
2
31
Plain Jar
Jar
CAL_C
530
324
16
3
2
31
Plain Jar
Jar
CAL_U
535
Chem
Macroreg. Chem
Vessel
Form
Group Grp
Frying Pan CAL_U
Censer
Frying Pan CAL_C
Censer
Frying Pan CAL_C
Censer
Simple
CAL_A
bowl
Simple
CAL_U
bowl
Bowl, legs CAL_U
unknown
Simple
CAL_E
East
bowl
Morelos
Simple
CAL_A
bowl
Simple
CAL_B
bowl
Simple
CAL_B
bowl
Simple
CAL_A
bowl
Bowl, legs CAL_E
West
unknown
Morelos
Simple
CAL_U
bowl
Bowl, legs CAL_U
unknown
Tripod
CAL_C
bowl
Tripod
CAL_C
bowl
Simple
CAL_C
bowl
Tripod
CAL_U
bowl
Jar
CAL_U
ACH
Context
Samp. # Unit Locus Lot Phase
Type
Code Ceramic Type
Vessel
Form
531
324
16
3
2
255
Jar
CAL_U
532
324
16
4
2
255
Jar
CAL_U
533
324
16
4
2
31
Matlatzinca Red-on-Buff Jar
(variant unknown)
Matlatzinca Red-on-Buff Jar
(variant unknown)
Plain Jar
Jar
CAL_C
534
324
16
4
2
31
Plain Jar
Jar
CAL_U
535
324
16
5
2
31
Plain Jar
Jar
CAL_U
536
324
16
3
2
45
Eroded Comal
Comal
CAL_U
537
324
16
3
2
267
Biconical Spiked Censer
Censer
CAL_U
538
324
16
4
2
134
Crude Unfinished
CAL_U
539
324
16
4
2
134
Crude Unfinished
540
324
16
4
2
262
Toluca Valley Sahumador
541
324
1
5
6
202
542
324
1
5
6
264
543
324
4
4
6
11
Matlatzinca Red (variant
unknown)
Matlatzinca Red or Red-on-Buff
fragment
Plain Bowl
544
324
4
4
6
11
Plain Bowl
545
324
4
4
6
11
Plain Bowl
546
324
4
4
6
11
Plain Bowl
547
324
4
4
6
202
548
324
4
4
6
264
549
324
4
5
6
10
Matlatzinca Red (variant
unknown)
Matlatzinca Red or Red-on-Buff
fragment
Eroded Bowl
550
324
4
5
6
222
551
324
6
4
6
264
552
324
6
4
6
200
Matlatzinca Red-on-Buff
(interlocking triangle variant)
Matlatzinca Red or Red-on-Buff
fragment Red
rim bowl
553
324
6
4
6
11
Plain Bowl
Minature
Vessel
Minature
Vessel
Frying Pan
Censer
Tripod
bowl
Simple
bowl
Simple
bowl
Simple
bowl Bowl,
legs
unknown
Simple
bowl
Tripod
bowl
Simple
bowl Bowl,
legs
unknown
Bowl, legs
unknown
Bowl, legs
unknown
Tripod
bowl Bowl,
legs
unknown
536
Chem
Group
CAL_U
CAL_U
CAL_A
CAL_U
CAL_U
CAL_A
CAL_U
CAL_A
CAL_A
CAL_A
CAL_A
CAL_C
CAL_A
CAL_U
CAL_A
Macroreg. Chem
Grp
Table C.2 Catalog of type-based sample of ceramic sherds submitted for INAA, with
project chemical source group assignments and macroregional matches. For a full list of
ceramic type descriptions, see Table A.2.
537
MES
Context
Type
Samp. # Unit Locus Lot Phase Code Ceramic Type
101
304
1
2
4
192 Guinda Black-on-Red (Fine line
variant)
102
307
11
1
6
Vessel
Form
Chem
Group
Simple
bowl
CAL_Gui
nda
Import?
CAL_Gui
192 Guinda Black-on-Red (Fine line
variant)
Simple
bowl
103
315
7
1
2
192 Guinda Black-on-Red (Fine line
variant)
Simple
bowl
104
316
9
2
4
192 Guinda Black-on-Red (Fine line
variant)
Simple
bowl
105
307
16
1
6
208 Guinda Black-on-Red
Simple
bowl
106
315
12
3
2
208 Guinda Black-on-Red
Simple
bowl
107
307
20
8
4
208 Guinda Black-on-Red
Simple
bowl
nda
Import?
CAL_Gui
nda
Import?
CAL_A
CAL_Gui
nda
Import?
CAL_Gui
nda
Import?
CAL_Gui
Tenochtitlan
Tenochtitlan
nda
Import?
CAL_Gui
nda
Import?
CAL_U
108
316
11
5
4
208 Guinda Black-on-Red
Simple
bowl
110
310
4
4
2
207 Interior Incised Red
Tripod
bowl
111
315
18
1
2
207 Interior Incised Red
Tripod
bowl
CAL_G
112
315
21
11
2
207 Interior Incised Red
Tripod
bowl
CAL_U
113
323
11
3
2
207 Interior Incised Red
Tripod
bowl
CAL_E
114
307
8
7
4
207 Interior Incised Red
Tripod
bowl
CAL_G
115
303
1
6
4
137 Exterior Incised Red
Tripod
bowl
CAL_F
116
307
16
1
6
137 Exterior Incised Red
Tripod
bowl
CAL_U
117
315
17
2
2
137 Exterior Incised Red
Tripod
bowl
CAL_U
118
315
12
2
2
137 Exterior Incised Red
Tripod
bowl
CAL_G
119
315
17
1
2
137 Exterior Incised Red
Tripod
bowl
CAL_G
120
307
8
7
4
137 Exterior Incised Red
Tripod
bowl
CAL_E
121
313
1
16
4
204 Toluca Valley Black-and-White-on- Tripod
Red
bowl
CAL_U
538
Macroreg.
Chem Grp
East
Morelos
MES
Context
Type
Samp. # Unit Locus Lot Phase Code Ceramic Type
122
316
8
3
6
204 Toluca Valley Black-and-White-onRed
Vessel
Form
Chem
Group
Tripod
bowl
CAL_A
123
307
20
8
4
204 Toluca Valley Black-and-White-on- Tripod
Red
bowl
CAL_A
124
307
20
8
4
204 Toluca Valley Black-and-White-on- Tripod
Red
bowl
CAL_A
125
320
17
6
2
204 Toluca Valley Black-and-White-on- Tripod
Red
bowl
CAL_A
126
320
17
6
2
204 Toluca Valley Black-and-White-on- Bowl, legs
Red
unknown
CAL_A
127
304
1
7
4
46
Plain Comal
Comal
CAL_B
128
307
14
1
6
46
Plain Comal
Comal
CAL_A
129
307
20
3
6
46
Plain Comal
Comal
CAL_U
130
317
10
3
9
46
Plain Comal
Comal
CAL_U
131
316
8
5
4
46
Plain Comal
Comal
CAL_A
132
316
9
3
4
46
Plain Comal
Comal
CAL_U
133
307
20
4
6
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_A
134
311
11
5
4
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_U
135
311
10
5
4
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_B
136
307
17
3
4
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_B
137
307
17
3
4
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_F
138
307
19
2
4
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_C
139
316
22
3
4
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_D
140
309
7
2
6
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_F
141
309
18
2
6
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_B
539
Macroreg.
Chem Grp
MES
Context
Type
Samp. # Unit Locus Lot Phase Code Ceramic Type
142
316
6
1
6
271 Aztec III Black-on-Orange, Local
Variant A
Vessel
Form
Chem
Group
Tripod
bowl
CAL_B
143
317
13
3
6
271 Aztec III Black-on-Orange, Local
Variant A
Tripod
bowl
CAL_U
144
316
6
1
6
274 Aztec III/IV Black-on-Orange, Local Tripod
Variant A
bowl
CAL_B
145
316
6
1
6
274 Aztec III/IV Black-on-Orange, Local Tripod
Variant A
bowl
CAL_U
146
317
10
3
9
274 Aztec III/IV Black-on-Orange, Local Tripod
Variant A
bowl
CAL_U
147
317
14
7
6
274 Aztec III/IV Black-on-Orange, Local Tripod
Variant A
bowl
CAL_B
148
310
4
4
2
262 Toluca Valley Sahumador
Frying Pan
Censer
CAL_A
149
310
4
2
2
262 Toluca Valley Sahumador
Frying Pan
Censer
CAL_A
150
315
21
1
2
262 Toluca Valley Sahumador
Frying Pan
Censer
CAL_A
151
315
13
2
2
262 Toluca Valley Sahumador
Frying Pan
Censer
CAL_A
152
315
21
2
2
262 Toluca Valley Sahumador
Frying Pan
Censer
CAL_A
153
323
10
6
2
262 Toluca Valley Sahumador
Frying Pan
Censer
CAL_U
154
315
24
2
2
215 Toluca Valley Red-on-White
Tripod
bowl
CAL_A
156
315
22
2
2
188 Toluca Valley Red-on-White (with
panels)
Tripod
bowl
CAL_E
157
323
10
8
2
188 Toluca Valley Red-on-White (with
panels)
Tripod
bowl
CAL_E
158
309
14
2
6
215 Toluca Valley Red-on-White
Tripod
bowl
CAL_A
159
307
8
6
4
215 Toluca Valley Red-on-White
Tripod
bowl
CAL_E
160
307
20
8
4
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
161
315
24
2
2
213 Toluca Valley Polychrome
Tripod
bowl
CAL_H
162
315
24
2
2
213 Toluca Valley Polychrome
Tripod
bowl
CAL_H
540
Macroreg.
Chem Grp
East
Morelos
MES
Context
Type
Samp. # Unit Locus Lot Phase Code Ceramic Type
163
315 21
2
2
213 Toluca Valley Polychrome
Vessel
Form
Chem
Group
Tripod
bowl
CAL_A
164
315
31
2
2
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
165
316
4
9
4
213 Toluca Valley Polychrome
Tripod
bowl
CAL_E
166
317
31
8
6
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
167
307
20
5
6
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
168
323
10
5
2
213 Toluca Valley Polychrome
Tripod
bowl
CAL_E
169
307
8
6
4
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
170
307
20
7
4
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
171
307
20
7
4
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
172
316
11
4
6
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
173
316
11
4
6
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
174
322
2
2
4
213 Toluca Valley Polychrome
Tripod
bowl
CAL_A
175
307
20
6
6
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
176
307
20
9
4
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
177
307
5
1
6
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
178
315
8
1
2
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
179
315
13
1
2
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
180
315
13
1
2
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
181
315
25
1
2
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
182
315
21
2
2
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
541
Macroreg.
Chem Grp
East
Morelos
MES
Context
Type
Samp. # Unit Locus Lot Phase Code Ceramic Type
183
307 20
8
4
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Vessel
Form
Chem
Group
Tripod
bowl
CAL_A
184
323
13
6
2
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_U
185
307
8
6
4
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_U
186
307
8
7
4
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
187
307
8
7
4
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
188
316
5
1
6
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
189
307
20
7
4
222 Toluca Valley Red-on-Buff (int. &
ext. thick red band)
Tripod
bowl
CAL_A
190
307
20
7
4
225 Toluca Valley Red-on-Buff (interior Simple
panels)
bowl
CAL_A
191
310
4
4
2
225 Toluca Valley Red-on-Buff (interior Simple
panels)
bowl
CAL_A
192
310
4
4
2
225 Toluca Valley Red-on-Buff (interior Tripod
panels)
bowl
CAL_C
193
310
4
2
2
225 Toluca Valley Red-on-Buff (interior Bowl, legs
panels)
unknown
CAL_A
194
315
23
2
2
225 Toluca Valley Red-on-Buff (interior Bowl, legs
panels)
unknown
CAL_A
195
315
17
2
2
225 Toluca Valley Red-on-Buff (interior Bowl, legs
panels)
unknown
CAL_A
196
316
13
7
4
225 Toluca Valley Red-on-Buff (interior Tripod
panels)
bowl
CAL_A
197
326
3
6
2
225 Toluca Valley Red-on-Buff (interior Tripod
panels)
bowl
CAL_A
198
315
22
10
2
225 Toluca Valley Red-on-Buff (interior Tripod
panels)
bowl
CAL_U
199
323
10
7
2
225 Toluca Valley Red-on-Buff (interior Tripod
panels)
bowl
CAL_U
200
316
15
2
6
225 Toluca Valley Red-on-Buff (interior Tripod
panels)
bowl
CAL_A
201
316
15
2
6
225 Toluca Valley Red-on-Buff (interior Simple
panels)
bowl
CAL_A
202
316
15
2
6
225 Toluca Valley Red-on-Buff (interior Tripod
panels)
bowl
CAL_A
542
Macroreg.
Chem Grp
MES
Context
Type
Samp. # Unit Locus Lot Phase Code Ceramic Type
203
324
4
4
6
225 Toluca Valley Red-on-Buff (interior
panels)
Vessel
Form
Chem
Group
Tripod
bowl
CAL_A
204
324
4
4
6
225 Toluca Valley Red-on-Buff (interior Bowl, legs
panels)
unknown
CAL_C
205
307
2
3
2
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
206
303
1
12
4
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
207
307
20
8
4
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
208
307
5
1
6
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_U
209
315
18
2
2
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_H
210
315
15
1
2
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
211
324
16
3
2
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
212
323
2
9
2
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
213
307
20
7
4
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
214
307
8
6
4
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_U
215
316
4
9
4
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
216
316
7
2
6
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
217
316
7
2
6
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_U
218
316
15
2
6
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
219
316
15
2
6
277 Toluca Valley Red-on-Buff (wide
bands)
Tripod
bowl
CAL_A
220
310
3
2
2
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
221
307
20
8
4
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
222
307
4
2
6
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_U
543
Macroreg.
Chem Grp
MES
Context
Type
Vessel
Samp. # Unit Locus Lot Phase Code Ceramic Type
Form
223
310
4
4
2
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
Chem
Group
CAL_A
224
315
13
2
2
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
225
315
25
2
2
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
226
315
26
2
2
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
227
307
8
9
6
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
228
326
3
6
2
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_E
229
316
16
5
4
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
230
323
5
10
2
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
231
307
20
7
4
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
232
324
4
4
6
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
233
316
15
2
6
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
234
316
15
2
6
256 Toluca Valley Red-on-Buff (parallel Jar
line jars)
CAL_A
235
304
1
2
4
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_D
236
307
20
7
4
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_D
237
307
9
4
6
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_D
238
307
13
1
6
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_B
239
307
20
2
6
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_D
240
316
13
8
4
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_D
241
316
6
1
6
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_D
242
316
9
2
4
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_B
544
Macroreg.
Chem Grp
MES
Context
Type
Samp. # Unit Locus Lot Phase Code Ceramic Type
243
317 28
6
6
21 Aztec III Black-on-Orange
Vessel
Form
Chem
Group
Tripod
bowl
CAL_D
244
307
10
2
6
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_B
245
307
20
7
4
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_D
246
307
20
7
4
21
Aztec III Black-on-Orange
Tripod
bowl
CAL_D
247
316
5
1
6
114 Aztec III/IV Black-on-Orange
Tripod
bowl
CAL_B
248
316
7
2
6
114 Aztec III/IV Black-on-Orange
Tripod
bowl
CAL_B
249
324
1
8
6
182 Aztec III/IV Black-on-Orange, Local Tripod
variant B
bowl
CAL_B
250
316
6
1
6
114 Aztec III/IV Black-on-Orange
Tripod
bowl
CAL_D
251
317
28
5
6
114 Aztec III/IV Black-on-Orange
Tripod
bowl
CAL_U
252
317
35
4
6
114 Aztec III/IV Black-on-Orange
Tripod
bowl
CAL_B
253
307
20
8
4
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_D
254
307
20
1
6
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_D
255
317
14
6
6
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_D
256
317
35
4
6
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_U
257
317
28
5
6
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_U
258
322
2
4
4
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_D
259
322
2
2
4
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_D
260
324
6
4
6
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_D
261
307
19
2
4
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_B
262
311
10
5
4
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_D
545
Macroreg.
Chem Grp
MES
Context
Type
Samp. # Unit Locus Lot Phase Code Ceramic Type
263
317 64
3
6
181 Aztec III Black-on-Orange, Local
variant B
Vessel
Form
Chem
Group
Tripod
bowl
CAL_E
264
309
17
3
6
181 Aztec III Black-on-Orange, Local
variant B
Tripod
bowl
CAL_U
265
309
17
3
6
182 Aztec III/IV Black-on-Orange, Local Tripod
variant B
bowl
CAL_B
266
317
14
7
6
182 Aztec III/IV Black-on-Orange, Local Tripod
variant B
bowl
CAL_D
267
309
14
2
6
182 Aztec III/IV Black-on-Orange, Local Tripod
variant B
bowl
CAL_U
268
316
15
1
6
182 Aztec III/IV Black-on-Orange, Local Tripod
variant B
bowl
CAL_B
269
317
13
4
6
182 Aztec III/IV Black-on-Orange, Local Tripod
bowl
variant B
CAL_B
270
307
20
3
6
271 Aztec III Black-on-Orange, Local
Variant A
CAL_B
546
Tripod
bowl
Macroreg.
Chem Grp
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
APPENDIX D
PETROGRAPHIC SAMPLE LIST AND FABRIC DESCRIPTION REPORT
580
This appendix includes a catalog of ceramic samples submitted to Dr. Jennifer
Meanwell for thin section petrography and her report describing the paste groups that she
identified based on those samples. The petrographic batch codes refer to the year of
sampling, as well as the reason for sample selection.
Table E.1 Catalog of ceramic samples submitted for thin section petrography, with
resulting petrographic fabric group assignments.
581
Batch
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-3
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
2009-2
Code
TSN 1
TSN 2
TSN 3
TSN 4
TSN 5
TSN 6
TSN 7
TSN 8
TSN 9
TSN 10
TSN 11
TSN 12
TSN 13
TSN 14
TSN 15
TSN 16
TSN 17
TSN 18
TSN 19
TSN 20
TSN 21
TSN 22
TSN 23
TSN 24
TSN 25
TSN 26
TSN 27
TSN 28
TSN 29
TSN 30
TSN 31
TSN 32
TSN 33
TSN 34
TSN 35
TSN 36
TSN 37
TSN 38
TSN 39
Provenience
Unit Locus Lot
307 20
4
307 20
4
307 20
6
307 20
5
307 20
4
307 20
3
307 20
6
307 20
6
307 20
4
307 20
3
307 20
3
307 20
3
307 20
5
307 20
3
307 20
6
307 20
3
307 20
6
307 20
5
307 20
3
307 20
3
307 20
6
307 20
5
307 20
7
307
8
7
307 20
7
307 20
7
307 20
7
307
8
7
307
8
7
307
8
7
307 20
7
307
8
6
307
8
6
307 20
7
307
8
7
307
8
7
307 20
7
307 20
7
307 20
7
INAA Petro
Phase Type Macro- Paste Group Fabric
6
167
C
I-C
6
167
C
X
6
167
C
I-A
6
31
A
III
6
167
E
I-C
6
253
E
I-A
6
266
E
II
6
31
E
I-A
6
208
G
XIII
6
208
G
XIII
6
132
C
I-C
6
11
C
I-A
6
226
E
III
6
200
E
III-var?
6
202
E
II
6
264
E
I-C
6
267
A
III
6
262
D
XIV
6
68
A
XV
6
45
E
I-A
6
267
E'
I-A
6
71
E
XIV
4
167
C
III
4
31
D
I-A
4
31
D
I-C
4
31
D
I-C
4
167
C
III
4
31
E
I-A
4
31
E'
I-A
4
31
E
I-A
4
209
J
X
4
11
C
X
4
11
B
XIV
4
11
D
XV
4
133
E'
III
4
202
E'
I-C
4
222
E
I-C
4
200
E
I-C
4
169
C
II
582
Batch
2009-2
2009-2
2009-2
2009-2
2009-2
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-1
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
Code
TSN 40
TSN 41
TSN 42
TSN 43
TSN 44
TSN 45
TSN 46
TSN 47
TSN 48
TSN 49
TSN 50
TSN 51
TSN 52
TSN 53
TSN 54
TSN 55
TSN 56
TSN 57
TSN 58
TSN 59
TSN 60
TSN 61
TSN 62
TSN 63
TSN 64
TSN 65
TSN 66
TSN 67
TSN 68
TSN 69
TSN 70
TSN 71
TSN 72
TSN 73
TSN 74
TSN 75
TSN 76
TSN 77
TSN 78
Provenience
Unit Locus Lot
307 20
7
307
8
6
307 20
7
307
8
6
307 20
7
307
5
3
307
5
2
307 12
4
307 12
4
307
5
2
307
2
4
307 12
3
307 12
3
307
2
4
307 12
4
307
5
2
307
2
3
307
2
3
307
2
3
307
2
4
307 12
4
307 12
3
307 12
4
307 12
4
307 12
4
307 12
4
307 12
4
300 403
1
300 403
1
300 11
1
300 396
1
300 403
1
300 454
1
300 153
1
300
0
0
300 357
1
300 403
1
300
5
1
300 357
1
INAA Petro
Phase Type Macro- Paste Group Fabric
4
101
D
III
4
60
J
X
4
267
E'
I-C
4
267
E
I-A
4
170
E'
I-A
2
31
C
II
2
31
D
III
2
31
D
I-A
2
31
D
I-A
2
31
E'
III
2
31
E'
III
2
31
E'
I-C
2
31
E
I-C
2
11
D
III
2
11
C
I-A
2
11
J
I-B
2
11
D
III
2
277
E'
I-C
2
11
F
I-B
2
213
E'
I-C
2
11
E
I-B
2
134
D
I-C
2
134
D
I-C
2
134
D
II
2
267
E'
I-A
2
267
F
III
2
169
E
I-B
11
A
III
225
A
III
30
A
X
30
A
III
11
A
III
10
A
X
11
B
XIV
31
B
X
11
B
II
225
B
XIII
10
B
II
11
B
XIV
583
Batch
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-10
2009-5
2009-5
Provenience
Code
Unit Locus Lot
TSN 79 300 569
1
TSN 80 300 781
1
TSN 81 300 376
1
TSN 82 300 569
1
TSN 83 300 365
1
TSN 84 300 60
1
TSN 85 300 60
1
TSN 86 300
0
0
TSN 87 300
0
0
TSN 88 300
0
0
TSN 89 300
0
0
TSN 90 300 60
1
TSN 91 300 153
1
TSN 92 300 53
1
TSN 93 300
4
1
TSN 94 300 353
1
TSN 95 300 153
1
TSN 96 300 353
1
TSN 97 300 153
1
TSN 98 300 371
1
TSN 99 300 158
1
TSN 100 300 371
1
TSN 101 300 354
1
TSN 102 300 232
1
TSN 103 300 153
1
TSN 104 300 466
1
TSN 105 300 53
1
TSN 106 300 357
1
TSN 107 300 153
1
TSN 108 300 365
1
TSN 109 300 407
1
TSN 110 300 796
1
TSN 111 300 796
1
TSN 112 300 232
1
TSN 113 300 810
1
TSN 114 300 153
1
TSN 115 300 354
1
TSN 116 307 20
4
TSN 117 307 20
5
INAA Petro
Phase Type Macro- Paste Group Fabric
167
C
I-C
11
C
II
167
C
II-var
10
C
XV
264
C
XIV
11
C
II
31
D
I-C
30
D
III
134
D
I-var
11
D
I-var
10
D
III
169
D
I-D
134
E
IX
30
E
I-B
10
E
III
30
E
I-C
31
E
I-A
31
E
I-C
256
E'
I-C
31
E'
I-A
31
E'
III
31
E'
I-D
264
E'
II
11
E'
I-B
31
F
I-C
213
F
II
30
F
III
31
F
III
225
F
I-B
31
F
I-B
208
G
XIV
31
J
I-A
30
J
I-C
167
J
XV
167
J
I-D
167
J
I-A
31
J
IX
6
114
aztec
XI
6
215
high bio
XV
584
Batch
2009-5
2009-5
2009-5
2009-4
2009-4
2009-4
2009-4
2009-4
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-6
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
Provenience
Code
Unit Locus Lot
TSN 118 307 20
3
TSN 119 307 20
3
TSN 120 307 20
6
TSN 121 307
8
6
TSN 122 307 20
7
TSN 123 307
8
7
TSN 124 307
8
7
TSN 125 307 20
7
TSN 126 316 15
4
TSN 127 316
7
7
TSN 128 316
7
7
TSN 129 316 15
4
TSN 130 316 15
4
TSN 131 316 15
4
TSN 132 316
6
5
TSN 133 316
6
5
TSN 134 316 15
4
TSN 135 316
7
7
TSN 136 316 15
4
TSN 137 316
7
7
TSN 138 316
7
7
TSN 139 316
5
6
TSN 140 316
6
5
TSN 141 316 15
4
TSN 142 316
5
6
TSN 143 316
5
4
TSN 144 316
7
7
TSN 145 316
7
7
TSN 146 316 15
4
TSN 147 316
7
7
TSN 148 316
6
3
TSN 149 316
6
4
TSN 150 316 15
3
TSN 151 316
6
3
TSN 152 316 16
4
TSN 153 316 22
2
TSN 154 316 16
5
TSN 155 316
6
3
TSN 156 316 22
2
INAA Petro
Phase Type Macro- Paste Group Fabric
6
181
B
XIV
6
21
aztec
XI
6
182
B
XI
4
180
D
I-C
4
180
F
I-A
4
207
E
CAL_G VIII
4
137
D
CAL_E VII
4
134
E'
I-A
2
31
E'
I-A
2
256
E'
I-B
2
31
E
I-C
2
31
E'
I-B
2
30
A
X
2
31
D
I-A
2
30
D
III
2
38
B
X
2
11
F
I-C
2
225
F
I-B
2
277
E
I-D
2
225
F
I-B
2
132
A
I-B
2
11
A
III
2
11
C
II
2
11
A
II
2
267
E'
I-A
2
267
F
I-B
2
170
F
III
2
180
D
I-C
2
134
D
III
2
68 A?salt vessel
XV
4
31
E
I-D
4
31
E'
I-B
4
167
E
I-D
4
31
F
I-B
4
256
A
III
4
30
C
XIV
4
31
A
XIV
4
31
D
III
4
11
F
I-B
585
Batch
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-7
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2009-8
2013
2013
2013
2013
Code
TSN 157
TSN 158
TSN 159
TSN 160
TSN 161
TSN 162
TSN 163
TSN 164
TSN 165
TSN 166
TSN 167
TSN 168
TSN 169
TSN 170
TSN 171
TSN 172
TSN 173
TSN 174
TSN 175
TSN 176
TSN 177
TSN 178
TSN 179
TSN 180
TSN 181
TSN 182
TSN 183
TSN 184
TSN 185
TSN 186
TSN 187
TSN 188
TSN 189
TSN 190
TSN 191
ACH 181
ACH 193
ACH 204
ACH 214
Provenience
Unit Locus Lot
316
6
3
316 16
3
316
6
4
316 16
4
316 22
2
316 16
2
316 16
5
316 15
3
316 15
3
316 22
2
316
6
3
316 16
5
316 16
4
316
5
1
316
5
1
316
5
1
316 15
2
316
5
1
316
5
1
316 15
2
316
6
2
316
6
2
316 15
2
316
5
1
316 15
2
316
5
1
316
6
2
316
5
1
316
5
2
316
6
2
316
5
1
316 15
1
316 15
2
316
6
2
316 15
2
309
7
2
309 18
3
309 25
3
311
1
6
Phase Type Macro- Paste
4
11
F
4
222
E'
4
203
F
4
11
B
4
11
D
4
10
aztec
4
10
D
4
267
E'
4
262
E'
4
267
E
4
134
J
4
134
D
4
68 A?salt vessel
6
167
E
6
31
F
6
31
E'
6
167
E
6
167
C
6
256
D
6
31
A
6
167
A
6
11
F
6
264
F
6
10
E'
6
269
E'
6
192
G
6
114
Aztec
6
182
B
6
182
B
6
267
E'
6
134
E'
6
267
E
6
180
D
6
46
Aztec
6
46
A
6
271
C
6
132
E
6
30
E-prime
4
11
E-prime
586
INAA Petro
Group Fabric
I-B
I-A
III
XIV
I-B
XIII
III
I-A
I-B
II
I-B
III
XV
I-C
II
I-C
I-C
I-A
I-B
X
I-A
I-B
III
I-B
III
XIII
XI
XIII
XII-var
I-C
I-B
I-B
I-B
XIV
XV
CAL_A II
CAL_A I-A
CAL_A I-B
CAL_A I-A
Batch
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
2013
Code
ACH 224
ACH 234
ACH 481
ACH 499
ACH 507
MES 128
MES 131
MES 133
ACH 267
ACH 392
MES 127
MES 141
MES 144
MES 147
MES 248
ACH 387
ACH 413
ACH 510
ACH 533
MES 204
ACH 45
MES 239
MES 250
MES 253
MES 255
ACH 37
MES 113
MES 157
MES 168
MES 228
ACH 4
ACH 8
ACH 24
ACH 28
MES 137
ACH 494
ACH 505
MES 109
MES 114
Provenience
Unit Locus Lot
311 11
6
311 21
3
323
2
9
323 13
2
323
2
12
307 14
1
316
8
5
307 20
4
311
5
6
317 12
8
304
1
7
309 18
2
316
6
1
317 14
7
316
7
2
316 15
1
317 28
6
323 13
2
324 16
4
324
4
4
303
1
7
307 20
2
316
6
1
307 20
8
317 14
6
303
1
8
323 11
3
323 10
8
323 10
5
326
3
303
6
9
303
6
9
303
6
9
303
6
9
307 17
3
323 13
3
323 13
3
300-SURVEY
307
8
7
INAA
Phase Type Macro- Paste Group
4
213
E-prime
CAL_A
4
30
Other
CAL_A
2
11
E
CAL_A
2
31
E-prime
CAL_A
2
31
J
CAL_A
6
46
C
CAL_A
4
46 Coarse Orange CAL_A
6
271
C
CAL_A
4
46 Coarse Orange CAL_B
6
11
Aztec
CAL_B
4
46 Coarse Orange CAL_B
6
271
C
CAL_B
6
274
C
CAL_B
6
274
C
CAL_B
6
114
Aztec
CAL_B
6
45
E-prime
CAL_C
6
167
Other
CAL_C
2
31
C
CAL_C
2
31
E
CAL_C
6
225
E
CAL_C
4
11
Aztec
CAL_D
6
21
Aztec
CAL_D
6
114
Aztec
CAL_D
4
181
B
CAL_D
6
181
B
CAL_D
4
166
C
CAL_E
2
207
E
CAL_E
2
188
CAL_E
2
213
E
CAL_E
2
256
E
CAL_E
4
11
E-prime
CAL_F
4
185
C
CAL_F
4
31
F
CAL_F
4
134
E
CAL_F
4
271
C
CAL_F
2
11
E
CAL_G
2
31
A
CAL_G
n/a
207
E
CAL_G
4
207
CAL_G
587
Petro
Fabric
I-D
XV
I-B
I-A
X
III
III
II
XII
XII
XII
XII-var
XII
XII
XI
I-A
I-A
I-A
I-C
I-C
XV
XI
XI
XI-var
XV
VII
VII
VI
VII
III
I-A
I-B
III
III
II
I-C
IX
XV
VIII
Batch
2013
2013
2013
2013
2013
2013
2013
2013
2013
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
2014
Code
MES 119
MES 103
ACH 271
ACH 272
ACH 273
ACH 289
MES 162
MES 129
MES 130
ACH 277
ACH 287
ACH 300
MES 136
ACH 119
MES 265
ACH 529
ACH 523
ACH 524
MES 235
MES 240
MES 266
MES 120
MES 165
ACH 491
MES 111
MES 118
ACH 398
ACH 280
ACH 299
ACH 295
ACH 65
ACH 87
ACH 66
ACH 103
ACH 113
ACH 117
ACH 222
ACH 240
ACH 243
Provenience
Unit Locus Lot
315 17
1
315
7
1
315 12
10
315 12
10
315 13
9
315 21
10
315 24
2
307 20
3
317 10
3
315 13
13
315 13
13
315 22
10
307 17
3
307 20
8
309 17
3
324 16
3
324 16
4
324 16
4
304
1
2
316 13
8
317 14
7
307
8
7
316
4
9
323 13
2
315 18
1
315 12
2
317 34
4
315 14
8
315 17
7
315 28
1
307
2
3
307
5
2
307
5
2
307 20
8
307 20
8
307 20
8
311
5
7
311 21
3
311
2
8
INAA Petro
Phase Type Macro- Paste Group Fabric
2
137
E
CAL_G IX
2
192
G
CAL_GuinXIII
2
222
E-prime
CAL_H III
2
170
E-prime
CAL_H I-C
2
11
C
CAL_H I-A
2
31
E-prime
CAL_H III
2
213
E
CAL_H I-C
6
46 Coarse Orange CAL_U III
9
46 Coarse Orange CAL_U III
2
132
E-prime
CAL_A I-D
2
31
E
CAL_A I-A
2
134
E
CAL_A XV
4
271
C
CAL_B XI
4
46
Coarse Or CAL_B XV
6
182
B
CAL_B XII
2
31
E-prime
CAL_C I-A
2
230
E
CAL_C III
2
173
E-prime
CAL_C III
4
21
Aztec
CAL_D XI
4
21
Aztec
CAL_D XII
6
182
B
CAL_D XII
4
137
CAL_E VII
4
213
E
CAL_E X
2
214
E
CAL_E VII
2
207 High Biotite CAL_G VIII
2
137 High Biotite CAL_G I-A
6
202
E-prime
CAL_G X
2
225
E-prime
CAL_H I-D
2
101
E-prime
CAL_H I-B
2
31
F
CAL_H I-D
2
277
E-prime
CAL_U
2
42
E-prime
CAL_U
2
202 High Biotite CAL_U
4
222
E-prime
CAL_U III
4
256
E-prime
CAL_U I-D
4
267
E
CAL_U I-A
4
11
E-prime
CAL_U I-D
4
46
E-prime
CAL_U I-A var
4
221
E
CAL_U IX
588
Batch
2014
2014
2014
2014
2014
2014
Code
ACH 416
ACH 401
ACH 415
ACH 492
ACH 493
ACH 500
Provenience
Unit Locus Lot
317
7
3
317 35
4
317 28
6
323 13
2
323 13
2
323 13
2
INAA
Phase Type Macro- Paste Group
6
100
G
CAL_U
6
21
Aztec
CAL_U
6
167
E
CAL_U
2
225
E-prime
CAL_U
2
230
E-prime
CAL_U
2
31
F
CAL_U
Petro
Fabric
I-A
XI
I-A
I-A
III
III-var?
Calixtlahuaca Master List of Petrographic Fabric Designations
Jennifer Meanwell
The petrographic fabrics identified at Calixtlahuaca include a set of fabrics that appear
to be made from locally available materials (within 5-10 km of the site, most likely). These
fabrics all contain similar inclusions that are consistent with the local geology around
Calixtlahuaca, and are divided based on the optical activity of the matrix, the size and sorting of
the inclusions, and slight variations in the relative frequency of different inclusions. The samples
that were analyzed via chemical analysis as well as petrographic analysis also place the “local”
fabrics into the local chemical groups. A number of presumably non-local fabrics have also been
identified, and outlier sherds were placed into a presumed region of origin where possible. A
summary table of the fabrics (with subgroups) and their designations appears first, followed by a
more detailed description of each fabric. The fabrics have been organized into three different
regions: local, elsewhere in the Toluca Valley, and Basin of Mexico. All fabrics are now given a
Roman numeral designation (skipping IV and V to avoid confusion with an earlier iteration of
fabric designations), although former designations are noted in the table as well.
Fabric
Code
I-A
Former
Code
Region
Distinctive Characteristics
Local
Typical mineralogy suite, darker colored amphibole, low to
moderate optical activity in the clay matrix
Typical mineralogy suite, lighter colored amphibole,
moderate optical activity in the clay matrix
Typical mineralogy suite, lighter colored amphibole, very
high optical activity in the clay matrix (likely low fired)
Typical mineralogy suite, darker colored amphibole,
moderate to high optical activity in the clay matrix
Typical mineralogy suite, inclusions usually smaller and
I-B
Local
I-C
Local
I-D
Local
II
Local
589
III
VI
VII
VIII
IX
X
XI
XII
XIII
XIV
XV
better sorted than Fabric I, clay matrix has low optical
activity and a light beige color
Local
Typical mineralogy suite, moderate to high optical activity in
the clay matrix, which fires to red and brown shades, some
examples have iron minerals present in volcanic rock
fragments and may have grog temper
OTHER Local
Outlier fabrics that look local in origin
COYO
Toluca
Typical mineral suite, low volume fraction of inclusions, grog
Valley
or clay lumps present, incomplete clay mixing or processing
likely, bimodal size distribution in minerals (likely tempered)
GR
Toluca
Similar texture to Fabric VII above, but higher volume
Valley
fraction of grog/clay lumps relative to mineral inclusions,
mineralogy seems more felsic than most of the Calixtlahuaca
sherds, with higher quartz and biotite, bimodal size
distribution in the minerals (likely tempered)
OA
Toluca
Likely variant from S/SW Toluca Valley, characterized by very
Valley
high optical activity in the clay matrix, bimodal size
distribution in the inclusions (likely tempered), and more
felsic minerals, such as quartz and biotite
OTHER, Toluca
Outlier fabrics that seem to match the other S/SW Toluca
V
Valley
Valley fabrics in mineralogy (relatively more felsic)
RA
Basin of
Extremely fine grained and well sorted, typical mineral suite,
Mexico
reddish clay matrix with generally low optical activity (likely
high fired)
B
Basin of
Very low optical activity in the core (likely high fired), typical
Mexico
mineral suite, varied size and texture in the inclusions, some
examples exhibit bimodal size distribution
ORG,
Basin of
Typical mineral suite, large organic temper burnout voids,
Guinda Mexico
some still containing remnant carbon, clay matrix is
generally fine and well sorted
OTHER, Basin of
Outlier fabrics that seem to match the texture and
IV
Mexico
processing consistent with a Basin of Mexico origin
OTHER Unknown Outliers that cannot be associated with any region studied
within this project
Detailed Fabric Descriptions:
Fabric Groups
Fabric I:
590
Fabric I is characterized by typically intermediate to mafic inclusions, primarily plagioclase
feldspar, with lesser amounts of quartz, amphibole, minerals with a high iron content, and
volcanic rock fragments. Other accessory minerals present in smaller amounts include biotite,
orthopyroxene, and clinopyroxene, as well as fragments of what appears to be volcanic glass.
Many of the plagioclase feldspar grains are zoned. The amphiboles are present in two forms,
which is one of the characteristics used to divide this fabric into subgroups. One amphibole is
consistent with hornblende, and has a deep color, strong pleochroism, and second-order
interference colors. The other amphibole displays a paler green color, weaker pleochroism, and
first-order interference colors, but still exhibits the typical cleavage and crystal habit of
amphiboles. All of the inclusions are generally subangular and range in size from silt sized to
coarse sand, with a few larger fragments seen occasionally. Very few sherds in the collection
show strong evidence for deliberate tempering, and most of these are clearly imported from the
Valley of Mexico or elsewhere.
Fabric I-A contains the deeper-colored amphibole and the clay matrix exhibits low to moderate
optical activity. Fabric I-B contains the lighter-colored amphibole and the clay matrix has
moderate optical activity. Fabric I-C contains the lighter-colored amphibole with a very optically
active clay matrix. Fabric I-D is similar to Fabric I-A, except that the clay matrix is more optically
active.
Fabric II
Fabric II contains the same suite of inclusions present in Fabric I, including plagioclase feldspar,
quartz, amphibole, volcanic rock fragments, volcanic glass, pyroxene, and biotite, and it can be
distinguished from Fabric I primarily based on texture and the clay matrix. Fabric II samples
contain smaller grains and fewer voids. The clay matrix generally displays low optical activity,
and it tends to be lighter in color in plane-polarized light than the clays used for Fabric I. Due to
the strong similarity to Fabric I, Fabric II is likely a manufacturing variant, either using a local clay
deposit that naturally contains smaller inclusions, or by deliberately processing the clay to
contain smaller inclusions.
Fabric III
Fabric III again contains the same general suite of inclusions found in Fabrics I and II, with
plagioclase feldspar being the most common inclusion, followed by quartz, volcanic rock
fragments, amphibole, volcanic glass, pyroxene, and biotite. The major difference is that the
amount of iron present in both the clay matrix and among the inclusions seems to be higher.
The clays fire to a rich reddish brown color (when they are not reduced to a black or grey shade),
and many of the volcanic rock fragments show iron staining. The clay is generally moderately to
very optically active, and the volcanic glass fragments are more often orange, rather than the
typical brownish-gray.
591
Several of the Fabric III examples include grog temper in small amounts. While the sherds that
have grog always have more than one piece of grog in them, it is not the only non-plastic
inclusion within the fabric. In general, though, the grog pieces contain the same set of inclusions
as the surrounding clay matrix, so it likely that other locally produced vessels were added to the
clay mixture for Fabric III, in some cases.
Fabric VII
This small group of sherds is characterized by a fabric that has a low volume fraction of
inclusions with an inhomogeneous clay matrix that is optically active. These sherds also have
fragments of grog or clay lumps in them. The clay matrix preserves clay domains that are likely
related to the production process. This clay does not seem to have been well mixed prior to use,
and the clay preserves evidence of this incomplete mixing, with randomly oriented domains that
are not oriented parallel to the vessel walls.
The inclusions present in the Fabric VII sherds are similar to those found in other Calixtlahuaca
sherds, with plagioclase feldspar, amphibole, quartz, pyroxene, and iron rich inclusions
predominating. The inclusions are generally well sorted and demonstrate a bimodal size
distribution, suggesting that the inclusions were deliberately added as tempering material to the
base clay. The grog fragments do not always match the fabric of the surrounding sherd, so it is
possible that a variety of sherds were processed as grog to be added to Fabric VII sherds.
Fabric VIII
Although grog is found as an inclusion in the local Calixtlahuaca Fabric III, the sherds placed in
this fabric are full of grog fragments. The Fabric VIII sherds are made from a clay fabric that has a
dark core (dark brown in thin section, but appears black in hand sample) with a red surface
layer. The clay is moderately optically active, but not quite as active as that seen in the Fabric VII
fabrics. The clay domains are also more oriented parallel to the vessel walls, which may indicate
additional clay mixing and processing during manufacture.
The inclusions are also subtly different. Although the texture and sorting is similar to the
inclusions found in Fabric VII, including the bimodal size distribution that indicates tempering,
the mineralogy is different. These sherds contain more quartz relative to the amount of
plagioclase feldspar, although they also contain the amphibole, pyroxene, and iron rich
inclusions commonly found in the Calixtlahuaca material. In addition, small fragments of biotite
and muscovite are seen, as well as polycrystalline quartz. This suggests that this fabric comes
from a more felsic source region.
Fabric IX
The sherds that are placed in fabric IX were not necessarily made in the same location, although
they do share several production techniques. This fabric shows considerable similarities to
592
fabrics VII and VIII, just made with different clays or with shorter firing times. The fabric is
characterized by extremely high optical activity in the clay matrix. Sherds exhibit a bimodal
inclusion size distribution as well as unusual mineralogy for the Calixtlahuaca material. Although
the typical inclusions are present, quartz is more prevalent than in the Calixtlahuaca material,
and more felsic inclusions are seen, including biotite and possibly a small fragment of muscovite.
NOTE: With further investigation and a larger sample size, it is possible that some combination
of fabric VII, VIII, and IX will happen. These seem similar, and may overlap with a larger set.
Fabric XI
This fabric appears only in sherds that were visually sorted to Aztec black-on-orange types
(project types 114, 21, 271, and 182). The fabric is generally extremely fine-grained and wellsorted. The suite of inclusions is not dissimilar to the intermediate to mafic minerals found in
the local Calixtlahuaca material, as would be expected given the generally similar volcanic
conditions within the Basin of Mexico, however, the volume fraction of inclusions is much lower
within fabric XI than is found within Fabric XII or the local Calixtlahuaca material. The most
common inclusion is plagioclase feldspar, with amphibole, quartz, pyroxene, volcanic rock
fragments, and biotite also present. The clay matrix is generally a reddish shade, and shows
evidence of being fired at a high temperature, not dissimilar to Fabric XI. The clays used to make
these sherds was highly processed, likely using levigation or some other method of extracting
the largest inclusions.
Fabric XII
The most characteristic feature of Fabric XII is a consistently high temperature firing. These
sherds all have a deep red, non-optically active core. The inclusions are generally consistent with
the Calixtlahuaca material, with plagioclase feldspar, amphibole, and quartz being the most
common, with lesser amounts of volcanic rock fragments, pyroxene, iron-rich minerals, and very
occasional biotite. The amphiboles are mainly deep red in plane-polarized light, suggesting that
they were exposed to high temperatures, which is consistent with the deep core with no optical
activity.
The major source of variation within the Fabric XII samples is related to the size and texture
distribution of the inclusions. Most are as coarse as the locally produced plain-ware ceramics. At
least one example is finer (ACH-392) with a lower volume fraction of inclusions. It also displays a
bimodal distribution in the size of the inclusions, suggesting that it was processed to remove the
larger inclusions and then had additional sand added as temper. The chemical consistency in the
fabric may be related to the high temperature firing, as the amount of clay processing that
occurred during the production of each vessel varies.
Fabric XIII
593
Fabric XIII seems consistent with a Basin of Mexico origin, and is generally well sorted and fine
to medium in texture. The clay matrix is slightly optically active, and most have a dark core,
which is possibly linked to organic inclusions. The inclusion mineralogy is similar to other sherds
analyzed, with quartz, feldspar, amphibole, volcanic rock fragments, and iron-rich minerals all
present. There are also elongated voids present in the center of several of the sherds that are
consistent with the use of organic temper. This fabric is similar in processing to the fabric XII
sherds, but seems to have been made from a different clay with less iron content. TSN 76 is
placed in this fabric, despite a lack of visible organic burnout voids.
594
APPENDIX E
LITHIC TECHNOLOGICAL CLASSIFICATION CODES
595
This appendix includes a list of type codes used by Dr. Bradford Andrews to
classify lithics excavated at Calixtlahuaca.
Table E.1 Meanings of codes used to group lithic classification codes for analysis
Code
Technology
BC
FC
BIF
BP
U
Lap
Meaning
Blade-Core
Flake-Core
Bifacial
Bipolar
Unidentified
Lapidary
Production/Consumption (pro/con)
cs
core shaping
ips
initial pressure series
ps
prismatic
ps
prodcution
t
tool
crjuv
core rejuviation
cr
core recycling
Biface
5
6
t
Percussion
Pressure
finished tool
596
Table E.2 Lithic type codes used at Calixtlahuaca, with grouping codes used for analysis.
Artifact Code
MF
1UCB-DS
MB-gd
MB-ms
MB-ds
MB-ds-C
MB-f
SPPS-sf
Description
Macro flake
Unilateral crested blade-distal section
Macroblade, proximal section with
ground platform
Macroblade medial section
Macroblade distal section
Macroblade, distal section w/cortex
Macroblade fragment
Small percussion, proximal section, single
facet platform
Technology
BC
BC
BC
Pro/Con
Code
cs
cs
cs
BC
BC
BC
BC
BC
cs
cs
cs
cs
cs
SPPS-gd
Small percussion, proximal section,
ground platform
BC
cs
SPPS-gd-C
Small percussion, proximal section,
ground platform w/cortex
BC
cs
SPPS-cr
Small percussion, proximal section,
crushed platform
Small percussion, proximal section,
crushed plat w/cortex
BC
cs
BC
cs
Small percussion blade, medial section
Small percussion, medial section w/hinge
Small percussion, medial section
w/cortex
Small percussion, distal section
Small percussion, distal section w/ cortex
Small percussion blade fragment
Platform overhang removal flake,
percussion
Small percussion flake, nacelle flake
Plunging blade distal section, percussion
Macroblade scraper
Macroblade end scraper
Small percussion, distal section, end
modified
Small percussion, proximal section,
scraper
BC
BC
BC
cs
cs
cs
BC
BC
BC
BC
cs
cs
cs
cs
BC
BC
BC
BC
BC
cs
cs
t
t
t
BC
t
SPPS-cr-C
SPMS
SPMS-h
SPMS-C
SPDS
SPDS-C
SPB-f
PORF-perc
SPNacelle
DS-pl-perc
MB-scr
MB-es
SPDS-em
SPPS-gd-scr
597
Biface
Code
TPB-gd
TPS-gd
Triangular prismatic blade, ground
platform
Triangular prismatic proximal section,
ground platform
BC
cs
BC
cs
Artifact Code
1MS
Description
1st series prismatic blade, medial section
Technology
BC
Pro/Con
Code
ips
1PB-gd
1st series prismatic blade, ground
platform
1st series prismatic blade, single facet
platform
1st series prismatic blade, proximal
section, ground platform
2nd series prismatic blade, medial
section
Prismatic blade complete
Prismatic blade, proximal section, cortical
platform
BC
ips
BC
ips
BC
ips
BC
ips
BC
BC
ps
ps
PS-sf
Prismatic blade, proximal section, single
facet platform
BC
ps
PS-gd
Prismatic blade, proximal section, ground
platform
BC
ps
PS-gd-C
Prismatic blade, proximal section, ground
plat w/cortex
BC
ps
PS-cr
Prismatic blade, proximal section,
crushed platform
BC
ps
PS-mf
Prismatic blade, proximal section, multi
facet platform
BC
ps
MS
MS-C
DS
DS-C
DS-pl
Prismatic blade medial section
Prismatic blade medial section, w/cortex
Distal prismatic blade section
Distal prismatic blade section, w/cortex
Distal prismatic blade section, plunging
blade
BC
BC
BC
BC
BC
ps
ps
ps
ps
ps
DS-pl-CT-gd
Distal prismatic blade section, plunging
blade w/distal ground core top
BC
ps
BF
Prismatic blade fragment
BC
ps
1PB-sf
1PS-gd
2MS
PB-c
PS-cor
598
Biface
Code
DRS-sf
Pres-B
BNF
Distal reverse prismatic blade section
w/distal single facet platform
Pressure bifacial thinning flake, thinning
of prismatic blade
Flake made from notching a prismatic
blade section
BC
ps
BC
ps
BC
ps
Technology
BC
Pro/Con
Code
p
Prismatic blade proximal section,
ground plat w/dorsal "j" hinge
removal scar
Prismatic blade proximal section,
single facet plat w/dorsal "hb" hinge
removal scar
BC
p
BC
p
Prismatic blade medial section with
hinge termination
Prismatic blade medial section
w/dorsal "hb" hinge removal scar
Pressure derived nacelle flake (flex
tablet, created during blade
removal)
Platform overhang removal flake,
pressure
BC
p
BC
p
BC
p
BC
p
PS-gd-SM-EM Proximal prismatic section, ground
platform, side & end modified
BC
t
PT
MNB
Pointed top prismatic blade section
Multi-notched prismatic blade
section
Diagonal tipped prismatic blade
section
End-modified prismatic blade
section
End-modified and side modified
prismatic blade section
BC
BC
t
t
BC
t
BC
t
BC
t
Needle-tipped prismatic blade
section
Notched prismatic blade section,
complete
BC
t
BC
t
Notched prismatic blade section
BC
t
Artifact Code
PS-gd-h
Description
Prismatic blade proximal section,
ground plat w/hinge term
PS-gd-j
PS-sf-hb
MS-h
MS-hb
Nacelle
PORF
DT
EM
EM-SM
NT
NB-c
NB-f
599
Biface
Code
Haft-B
Haft-B-sn
LP-f
LP-sn
fragment
Hafted prismatic blade section
Hafted prismatic blade section,
notched on its sides
Large point prismatic blade section,
fragment
Large prismatic blade section, sidenotched
Artifact Code Description
LP-sn-cnv
Large prismatic blade section, sidenotched with a convex base
BC
BC
t
t
BC
t
BC
t
Technology
BC
Pro/Con
Code
t
LP-st
Large prismatic blade section,
stemmed base
BC
t
SP-bn
Small prismatic blade section,
basally notched
BC
t
SP-bs-cnv
SP-bs-sn
SP-cnv
SP-sn
SP-sn-cnv
SP-st
SP-f
X-cres
X-cres-f
X-multi
X-tri
X-tri-f
WBF
SBF
CT-gd-bp
Small prismatic blade section
Small prismatic blade section
Small prismatic blade section
Small prismatic blade section
Small prismatic blade section
Small prismatic blade section
Small prismatic blade section
Prismatic blade section
Prismatic blade section
Prismatic blade section
Prismatic blade section
Prismatic blade section
Worked prismatic blade fragment
Snapped prismatic blade fragment
Ground prismatic core top,
bipolared
Recycled prismatic core
Recycled prismatic core, scraper
Prismatic core section flake
Prismatic core distal tip
Prismatic core split platform flake
Prismatic core platform preparation
flake
Prismatic core interior platform
preparation flake
BC
BC
BC
BC
BC
BC
BC
BC
BC
BC
BC
BC
BC
BC
BC
t
t
t
t
t
t
t
t
t
t
t
t
t
t
crjuv
BC
BC
BC
BC
BC
BC
cr
t
crjuv
crjuv
cr
crjuv
BC
crjuv
CR
CR-scr
CSF
CDT
SPF-f
PPF
PPF-int
600
Biface
Code
DOF-j
DOF-pres
DOF-scr
PF
PF-gd
Artifact Code
PD
SD
IC
IS
FC
FC-scr
SDBT
EPBT
EPBT-B
EPBT-det
BT-alt
BT-alt-C
BT-bulb
BT-ed
Prismatic core distal orientation
flake w/j flake hinge removal scar
Prismatic core distal orientation
flake, removed with pres
Prismatic core distal orientation
flake
Prismatic core single facet platform
flake
Prismatic core, ground platform
flake
BC
crjuv
BC
crjuv
BC
t
BC
cr
BC
crjuv
Description
Technology
Primary decortication flake
FC
Secondary decortication flake
FC
Interior core flake w/cortical
FC
platform
I core flake w/ single facet platform
FC
Flake core
FC
Flake core used as scraping tool
FC
Secondary flake w/characteristics of
BIF
perc biface thinning flake
Early percussion biface thinning
BIF
flake
Early percussion biface thinning
BIF
flake, from thinning a prismatic
blade
Early percussion biface thinning
BIF
flake w/remnant detachment scar
Percussion biface thinning, alternate
BIF
flake
Percussion biface thinning
BIF
Percussion biface thinning, bulb
BIF
removal flake
Percussion biface thinning, edge
BIF
preparation flake
Pro/Con
Code
p
p
p
Biface
Code
p
p
t
p
5
p
5
p
5
p
5
p
5
p
p
5
5
p
5
BT-ed-C
Percussion biface thinning, edge
preparation flake - remnant cortex
BIF
p
5
BT-mar
Percussion biface thinning, margin
removal flake
BIF
p
5
LPBT
Late percussion biface thinning flake
BIF
p
5
601
LPBT-det
EPres
EPres-det
LPres
NF
BIF-bs
BIF-bs-sb
BIF-bs-sn
Late percussion biface thinning,
remnant detachment scar
Early pressure biface thinning flake
Early pressure biface thinning flake,
remnant detachment scar
Late pressure biface thinning flake
Pressure notch flake
Biface base fragment
Biface base, straight base
Biface base, side notched
602
BIF
p
5
BIF
BIF
p
p
6
6
BIF
BIF
BIF
BIF
BIF
p
p
t
t
t
6
6
t
t
t
Description
Biface base, convex base
Biface, discoid scraper
Biface fragment
Biface mid-section
Biface tip
Biface tip w/cortex
Biface, straight base
Biface scraper
Biface scraper w/cortex
Biface, side-notched
Biface, side-notched & basally
notched
Biface, stemmed base
Small point, flake blank, corner
notched
Small point, flake blank, side
notched
Uniface
Uniface w/cortex
Unifacial discoid scraper
Uniface discoid scraper w/cortex
Uniface-fragment
Uniface scraper
Uniface side scraper
Uniface side scraper w/cortex
Biface eccentric fragment
Early percussion biface thinning
flake, scraper
Recycled prismatic blade core,
bipolared
Small percussion blade medial
section, bipolared
Technology
BIF
BIF
BIF
BIF
BIF
BIF
BIF
BIF
BIF
BIF
BIF
Pro/Con
Code
t
t
t
t
t
t
t
t
t
t
t
Biface
Code
t
t
t
t
t
t
t
t
t
t
t
BIF
BIF
t
t
t
t
BIF
t
t
BIF
BIF
BIF
BIF
BIF
BIF
BIF
BIF
BIF
BIF
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
BP
cr
BP
p
PS-sf-bp
Proximal prismatic blade section,
single facet platform, bipolared
BP
p
PS-gd-bp
Proximal blade prismatic section,
ground platform, bipolared
BP
p
SBF-bp
Snapped prismatic blade fragment,
bipolared
BP
p
MS-bp
Prismatic blade medial section,
bipolared
BP
p
Artifact Code
BIF-cnv
BIF-dsc
BIF-f
BIF-ms
BIF-tip
BIF-tip-C
BIF-sb
BIF-scr
BIF-scr-C
BIF-sn
BIF-sn-bn
BIF-st
SP-FB-cn
SP-FB-sn
UNIF
UNIF-C
UNIF-dsc-C
UNIF-dsc-C
UNIF-f
UNIF-scr
UNIF-ss
UNIF-scr-C
X-f
EPBT-scr
CR- bp
SPMS-bp
603
BIF-bp
BP
Artifact Code
BP-C
BP-ch
BP-ch-C
BP-n
DS-bp
MB-bp
SC-bp
SPB-bp
E
UC
UN
FF
FF-C
SH
SH-C
CH
CHB
WF
WF-C
WF-cir
UN-polish
SD-scr
Biface fragment, bipolared
Bipolared flake
BP
BP
Description
Bipolared flake w/cortex
Bipolared chunk
Bipolared chunk w/cortex
Bipolared flake w/notch
Prismatic blade distal section,
bipolared
Macroblade, bipolared
Bipolared scalar core
Small percussion blade section,
bipolared
Eraillure flake
Unidentified flake w/cortex
Unidentified flake
Flake fragment
Flake fragment w/cortex
Shatter
Shatter w/cortex
Chunk of toolstone
Battered chunk of toolstone
Worked fragment
Worked fragment with cortex
Worked fragment - circular piece
Unidentified flake w/polish
Secondary decortication flake, made
into a scraper
604
t
p
t
Technology
BP
BP
BP
BP
BP
Pro/Con
Code
p
p
p
t
p
BP
BP
BP
p
p
p
U
U
U
U
U
U
U
U
U
U
U
U
Lap
BIF
p
p
p
p
p
p
p
p
p
t
t
t
l
t
Biface
Code