Citation
Circles of glass and grain

Material Information

Title:
Circles of glass and grain economic differences between core and semi-peripheral zones, a study of public center lithics from the Tequila Valleys of west Mexico
Creator:
Wagner, John P. ( author )
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file (295 pages). : ;

Thesis/Dissertation Information

Degree:
Master's ( Master of arts)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Anthropology, CU Denver
Degree Disciplines:
Anthropology

Subjects

Subjects / Keywords:
Stone implements -- Mexico ( lcsh )
Indians of Mexico -- Antiquities ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
The dynamics of expanding polities and relationships between cultural core groups, peripheral populations and sites in semi-peripheral areas between the two groups are important topics in studies of complex societies. One area where these distinctions are clearly identified within the settlement pattern formed by the relationship between the cultural and the natural landscape is the Tequila Valleys of Western Mexico. The Teuchitlan culture of the Late Formative and Early Classic periods formed distinctive settlements around the edges of the valleys, which were also marginally bound to most complex social developments within the cultural core region near the center of the valleys. Semi-peripheral sites between cultural traditions are of particular interest as focal points for economic, political, and social relationships. This thesis focuses on two sites which occupied very different environments, namely Llano Grande and Las Navajas. I ask whether these sites show different degrees of emphasis on two basic economic strategies in ways which capitalized on the advantages of each site's respective environment. Specifically, did Llano Grande's relative physical isolation from the cultural core area, more distant location and differences in available resources reflect a greater reliance on trade via exported obsidian? Alternatively, did Navajas' closer relation to the core allow a continuance of the core's degree of emphasis on the staple-oriented economy, with less emphasis on obsidian production and rade than Llano Grande? This thesis draws upon the work of Earle (1991) to structure the analysis, particularly his contrast between wealth and staple finance. Past research is reviewed to develop expectations for ach model, which are tested using the analysis of obsidian debitage and products within the ritual centers of each site.
Thesis:
Thesis (M.A.)--University of Colorado Denver. Anthropology
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
Department of Anthropology
Statement of Responsibility:
by John P. Wagner.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
913747296 ( OCLC )
ocn913747296

Downloads

This item has the following downloads:


Full Text
CIRCLES OF GLASS AND GRAIN:
ECONOMIC DIFFERENCES BETWEEN CORE AND SEMI-PERIPHERAL ZONES, A STUDY OF PUBLIC CENTER LITHICS FROM THE TEQUILA VALLEYS OF
WEST MEXICO.
By
JOHN P. WAGNER
B.A.S. Lake Superior State University, 1989
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Masters of Arts Anthropology
2014


This thesis for the Masters of Arts degree by John P. Wagner has been approved for the Anthropology Program by
Christopher Beekman, Chair Julien Riel-Salvatore Tammy Stone
December 5, 2014
11


Wagner, John P. (MA, Anthropology)
Circles of Glass and Grain: Economic Differences Between Core and Semi-Peripheral Zones, a Study of Public Center Lithics from the Tequila Valleys of West Mexico
Thesis directed by Associate Professor Christopher Beekman
ABSTRACT
The dynamics of expanding polities and relationships between cultural core groups, peripheral populations and sites in semi-peripheral areas between the two groups are important topics in studies of complex societies. One area where these distinctions are clearly identified within the settlement pattern formed by the relationship between the cultural and the natural landscape is the Tequila Valleys of Western Mexico. The Teuchitlan culture of the Late Formative and Early Classic periods formed distinctive settlements around the edges of the valleys, which were also marginally bound to most complex social developments within the cultural core region near the center of the valleys. Semi-peripheral sites between cultural traditions are of particular interest as focal points for economic, political, and social relationships. This thesis focuses on two sites which occupied very different environments, namely Llano Grande and Las Navajas. I ask whether these sites show different degrees of emphasis on two basic economic strategies in ways which capitalized on the advantages of each site's respective environment. Specifically, did Llano Grande's relative physical isolation from the cultural core area, more distant location and differences in available resources reflect a greater reliance on trade via exported obsidian? Alternatively, did Navajas' closer relation to the core allow a continuance of the core's degree of emphasis on the staple-oriented economy, with less emphasis on obsidian production and trade than Llano Grande? This thesis draws upon the work of Earle (1991) to structure the analysis, particularly his
m


contrast between wealth and staple finance. Past research is reviewed to develop expectations for each model, which are tested using the analysis of obsidian debitage and products within the ritual centers of each site.
The form and content of this abstract are approved. I recommend its publication.
Approved: Christopher Beekman
IV


ACKNOWLEDGEMENTS
This thesis required much time and effort, and I am very grateful for the support, patience and persistence of the faculty and everyone else involved in all aspects of its creation. I would first like to thank my advisor, Dr. Christopher Beekman for the opportunity to work in his research area, as well as his tremendous attention to detail, added information, reviews and advice which have been invaluable in the process of refining the composition of this thesis. I also wish to thank Dr. Julien Riel-Salvatore for his advice on the lithic analysis, and in particular his input on research design for datasets which have proven quite challenging to compare. I want to thank Dr. Tammy Stone as well for her advice on the statistical portions of analysis, her overall advice and perspective on the purpose and focus of the thesis.
Some colleagues have also been tremendous assets, and I especially want to thank Lucas Hoedl, Catherine Johns and Nichole Abbott for their sharing of information and companionship over two lab seasons in Mexico, and also Tony DeLuca who also shared valuable new information on the latest surveys in the region of study. I want to also thank Dr. Verenice Heredia of the Colegio de Michoacan for providing the use of lab facilities, as well as taking care of the logistics for extended analysis during the 2012 lab season. Finally, I wish to extend thanks to Camilo Mireles for his insight into some of the more curious artifacts in the area of study for this thesis.
Several people outside of the academic setting have strongly encouraged me to pursue this very different path from my technical career background, and a few have encouraged me through the process of writing this thesis. I wish to thank my siblings, Deborah
v


Hentschel and Kristopher Wagner, for encouraging me to see this process through to completion. I would also like to extend thanks to two additional co-laborers in the thesis writing process who have also been sources of encouragement: my niece "in-law" Andrea Hentschel and Kathleen Chambers. I would also like to thank my good friend and natural sciences professional Mark Reichel for his interest, occasional inquiries, informative insight, discussions and further encouragement. Finally, to my parents and all others in my circle whom I have not already mentioned, thank you for your support and patience with my work on this thesis, which has absorbed so much of my time and energy over the last few years.
vi


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION..........................................................1
The Tequila Valleys and the Teuchitlan Culture........................1
Overview of Chapters..................................................8
II. THEORETICAL DIRECTION................................................12
Core-Periphery Relations.............................................12
Outposts and Remote Groups........................................16
Boundaries........................................................21
Unitary States and Segmentary States..............................23
Political Strategies.................................................25
Wealth vs. Staple Economies.......................................25
Wealth Relations and Boundary Permeability........................30
Network and Corporate Strategies..................................33
Connecting Economy and Strategy...................................35
Institutional Competition.........................................35
Combining Economies: Context and Matrix Control...................38
Structure, Economy and the Natural Environment.......................40
Examining Semi-periphery Economies...................................43
III. REGIONAL BACKGROUND..................................................45
The Natural Environment and Subsistence..............................46
Geology and Climate...............................................46
Geographic Area Relations to Political-Economic Zones.............47
vii


Subsistence Resources
50
Minerals.............................................................52
The Economic Impact of a Changing Climate............................52
Economic Strategies within the Teuchitlan Culture.......................53
Staple Finance.......................................................54
Wealth Finance.......................................................60
Economies and Ritual Space...........................................65
Timeline................................................................66
Distribution of Architecture............................................67
Core Sites...........................................................68
Semi-Peripheral Sites................................................69
Peripheral Sites.....................................................72
Descriptions of Compared Sites..........................................75
Los Guachimontones...................................................76
Navajas..............................................................79
Llano Grande.........................................................83
Sayula Basin.........................................................92
Conclusion...........................................................94
IV. LITHICS BACKGROUND.....................................................98
Overview of Mesoamerican Forms..........................................99
Polyhedral Core Blades..............................................102
Eccentrics..........................................................105
viii


Overview of West Mexican forms
105
Pikes and Rod Ornaments..............................................107
Laminar Flake Jewelry................................................Ill
Distribution...........................................................116
Workshops and Points of Exchange.....................................118
Challenges to Interpreting Inter-polity Exchange.....................119
Exchange in Mesoamerican contexts....................................120
Teuchitlan: Internal Distribution....................................138
Area and Site Studies..................................................147
Los Guachimontones...................................................149
The La Venta Corridor................................................155
The Sayula Basin.....................................................158
Inter-Area Comparisons...............................................161
Conclusion.............................................................165
V. METHODS.................................................................168
The Llano Grande and Navajas Datasets..................................168
Attributes...........................................................172
General Approach for Analysis..........................................181
Criteria for Comparing Staple and Wealth Lithic Economies..............182
The Nature of the Deposits...........................................182
Expectations and Tests for Economic Distinctions.....................185
Import Considerations................................................199
Outcome Analysis.....................................................200
ix


Conclusion
201
VI. RESULTS.................................................................202
Organization and Review of the Test Criteria...........................202
Lithic Type Proportion Comparison to Background Contexts...............204
General observations on the Collections Under Test.....................207
Overall Quantities and Site Coverage Differences......................208
The Nature of the Deposits..............................................210
Lithic Industry Tests.............................,....................212
Test 1: Formal Products vs. Debitage and Expedient Tools.............212
Test 2: Debitage to Ceramic Sherd Ratios..............................213
Test 3: Core Totals...................................................215
Test 4: Amounts of Dorsal and Platform Cortex on Flakes..............216
Test 5: The Percentage of Specialized Tools for Staple Processing....220
Test 6: The Percentage of Debitage Associated with Polyhedral Blades.221
Test 7: The Percentage of Unfinished or Failed Bifaces................222
Test 8: Emphasis on Production Related to Laminar Flakes.............223
VII. DISCUSSION & CONCLUSION.................................................229
The Origin and Purpose of Guachimonton Obsidian.........................233
Conclusion..............................................................235
Challenges, Limitations and Lessons Learned.............................241
Future Opportunities....................................................243
Continued Exploration at Llano Grande.................................244
Surveys and Additional Excavations....................................245
x


New Questions with Expanding Exploration.........................246
Other Paradigms..................................................247
REFERENCES..............................................................249
APPENDIX
A. Carta Land Use and Soil Map sections..............................269
B. Scatterplots of scar counts and size measurements.................278
xi


LIST OF TABLES
Table
3.1: Site occupation spans and related periods relevant to the current study....68
4.1: Platform preparation percentages, center vs. workshop.......................151
4.2: Lithics from the Los Guachimontones public center...........................152
4.3: Lithics from the Feature 83 workshop, with used item percentages............154
4.4: Lithics from the La Venta Corridor (nodules excluded).......................157
4.5: Percentages of each lithic type per location................................162
5.1: Llano Grande size frequency distribution comparison for different flake types ....178
5.1: Navajas size frequency distribution comparison for different flake types....178
6.1: Summary of compared background locations....................................204
6.2: Intersite comparison of lithic proportions by percentage relative to total item counts ............................................................................204
6.3: Percentages of formal product counts relative to total analyzed sample......213
6.4: Debitage and Sherd Total Comparison.........................................213
6.5: Core quantities as a percentage of debitage at each site....................215
6.6: Site percentages of flakes and blades which possess full or no cortex......216
6.7: Means and standard deviations for flakes and blades with partial cortex.....217
6.8: Percentage Comparisons of Blade Types by Count..............................221
6.9: Bifacial Blank Percentages by Count.........................................222
6.10: Correlations of flake scar density with dimension/size variables for all flake types ............................................................................225
6.11: Reduction flake correlations of flake scar density with dimension / size variables
...........................................................................225
xii


6.12: Shaping/finishing flake correlations of flake scar density with dim /size variables ...................................................................................226
xm


LIST OF FIGURES
Figure
1.1 Map of West Mexico............................................................2
1.2 Map of the Tequila Valleys showing core and semi-periphery...................3
1.3 The Teuchitlan Periphery (West Mexico) relative to the core and semi-periphery.4
2.1 A scaled representation of different forms of exchange......................19
3.1 Circle 2 guachimonton from the Los Guachimontones site......................55
3.2 Navajas Circle 5 guachimonton plan view.....................................56
3.3 Profile of a shaft tomb from the Atitlan Las Cuevas site....................61
3.4 Navajas central group with areas of excavation circled......................81
3.5 Site layout of Llano Grande.................................................88
3.6 Llano Grande central group, with 5 meter increment contours.................90
3.7 Plan of the Llano Grande guachimonton (with excavation units)...............91
4.1 Profile of a rhyolite dome from a lava flow, showing the obsidian layer and spines .............................................................................100
4.2 Polyhedral blade production sequence.......................................103
4.3 Prismatic blade production sequence defined................................104
4.4 Pike collection............................................................107
4.5 Rough pike from Navajas Circle 5, lateral and cross-sectional views........109
4.6 Refined pike fragment from Navajas Circle 5, lateral and cross-sectional views ....109
4.7 Anthropomorphic laminar jewelry from Navajas...............................112
4.8 Major areas and sites studied within the Tequila Valleys...................140
4.9 Tequila Valley obsidian source locations mentioned by Spence, et al........141
5.1 Llano Grande plan, with excavation units....................................169
xiv


5.2 Navajas plan, with excavation units..............................................170
5.3 Reduction flake..................................................................175
5.4 Small shaping/finishing flake....................................................175
5.5 Large shaping/finishing flake....................................................175
6.1 Llano Grande flake size distribution.............................................211
6.2 Navajas flake size distribution..................................................211
6.3 Flake and blade comparisons of cortex coverage distributions per site...........219
6.4 Core comparisons of Cortex Coverage Distributions at Llano Grande............219
xv


CHAPTERI
INTRODUCTION
The Tequila Valleys and the Teuchitlan Culture
The Tequila Valleys area of West Mexico is a portion of a chain of lake basins at the base of the Tequila Volcano, which form an open area within the forested hills of the Sierra Madre Occidental mountain range that runs parallel to the Pacific Coast (Beekman 2010:43-44, 63; Fig. 1.1). During the Mesoamerican Late Formative and Early Classic periods (ca. 300 B.C. 500 A.D), the Magdalena Basin on west side of the valleys contained a large lake, and the remaining basins were marshlands. The areas immediately surrounding the lake were interspersed with steep hills and valleys of marshes and fertile alluvial soils, which formed a secluded setting for the development of a complex culture known as Teuchitlan.
Aside from a general attraction to clay figurines and dioramas that have been quite popular with collectors since the turn of the 20th century (Townsend 1998a: 17-22; Weigand 1975:186, 223, 1985:47-54; cf. Breton 1903, 1905; Hrdlicka 1903:393; Lumholtz 1903), the region has generally been marginalized by earlier archaeologists and explorers as an area populated only with nomadic groups and simple chiefdoms (Beekman 1996b: 136; cf. Fernandez and Deraga 1988; Schondube 1998). However, more recent investigations have better recognized the area's true cultural significance. Around 1970, investigators began to recognize a pattern of ceremonial architecture that points to
1


a much larger and more complex society than previously assumed (Kelley 1974; Ohnersorgen and Varien 1996:103; Weigand and Beekman 1998). Today, the culture is
Figure 1.1: Map of West Mexico. Numbers indicate lake basins. Of particular interest to this thesis are the two main basins within the Tequila Valleys: Magdalena (2), and La Vega (3). Also of interest are more distantly associated sites in Sayula (7). (Beekman 2010:42).
known to a small but growing number of investigators as one unique to Mesoamerica with a social structure that has been described as something akin to either a complex chiefdom, or an early state (Beekman 1996a,b; Schondube 1998; Weigand and Beekman 1998). One area for potential research is site specialization. Although the majority of sites associated with the Teuchitlan culture are found in a core population zone south of the
2


Tequila Volcano (figure 1.2), others are scattered throughout the rest of the Tequila Valleys. Six of the sites outside the core zone, namely Estolanos Mesa, Mesa El
Llano Grande
Volcan de Tequila
Guachimontones
Core
population w zone
ivajas
18 Kilometers
Tequila Valleys, Jalisco $
A
$ B Valencia (iV: V
C lake basin
D
Figure 1.2: Map of the Tequila Valleys showing core and semi-periphery. Circled sites are within the core, and all sites outside the core population zone are considered semi-peripheral. Labeled sites are those under study. Navajas' location in a small exterior basin associates it more closely with the core zone. Adapted from Beekman in press a:Fig. 3.2.
Zacate, Cerro Tepopote, Penol Tepopote, Cerro Pipiole and Llano Grande, were built within or near passes in the surrounding hills. The exact purpose and nature of these surrounding sites has been a major focus of study (Beekman 1996a,b; Weigand and Beekman 1998). Within this paper, all sites within the Tequila Valleys but outside of the core zone are referred to as semi-peripheral sites because of their location between the
3


densely populated core habitation zones and populations outside the Tequila Valleys. Peripheral sites are outside the Tequila Valleys, yet within West Mexico (Figure 1.3).
Figure 1.3: The Teuchitlan periphery (West Mexico) relative to the core and semi-periphery.
The line of questioning pursued for this thesis is a continuation and extension of these core-periphery discussions, and utilizes evidence found in local obsidian production for additional interpretation of their economic dimensions. Substantial evidence from archaeological, ethnographic and ethno-historic sources shows that elites of West Mexico relied heavily on agricultural production to finance their positions in the social structure (Beekman 2003a; Butterwick 1998; Lopez and Ramos 2006a; Lumholtz 1903;
Schondube 1998; among several others). Contrastingly, the position of Las Navajas, a
4


southeastern semi-peripheral site on a high plateau with an expanse of fertile farmland, provided ample opportunity to continue the agricultural tradition of excess production needed for staple finance (Beekman 2007a; Appendix A:Figures 4-6). In contrast, the micro-environment of Llano Grande's chosen location, in a northwestern pass overlooking the Magdalena lake basin, was not conducive to agricultural production of any sort (Beekman 2001; Appendix A:Figures 1-3). It did, however, provide immediate access to an adjacent high quality obsidian source1 and much greater potential exposure to external groups. Did Llano Grande operate under a different economic strategy relative to the core population zones of the Teuchitlan culture? And, given the presence of obsidian, did Llano Grande represent a relative shift within the Teuchitlan culture from a surplus producing, internally competitive agrarian economy to one based on trade?
Studies and discussions of semi-peripheral site functions in the Tequila Valleys have thus far concentrated primarily on architecture and area wide surface surveys with minimal excavation (cf. Beekman 1996 a,b; Weigand and Beekman 1998), and therefore could not yet address some of the economic dimensions of site function. However, more recently a large sample area has been excavated in the Llano Grande ceremonial space in the hills just outside the valleys to the northwest, as well as the entirety of one other example of public architecture at the site of Navajas to the southeast. Ceramic and obsidian artifacts recovered from both sites have been the subjects of study over the last few years (Hoedl 2013; Johns 2014).
For the current study, the above questions are addressed through a comparative analysis of the lithic assemblages recovered from the excavated ritual centers within 1
1 The Llano Grande and Navajas obsidian sources mentioned in this thesis refer to those near the sites within the Tequila Valleys by the same name, and should not be confused with the more northern Llano Grande and Cerro de las Navajas lithic sources in Durango (cf. Darling 1993).
5


Llano Grande and Navajas. Wealth and staple based economic patterns discussed by Timothy Earle (1991) and related political strategies proposed by Richard Blanton and colleagues (1996) are used to frame the geographic and environmental implications of each site's economic potential in order to determine general economic expectations for the sites under test. Previously published lithic distribution analyses both within the valleys and in additional areas within Mesoamerica are used here to determine common lithic distribution patterns which relate to normative distribution strategies, and further refine specific expectations for each site's lithic industry, in relation to Earle's economic categories. Site contexts are discussed by local area, and include the primary core area near Los Guachimontones (Esparza 2003; Soto 1982, 1990) as well as three semiperipheral and peripheral areas. The latter areas include the La Venta Corridor (Beekman 1996a, 1996b), the Magdalena Basin (Spence et al. 2002) and the Sayula Basin (Reveles 2005). Because of the high volume and spatial breadth of comparative data, distribution patterns of outside groups necessarily include a limited number of studies that have the best potential to identify common patterns which may have influenced semi-peripheral groups. External groups discussed include Teotihuacan (Spence 1981, 1987), the Maya (focusing primarily on Nohmul and Colha, as discussed by Johnson [1996]), and earlier groups in the Valley of Oaxaca (De Leon et al. 2009).
The specific research questions addressed in this thesis are as follows: 1
1. Do the lithic assemblages at Llano Grande show a greater emphasis on production over use contexts than the assemblages at Navajas and other sites more closely affiliated with the core population zone?
6


2. Does Llano Grande show evidence for a greater emphasis on the production of potential trade than Navajas, including commonly exported forms, known elite items, and/or items which require specialized production techniques?
A positive answer to the first question supports the notion of a specialized area of production that may be defined as a workshop (Clark 1981; Spence 1981) at Llano Grande, either within or nearby the public architecture, which also suggests a possibility of systematic overproduction for some form of exchange. However, production could still be used for the general trade of common goods, or of minimally prepared raw materials.
The second question deals strictly with the forms of items produced, and a positive answer to the question may show increased production of potential ritual or trade items at Llano Grande. However, based on the second question alone, the use of these items is ambiguous in that they may be intended for elite trade, or strictly for internal use.
Positive answers to both questions would support an increased emphasis on both overproduction for trade and the development of wealth items in the same context, and therefore also an increase in the likelihood of a greater emphasis on the wealth trade economy.
The null hypothesis for this study is a lack of support for an increase in wealth trade at Llano Grande over Navajas. Therefore both questions must be positively answered for the null hypothesis to be rejected.
7


Lithicist George O'Dell's lament about his own specialty eighteen years ago is still essentially true today: "Although advances in lithic studies have been made in both the theoretical and methodological realms, yawning gaps exist in their articulation" (1996:4). This thesis is theory driven by design, and will serve as an attempt to bridge the "yawning gap" between the theoretical approach and methods through an explicit description of the relevant theory, the middle-range interpretive rules and assumptions, and the implications of the results on the theoretical foundation.
Overview of Chapters
Chapter II is devoted to describing the social theory which addresses and guides the questions under study. The chapter first covers theoretical discussions on the nature of core-periphery relations, and then turns to economic dichotomies that may explain how core areas may differ from peripheral or isolated semi-peripheral areas in economic strategies: Earle's wealth vs. staple elite financing (1991), and related networking/exclusion vs. corporate/inclusion power strategies (Blanton et al. 1996). The internal dynamics of these strategies as they are pursued in the same cultural groups and engage each other within designated community spaces, or arenas (Turner 1974) is also discussed. The chapter then discusses the reciprocal relationship between culture and environment, and how environmental factors may play a major role in elite decisions to emphasize certain economic strategies over others.
Chapter III covers Tequila Valleys background information. The chapter begins with the history of archaeology in the area and then turns to details of the area's prehistory,
8


which begins with past and current understanding of the timeline focusing on the transition from the Late Formative to the Early Classic periods. The prehistory section first briefly describes the distinctive elements of the culture (guachimontones, shaft tombs, ceramic dioramas/figurines, and unique forms of obsidian), then describes the demographics and other physical properties of core, semi-peripheral and peripheral sites. General descriptions and interpretations of apparent cultural uses of the shaft tombs and the guachimontones are then described in separate sub-sections. The chapter then discusses the current evidence for staple and wealth finance strategies. Feature details of the primary sites under study, Llano Grande and Navajas, are then discussed and compared to the primary core site, Los Guachimontones. The chapter then turns to environmental data, including resource and climate details. General area-wide details are presented before concentrating specifically on geography and climate differences between Llano Grande, Navajas and Los Guachimontones.
Chapter IV discusses the area lithic data, starting with an overview of Mesoamerican lithic forms, including details of West Mexican forms and discussions of proposed production methods which may offer ways to identify the manufacture of items which have been exported off-site from the remaining debitage. Since semi-peripheral area strategies may be as much or more influenced by external groups as they are by the core, lithic distribution strategies, including both products and debitage, three other Mesoamerican groups (Teotihuacan, Oaxaca and the Maya) are addressed before discussing the literature on lithic distribution patterns within the Tequila Valleys. Other relevant lithic attributes at the site and area level are then discussed and compared between Los Guachimontones, the La Venta Corridor and unaffiliated sites within the
9


neighboring Sayula Basin to find any detectible patterns of similarities and differences between core, semi-peripheral and peripheral sites.
Chapter V discusses the methods used to test the hypotheses outlined in this chapter. The chapter begins with a description of the datasets at both Navajas and Llano Grande, including the prior excavation process and associated issues encountered during excavation and artifact retrieval, the method of storage, and later retrieval and recording of the data. Expectations from the data for a wealth trade or an internal use (staple) strategy are then listed and described with the implications of all potential outcomes, and a series of tests used to determine whether these expectations are met are described.
Chapter VI reports on the outcomes of the tests outlined in Chapter V. The implications of each test's outcome on the research questions are also discussed, in relation to the site expectations.
Chapter VII concludes the thesis with a synthesis and summary of all outcomes and provides an overall interpretation of the sites under test in relation to the original questions and the implications that the results have to the theoretical foundations of the tests. A section on future work in pursuit of the same or related questions is also provided.
This project is not designed or intended to completely resolve the economic nature of all semi-peripheral sites associated with the Teuchitlan culture, or even completely resolve the issue specifically for Llano Grande and Navajas, as it is more of a theoretically guided probing study involving only two sites in detail, and general information from a few others. It is anticipated that, when employed on a much larger
10


scale, this form of theoretically informed technical analysis will bring West Mexico lithic research much closer to resolving these kinds of questions.
11


CHAPTER II
THEORETICAL DIRECTION
At the most fundamental level, this thesis seeks to determine whether economic differences can be seen in the material culture between sites associated with the same polity, but at different locations and positions relative to the core area, particularly when the chosen locations are within very different environments; and whether these economic differences represent differing emphases on wealth and staple economies (Earle 1991). Following Hirth (1996), economic differences between sites are seen here as relative differences in emphasis on multiple specific economies that may cross-cut staple and wealth categories, rather than a predictable dominance of one economic category over the other at the site level as originally proposed by Earle, and also Blanton et al. (1996). This study further tests whether observable elite choices in emphasis on different economies are influenced by the immediate area's economic resource options, as well as the distance and access to other internal and external groups.
Core-Periphery Relations
Differences in resource availability within different areas relate to patterns of settlement and social relationships originally described by Wallerstein's (1974) world-systems theory. Wallerstein describes a core group which is comprised of concentrated areas of consumption, and a periphery which contains the resources necessary for maintenance of the core's economy. In-between the core and periphery are semi-
12


periphery groups, which interact with both the core and periphery and create relations of exchange between the two areas (Wallerstein 1974:401-405). Wallerstein applied his theory exclusively to industrial capitalist societies with the technological capability to extend their reach globally and create dependent and unequal exchange relationships with less powerful groups, for the acquisition of agricultural and industrial bulk goods (1974:398-399). These relationships create a single system of labor division which cuts across cultural groups and polities (1974:390). The primary feature of Wallerstein's theory is the formation of a dependent and dominated periphery in an asymmetrical exchange relationship.
Schneider has suggested that Wallerstein's world system "suffers from too narrow an application of its own theory" in its strict application to capitalist systems (1977:47), and relies too heavily on bulk goods while dismissing luxury items as non-systematic trade items that had little effect on the overall economy (1977:52). Schneider (1977:52-53) contends that several luxuries, such as sugar, wine and precious metals, were also used quite effectively to manipulate peripheral groups before the world system began to form in the mid-seventeenth century. Schneider also referenced archaeologist Robert Adams' discussion of inter-polity social influence of long-distance luxury trade as far back as the earliest complex polities within Mesopotamia. Adams suggested that luxury trade was a "formidable socioeconomic force" (1974:247) which was often tumultuous, and used by the more powerful polities to coerce and dominate less powerful groups within Mesopotamia, given the presence of slave labor and very dynamic changes in exchange relationships revealed by the cuneiform records for luxury item trade (Adams 1974:247-249).
13


Blanton and Feinman (1984:674) suggested that the world system theory may be a more productive framework for the study of polity interaction within Mesoamerica, rather than the common approach to the region at the time, which treats the entire subcontinent as one massive culture area, which is yet multicultural, but held together by common ideologies and technologies that spread through various interpolity relations. These relations were identified through inter-polity commonality of some aspects of material culture, but essentially left unstudied regarding the nature of how and why they form and change over time. Blanton and Feinman do not advocate the application of all details of the theory to prehistoric Mesoamerica, and point out that no polity within Mesoamerican prehistory has ever extended its reach throughout the subcontinent, and some institutions associated with the capitalist system do not appear in prehistoric context. Blanton and Feinman's idea of a Mesoamerican "world system" utilizes Wallerstein's basic framework of cross-cultural labor division in a single multi-polity, cross-cultural system within the Mesoamerican context (Blanton and Feinman 1984:674; Hirth 1996:19). However, Blanton and Feinman assign a role for luxury goods which is as important as the role Wallerstein assigns to bulk goods (1984:675- 676), "through the calculated distribution of symbols of status that the elite controlled" (Blanton and Feinman 1984:676). They also suggested that multiple cores likely existed beyond their example of the Postclassic Central Highlands (areas exploited by the Aztecs), including Late Formative and Classic Period Teotihuacan, but "at present, we know less about them" (1984:679).
The principles of world-systems theory have since been utilized and modified by archaeologists to describe the economics of early prehistoric states, and theorists developed varying adaptations which included different dynamics between the core and
14


periphery. Algaze (1993) closely follows Wallerstein's definition for modern capitalist societies. Algaze interprets peripheral sites as "outposts" intentionally established within areas controlled by external groups with a very singular, intentional purpose of bringing foreign resources into an expanding core area, as the core's resource requirements increase due to its own growth. He further states that "by definition, marked developmental asymmetries always exist between pristine states and communities in the periphery." Informal trade relationships would then be established with a high probability of success, since peripheral groups would, at least initially, perceive some benefit from political relationship with the a more powerful core. However, Algaze describes outposts as "dendritic" entities entirely controlled by delegates of the core, therefore the trade-off to local elites is the cost of any real political control over the relationship. These outposts would then serve as remote collection points of goods from their respective locales (Algaze 1993:304).
Algaze further suggests that these outposts are, by nature, exploitive as he states that "early outposts reflect a system of economic hegemony whereby early emergent states attempted to exploit less complex polities located well beyond the boundaries of their direct political control and this system may be construed as imperialistic in both its extent and nature" (1993:305) .
Gil Stein (1999:36; 2014:55-56, among others) has criticized the view of a strong, dominant core exploiting a week, subaltern periphery in world system theory as oversimplified and monolithic. Stein views the nature of the relationships between groups as quite variable according to individual political, economic, ideological, and logistical factors (such as distance and transportation economics ) within each group. In many
15


cases, according to Stein, peripheral groups are active agents in the direction of trade relationships, and in some cases managed fully interdependent relationships, rather than asymmetrical ones.
Chase-Dunn and Hall (1991:19) also see a wider range of relationships involved in core and periphery relations, and offer a more inclusive definition in their model which distinguishes between core-periphery differentiation, or groups of different sizes and social structures interacting within the same system, and core-periphery hierarchy, the domination of one society over one or more others by means of political, economic and/or ideological control; differentiation does not assume dominance, and can involve favorable exchange for both sides. "Equality" of exchange is also difficult to measure, largely because different groups value the same objects of exchange differently within their own cultural contexts, especially if they utilize exchange items for different purposes (Chase-Dunn and Hall 1991:31). For the purpose of the current study, no initial assumption is made about the specific nature of the relationships between core and peripheral groups, and therefore Chase-Dunn and Hall's more flexible definition of core and peripheral sites are used.
Outposts and Remote Groups
Outpost settlements often occur at or beyond the farthest reaches of direct control, and are often strategically placed on natural transportation routes between the polity and external groups to facilitate both control and efficient transportation, collection and redistribution of traded goods, especially where transport is particularly difficult (Algaze
16


1993:321). Prehistoric Mesoamerica provides an excellent example of a region where transportation would have been especially problematic in many areas, due to a lack of pack animals and very rugged, hilly terrain in many places, especially along the Sierra Madre Occidental and Trans-Continental mountain ranges (Schondube 1998:206).
According to Algaze, collection and redistribution operate most efficiently when handled in outposts somewhere near the outer edge of a group's area of political control, in corridors which provide limited access. Outposts are also frequently placed very close to a desired resource in a remote area. Outposts vary greatly in size from small apparent way stations positioned along long distance trade routes, to large remote city settlements set up on the way to other major polities that provide high volume trade (1993:310-311).
The establishment of peripheral outpost trade is not generally considered an end-point to core expansion. European colonial trade enclaves could be initially benign and even welcome by peripheral groups, but they often became entry points for later expansion of core powers throughout the area. Expanding trade eventually radically transformed cultural landscapes around the world, including the American continent (Curtin 2000:4-5). Although the arrival of the Spanish empire was a chief concern for the natives in Central Mexico, the Spanish trade economy was likely not initially considered a threat compared to other colonial ventures in agriculture, ranching and the mining of gold and copper. However, eventually "trade empires" became the single most important factor driving European expansion as they successfully out-competed all ventures. Enclaves in the centers of the native populations ultimately became the Spanish political centers for the area (Curtin 2000:79).
17


Alternative models of economic redistribution have been proposed which do not emphasize Wallerstein's principle of core dominance. Stein's (2014:59) concept of a trade diaspora is quite similar to descriptions of outposts and enclaves, except that they may experience a wide range of power relationships with local polities. Diasporas can be very dominant as in World Systems models, but may also be marginalized by remote groups, or experience neutral positions of protected autonomy among their hosts if the local elites find them useful. Also, the distribution of goods may be relatively uncontrolled, or only distantly managed within open market systems. Kenneth Hirth's (1998:453) proposed approach to identifying centralized market activity notes several consistent spatial patterns used by some archaeologists as clues for identification, derived from Spanish contact period observations. Large markets were often set up in Mesoamerica along major trade routes and either near or adjacent to "administrative centers," and served as points of collection and redistribution of goods. These observations closely match many aspects of Algaze's definition of large trade-route outposts, and some areas interpreted as "outposts" may have actually operated as centralized marketplaces. Hirth describes a centralized marketplace as one strategically placed for access by multiple groups, which features unrestricted exchange (although likely sponsored and managed by elites), with equal access of all goods to all consumers regardless of social status. Centralized markets tend to be quite large and typically operate on a regional scale (1998:454-455). In contrast, Algaze's interpretation of peripheral exchange is a hegemonic one highly favoring the core polity. The observed ethnohistoric markets were typically regional exchanges set up in distinctive forms of permanent architecture surrounded by walls which also surround buildings directly associated with market administration, but as
18


Hirth (1998:455) cautioned, prehistoric exchanges may have ranged widely in scale from central exchanges much like those in historic accounts, to locations designated for occasional isolated partner trades which do not involve direct architectural evidence of the point of exchange. A few forms of trade also occur in venues in-between these two extremes in scale, including trade activity conducted directly at workshops locations ( Figure 2.1). Smaller scaled exchanges tend to be less centralized, and unequally accessed by certain segments of the population with access to trade networks of specific groups and an unequally high or even exclusive distribution of specific products and resources among network affiliates (Hirth 1998:455).
One likely example of a large outpost marketplace exchange area may be within the 6 km core city area of Matacapan, perhaps at or near the central plaza. Matacapan is situated about half way between Teotihuacan and Mayan territory, yet shows very strong material cultural continuity with Teotihuacan, including Teotihuacan's trademark apartment building-like residential structures (Algaze 1993:304, 310-312).
Centralized Decentralized
Exchanges Exchanges
Nucleated Clustered Dispersed
Interactions Interactions Interactions
Figure 2.1: A scaled representation of different forms of exchange. (Hirth 1998:455)
19


Although Algaze asserted that especially remote and/or isolated outposts must maintain a fragile connection to the core due to transportation and communication limitations (1993:325), Phil Weigand has argued that direct control of outposts is not necessary, since their function and reason for existence is defined by economics and dedication to external trade and interaction with outside economies, rather than political relations ( Weigand 2000:52; Weigand and Beekman 1998:44,45). However, Algaze refers to more distant outposts than Weigand, and the form of core control to which he refers is not political, but rather hegemonic and economic. Elite groups maintain their dominance through economic influence rather than institutionalized positions of power and authority.
Chase-Dunn and Hall have also argued for the use of world system theory on smaller scales in prehistoric contexts, on the basis that the range of interaction is actually part of the whole world of influence known to the system's participants, and is larger than any individual polity within the system. In prehistoric contexts, the authors also hold, along with Schneider (1977) and Blanton and Feinman (1984), that the prehistoric core/periphery connections tend to emphasize prestige goods; bulk exchanges in staple goods generally range over much smaller areas than exchanges in prestige items (Chase-Dunn and Hall 1991:12; Blanton and Feinman 1984:677).
Chase-Dunn and Hall also make use of Wallerstein's semiperiphery category of economic divisions in-between core and periphery sites. The semiperiphery is an economic and cultural intermediate point between the core and periphery, and is considered "both exploited and exploiter" in the economic process. The semi-periphery thus acts as a political buffer zone (and in this case, according to Wallerstein, the function
20


is indeed mainly political rather than economic) between core and peripheral areas, to avoid direct confrontation between starkly polarized regions, and thus promotes system stability (Wallerstein 1974:405). Semiperiphery sites may have hybridized core and peripheral organization strategies, and institutional features may show intermediate forms between those in the core and peripheral locations (Chase-Dunn and Hall 1991:21). Sites in these areas of transition are especially suited to social innovations since they can combine core and peripheral elements to create new cultural elements, and are less bound to the core due to their physical distance (Chase-Dunn and Hall: 1991:31). Blanton, et al. (1996:7) also note that such areas are among the most likely areas to establish trade networks, and also often occur in areas of low agricultural potential.
The Aztecs were well aware of the advantages of semi-peripheries, and chose to transform the Postclassic Central Highlands city of Tepeaca into a properly functioning semi-peripheral group. Tepeaca was a city on an Aztec trade route which at one time discouraged trade by killing and robbing merchants passing through the area. Upon assimilating the city, the Aztecs insisted that Tepeacans cease their usual practice with outsiders and welcome merchants from all areas. They were also instructed to increase their number of tlamemes (burden carriers), and the Aztecs established a marketplace for luxury goods within the city (Blanton and Feinman 1984:678).
Boundaries
Semi-peripheral sites can be difficult to identify in the archaeological record, and we must rely on multiple lines of evidence beyond spatial organization to do so. Core,
21


peripheral and semi-peripheral zones are defined by the division of labor across polities, which are independent of political territories. However, elites utilize inter-polity trade flows for their own political gain through the accumulation of goods for the building of socio-political hierarchy, and through the economic manipulation of groups beyond their direct political control (Hirth 1996:219). The functions of physical boundaries may therefore include attempts to control economic activity in semi-peripheral areas. Of course, boundaries may also represent a number of other functions, including physical defense, service as an observation post, message relay point, symbolic representation or embodiment of the state, administrative center, strategic activity center, and diplomatic center (Southall 1988:56). Also, boundaries are often invisible in archaeological data, although heavily controlled sections of a boundary are sometimes revealed by walls (Chase-Dunn and Hall 1991:18).
Cultural borders are also often ill-defined, and even when they are made explicit, lines drawn on a map in ink as "borders" as though indelible are actually quite transient, elusive entities which may move back and forth frequently with contention over territory. The existence of boundaries also belies the great deal of interaction that tends to occur across them; no society can be assumed to be completely isolated, regardless of the level of complexity (Schortman and Urban 1987:81; Wolf 1982). A boundary's functions can also vary both spatially and over time (Wolf 1982). A boundary may be very porous to economic goods, while the movement of people is closely monitored and heavily guarded (cf. Beekman 1996a:742). Yet at another time, or even concurrently at a different location along the boundary, it may be completely open and any physical manifestation may serve only as a historic symbol of the state. For these reasons, other lines of evidence should be
22


used to confirm the existence of a boundary. Ultimately, although boundaries can be complex, heterogeneous and therefore challenging to interpret, understanding the nature of a boundary means understanding something of the internal structure and strategy of the polity which created it (Southall 1988:56).
Unitary States and Segmentary States
Some early states may not even conceptualize a physical boundary as part of their self-identity. Southall has noted an organizational difference based on the degree of intentional definition of a polity's boundaries, between what he calls unitary and segmentary states. Unitary states are much more formally defined, centralized and territorial by nature, and therefore strive towards "stable and controlled boundaries" as a management tool. Segmentary states maintain their power structures through hegemonic control based on economic power and prestige, rather than formalized institutions of authority (1988:55). Segmentary states may be identified in material culture through secondary site emulation of primary site architecture and symbolism, as an indicator of hegemony. The developmental trajectory of a segmentary state may eventually transform it to a unitary one as control of more remote locations is consolidated under a centralized government, where previously only a hegemonic form of control had existed. Such a transition may not often be easily accomplished, since direct administration of remote sites in territorial polities is much more costly, and has only been completely demonstrated by very powerful, well centralized polities such as the Roman Empire (Beekman 1996b: 136). The limitations of transportation only by foot, hilly terrain, and a
23


very localized economy may make unitary state formation practically impossible in some areas, and control strategies may also vary within the same polity (Southall 1988:63,72).
Mesoamerican polities include a range of forms between unitary and segmentary states. A few of the most notable examples include the Aztecs along the Aztec-Tarascan border, the remainder of the Aztec empire, and the Classic Maya. The Aztecs near the border with their most powerful adversary were highly territorial and centrally controlled, with several fortresses along the border (Beekman 1996b: 136; Beekman and Houston 1993: 3). Yet, throughout the rest of their area of control, the Aztecs paid no attention to territories and only imposed hegemonic control with only a return of tribute from subordinate groups (Smith 1997:76). The Aztecs at the Tarascan border may have been heavily influenced by the Tarascans themselves, since the Tarascan empire appears to have been very highly centralized and organized as territories managed by a hierarchical administration (Pollard 2003:80-82). The Maya gained power through ceremonial and symbolic hegemony to form various city-states spread out through a large portion of the Yucatan Peninsula. Several especially powerful cities such as Tikal did reach some level of centralization as they subordinated several smaller groups under the rulership of "overkings", but there was no overall centralization, spatial organization, or territorial boundary that would ultimately unify the Maya, and although some walls do appear, territories still do not appear to be a defining element of affiliation with the Maya as a whole (Martin and Grube 2000:18-21).
24


Political Strategies
Wealth us. Staple Economies
The same division between luxury and bulk goods discussed above for the economic bases of world systems also has been described by Timothy Earle as a fundamental separation of economies utilized by elites in different groups to finance the maintenance of their social positions (1991). Earle uses this division to characterize economies of entire polities, although he does acknowledge that economies may occur "as admixtures in individual cases" (Earle 1991:3), which suggests that his economic division may occur at some component level within at least some polities. Although more complex views of how different economic bases and their related strategies are utilized have been adopted more recently (e.g. Blanton, et al. 1996; Hirth 1996), the basic definitions of wealth and staple economies remain important elements of more recent views.
Wealth economies involve long distance exchange and/or specialized craft production for symbols of status which elites use to legitimize their social positions, whereas staple economies include mainly food related resources utilized by elites as leverage for power. Staple economies include the hosting of feasts to attract followers and create a level of obligation to the hosts, and possibly other forms of food related payment (Earle 1991:3).
In wealth economies, elites incorporate exotic items from remote areas into group ideology as a means of creating exclusive control over objects that symbolically convey legitimacy of authority. These prestige items are either made from locally rare or unavailable materials, or they are difficult or expensive to produce, which makes them
25


more difficult to access and therefore easier for elites to control and monopolize their distribution (Beekman 2000:386, Earle 1991:7). Such objects are often utilized in ritual contexts to reinforce the elite ideology and validate elite authority (Reilly 1996:30). A common example of a wealth economy is the Classic Maya, who extensively traded and utilized eccentrics, ritual items and rare, distantly acquired commodities (e.g. Martin and Grube 2000:16; Johnson 1996). Mayan rulers legitimize their positions by "impersonating" (or perhaps, transform into, or join with, in their interpretation) the gods they represent using masks, ornaments and jewelry (Houston and Stuart 1996:291), and used very fine, specialized bloodletters in other ritual contexts (e.g. Carballo 2009:494; Johnson 1996:172; Joyce 2004:195).
In staple economies, feasting activity is used as a competitive display of power through abundance in order to gain political standing and an upper hand in economic and political negotiation. It is also very expensive as expectations of diverse and plentiful provision run high. Feasting is therefore also an indication of political competition among groups within a cultural tradition (Butterwick 1998:89-90,104-105). The extent of the sudden drain on staple resources required for feasting activity requires staple economies to rely heavily on a land and climate capable of consistent agricultural surplus (Hayden 1995: 62,63). Numerous ethnographic and archaeological accounts of feasting cultures have been identified through evidence of regular feasting behavior. Many favor livestock or wild mammals as the more prized fare over agricultural products, although a wide array of agricultural products are still included, and some extensive land resources and human labor are still necessary to support horticulture for highly populated areas. Some of the most common and elaborate feasts occur at funerals, which provide opportunities
26


for interaction when a void has been created in a social position, possibly requiring a shift in the social structure (Hayden 2009). Among the best documented staple groups are those within the Tana Toraja regency of Sulawesi, Indonesia, who are very highly stratified and hold six levels of funeral feasts, each with socially understood degrees of elaboration, based on the social status of the deceased. Expected expenditures (which hosts will go beyond if possible) range from 1-2 nights with a single pig and possibly a water buffalo for poor people, slaves and infants; to at least 27 nights, over 36 pigs and 16 water buffalos for only the very wealthiest individuals. Although the latter case may only occur once or twice in a lifetime, the former tends to occur approximately 20 times per year, leading to a consistently high expense in livestock (Adams 2004:64-65).
According to Kristiansen (1991:22), wealth economies tend to be individual oriented and horizontally structured since they involve individual peer level contacts and networks for exchange, whereas staple economies are more collective and vertically structured (such as the above example of Tana Toraja), as agricultural production involves the coordination of large numbers of laborers. Kristiansen and Earle both propose that although the two strategies differ by nature and thrive within different social structures, they can exist within the same polity. One economy will become dominant, and the "alternate" economy will become a dependent variable subordinate to the other (Kristiansen 1991:22; Earle 1991:8). In Kristiansen's view, overall trends in culture trajectories follow a usual progression towards increased complexity from wealth finance towards staple finance, but each of the two economies can follow its own trajectory into different patterns of increased complexity (Kristiansen 1991: fig. 2.2). However, individual trajectories have proven quite difficult to predict with any certainty. Past
27


attempts at creating any universal rules applicable to cultural trajectories have been thwarted by differences in environments, related technological developments and individual culture histories. For example, some groups considered to be members of larger state-like organizations may exhibit greater autonomy because they are cut off in some way from their parent culture. Earle treats these situations as exceptional cases, and termed such groups devolved societies (Earle 1991:14-15,25). A trajectory may also be altered by political reorganization, which may increase the demand for rare and exotic items and/or create a period of resource intensification as factions vie for control in a power vacuum (Beekman 2010:71, c.f. Beekman and Christensen 2003:145-149, Pollard and Cahue 1999). Therefore, the particular history and environment of a group must be considered before attempting to interpret their economic base.
Despite Earle's and Kristiansen's separation of the two economies, actual examples consistently show a trend towards a strong presence of both wealth and staple economies within the same social contexts. The Mayan elites are well known for their use and trade of a broad array of eccentrics, ritual items and luxury goods, as shown in grave goods public ceremonial spaces and other ritual and elite contexts (Johnson 1996; Martin and Grube 2000:16; Moholy-Nagy 1999:310; Spence 1996; and several others). Known wealth items for the Maya include jade jewelry, elaborately decorated ceramics jaguar pelts, cacao, shells and other marine items from the coasts, exotic feathers from other regions, clay and carved wood figures of gods (Barrett 2004:60; Martin and Grube 2000:16) as well as chipped stone eccentrics and ritual items. Elite Mayan stone items are typically made from both local chert, and rare obsidian from the Guatemalan Highlands and areas as far as Teotihuacan (1200 km), which controlled the nearby Pachuca source
28


of an unusually pure green obsidian ( Johnson 1996; Moholy-Nagy 1999:300-303;
Spence 1981, 1996).
But the Classic Maya also heavily emphasized maize and fertility, and elite connections to elements related to the staple yield, including a maize god which actually represented all life cycles, including those of people. To the Classic Maya, maize was a central component of life, and therefore all life cycles are related to the pattern represented in the life, death and regeneration of maize (Martin and Grube 2000:16). The Maya also emphasized regularly scheduled feasts. The Mesoamerican calendar's 18 month cycle recognizes monthly feasts (Townsend 1992:212 215). Also, in the ancient Mayan language, the phrase for an esteemed lineage leader translates literally as head of the banquet (McAnany 1995:31).
Across polities, it does not appear that an increase in staple reliance necessarily relates to a decrease in emphasis on wealth goods, or vice-versa. In fact, quite the opposite appears to be the case for Blue Creek in the Early Classic to Middle Classic periods, which exhibits an unusually high quantity of various luxury goods (Guderjan 2007:91), and especially of jade (Barrett 2004:109,155), considered one of the most valued luxury items based on its rarity and consistency of association with elite contexts (Barrett 2004:23-24). Yet Blue Creek's main economy appears to be based on the overproduction and trade of various agricultural items, including staples such as corn, beans, squash and other foods evident in phytolith evidence examined from the very fertile bajos and highland fields in the area; and an elaborate system of canals and ditches which would have served to regulate the flow of water between the rainy and dry seasons (Guderjan 2007:91-101). This dual economy coincides with Blue Creek's fortunate combination of a
29


strategic position at the end of a long trade route to encourage its wealth economy, as well as a location among pockets of very fertile land to feed its staple production (Barrett 2004:155). A group's location along the cultural and physical landscapes, as well as the natural resources which surround it, may therefore have more to do with a group's chosen economic emphasis than its current organizational state within a structural trajectory.
Wealth Relations and Boundary Permeability
Kowalewski et al. (1983:39) have proposed that decreased centralization of a social structure and increasing group population size are both directly proportional to a boundary's permeability, where permeability is the relative amount of energy flow across the boundary. In more concrete terms, permeability can be viewed as the amount of cross-boundary interaction, which is most visible archaeologically as economic flow of goods evident in the quantity of similar materials and artifacts deposited on either side of the boundary. Centralization is defined as the amount of interaction between the most active node or settlement group and other internal social groups, relative to the overall amount of interaction within the internal social system; and size is defined as the number of interacting social groups (Kowalewski, et al. 1983:35). Since Kowalewski, et al.'s measure of size does not include group populations or areas, but only a group count, it may be best viewed as more of a measure of population dispersion, rather than overall population count or density. Group identity, according to Kowalewski and colleagues, is based on the degree of integration between component groups, where their most
30


actionable definition of integration is "the total amount of flow of matter, energy, or information between system components" (Kowalewski, et al. 1983:35).
As noted by Beekman (1996a:8), Kowalewski et al.'s theory is a proposed abstract relationship between social parameters and is not tied to any particular social model. However, the connection to semi-peripheral area economics, especially where a clear boundary is present, does find some relevance here because it points to factors that may influence the degree to which economics of semi-peripheral sites may be directed to external groups; greater boundary permeability means that semi-peripheral sites are more heavily engaged in externally directed interaction. Increased interaction also increases the potential for trade relationships which would most likely include remote items involved in wealth trade.
Elites may attempt to increase their power by artificially increasing the size of the system (i.e., adding to the number of social groups), and thus the required level of integration between groups. This strategy, however, may not succeed every time simply because the available resources are not enough to increase the system size and maintain an adequate level of interaction to avoid fragmentation (Kowalewski, et al. 1983:37). Since elite-sponsored attempts at expansion often take the form of outposts and semiperipheral area sites (Algaze 1993, Chase-Dunn and Hall 1991), expansion also may maximize exposure to outside groups, which in turn increases elite opportunity for acquiring prestige goods.
Kowalewski, et al. tested their hypothesis on count and location data for both sites and artifacts from the valley of Oaxaca, from the Middle Formative Period to the Contact Period, which include the temporally sequential groups of San Jose Mogote and Monte
31


common public. This occurred with the Early Postclassic spread of prismatic blade-core platform grinding, which made the production of prismatic blades easy enough for nonspecialists to regularly engage in the practice for their own uses (Healan 2009). In such situations, elites may need to define new forms of prestige items to symbolize their status.
The strategy is also labeled "exclusionary" because the act of selecting specific contacts for the exchange of goods also channels wealth and power away from other groups by taking advantage of resource monopolies. As a result, the practice limits the number of people which have access to certain products (Blanton, et al. 1996:5).
Groups engaged in a corporate strategy form a group identity through multiple peer institutions. The corporate group is unified through an ideology that is communicated and reinforced by ceremonial and ritual activity. Corporate power strategies limit or prevent domination of one group through a corporate code of behavior. Teotihuacan may be the largest and best known Mesoamerican example of a corporate strategy. Blanton, et al. (1996) suggest that the material culture close to the city center reflects a pattern of collective identity and action rather than sovereign rulership of a particular lineage (see also DeLucia 2008, DeMarrais et al. 1996). This is not to say that hierarchy is replaced by egalitarianism within each component institution, but rather that elite power images are suppressed in corporate level contexts such as ceremonies, collaborations on labor projects, and defense, in favor of symbols and images which support corporate group identity (Beekman 2008:414-415). Since corporate strategies are inclusive, they require large public spaces, whereas space requirements for networking strategies are very small and private due to the need for exclusion (Beekman in press a:57-60).
34


Connecting Economy and Strategy
Blanton, et al (1996, Table 3) specifically associate their networking/exelusionary strategy with Earles wealth finance, and also associate their corporate/inclusion strategy with Earle's staple finance. According to Blanton, et al., wealth economies are controlled via trade relationships which form networks to obtain, transport and exchange prestige goods so that their distribution can be tightly controlled. Prestige goods can then be used to legitimize the networked group's position of authority. Conversely, intensive agriculture practices required to maintain a staple economy (see also Hayden 1995:
62,63) require vast territorial assets and very large labor pools. Elites may form intergroup alliances to better manage and coordinate agricultural efforts, and the cooperative arrangements between several groups create a competitive advantage over any remaining smaller groups, as well as collective control over a much larger population. Such arrangements may also reduce or eliminate conflict over resources between groups within the alliance. Cooperative arrangements may then be solidified over time as long term corporate entities through an ideology that promotes the corporate bond.
Institutional Competition
Blanton, et al. present the two strategies as incompatible, even though they are commonly both used, at least to some extent, within the same groups (Beekman in press a; Blanton, et al. 1996:2), but usually among separate institutions (Beekman in press a: 57). Blanton and colleagues further state that concurrent use of both inclusive and
35


exclusive strategies may lead to conflict, and suggest that they may need to be separated at different sites with very different functions for stability within the polity (Blanton, et al. 1996:7).
Although these strategies may well co-exist with some tension and conflict in their competition for site dominance, they do not tend to contend for the same resources. As a result, their co-location is possible and appears likely in areas where both staple and wealth trade resources are abundant, such as the above example for the Maya at Blue Creek. At present, we don't know the details of how institutions utilizing each of these strategies were coordinated within Blue Creek's political and economic systems.
Bourdieu views institutions as groups which occupy certain positions in social space, which is "a space of differences, in which classes exist in some sense in a state of virtuality, not as something given but as something to be done (Bourdieu 2002:275, emphasis Bourdieu's). Institutional actions cross multiple social contexts that Bourdieu refers to as fields (Wacquant 1989:38-40) as they vie for dominance in the power structure. Social actions also are played out in physical space, and are reflected in the patterns of spatial distribution of material culture in the archaeological record. Bourdieu's fields are not directly tied to physical space, but different architectural contexts are designed around different performative strategies that may include ritual, ceremony and other forms of social interaction (Beekman in press a), and therefore a designated physical space becomes an aspect of one or more fields.
A multi-purpose community space may also become a manifestation of Turners arena, a built space where competing institutions engage. An arena can be as obvious as a literal battlefield, or as subtle as a discussion between members of opposed institutions
36


(1974:133). Community spaces are often regularly used for open festivals and other public rituals, and bring multiple institutions together to interact in one place. These spaces may regularly become ad hoc arenas for various combinations of institutions and fields. Arenas also are expected to eventually produce some ultimate decision over power, "even if it is the decision to leave things temporarily undecided" (Turner 1974:135). Despite what expectations may be present, the shifting balance of power may become an ongoing struggle, as has been the case with most Mesoamerican cultures between the Middle Formative and Postclassic periods which, according to Blanton et al. (1996:8-12) have oscillated between corporate and network strategies over time. Since institutions which employ different power strategies tend to utilize different kinds of assets, political competition becomes, in effect, a struggle over which form of capital dominates the societys economy (Beekman in press a 57-60).
The Nayari of West Mexico have demonstrated co-location of networking institutions in the form of descent groups and a community administration, both of which have their own spatially separated ritual expressions of ideals intended to instill loyalty to the institution. The relationship has been far from stable, however, as the lineages and the community administration have been in constant conflict, and emphasis has shifted back and forth between the two strategies throughout their history, much like the situation described by Blanton et al. (1996) for much of Mesoamerican prehistory. Instability in the Nayari case, however, is mired in direct interference from the Spanish during the colonial period, and the dominant American governments through much of the 20th century, which overturned and altered much of what may have been a pre-existing institutional balance. Beekman (in press b:7,8) has suggested that two factors may need
37


to be in place for both institutions to coexist in a stable relationship: Explicit benefits of belonging to each institution need to be understood by its members, and some form of cross-legitimization of institutions pursuing both strategies may need to be embedded in the group's ideology. Such cross-legitimization, which may perhaps bind networking institutions, rather than individuals, more directly to the community institution may have been lost after Spanish contact.
Another problem in the Nayari case is that the traditional community ritual does not show any associated resources or capital which may benefit participants. The only potential economic benefit is the elder responsibility of insuring the rainy season cycle, but land and agriculture appear to be the sole property of the lineages (Beekman in press b:5, 7).
Combining Economies: Context and Matrix Control
Earle and Blanton each proposed a form of social separation between economies. Earle proposed a separation of wealth and staple economies by polity, and Blanton, et al. suggested the coexistence of the two economies within the same polity, but within separate institutions and probably also separate sites. Kenneth Hirth (1996) has criticized this tendency towards compartmentalization of prehistoric economies:
"The dichotomization between food and nonperishable commodities obscures the fact that (1) both food and luxury items play complementary roles in the development of political economies, and (2) elites may
38


reorganize agricultural production to generate surpluses to trade for luxury items or control the production of craft goods or luxury items to procure food." [Hirth 1996:208]
Hirth developed a model which allows for the control and activities involved in both wealth and staple economies to co-exist at the same site. Hirth's matrix control principle suggests that elites may actively accumulate and diversify economies which they control even across wealth and staple boundaries as a risk management strategy. Elites accomplish this by including diverse strategies as separate lower-level institutions within the social structure. Production and accumulation of resources can then be managed under separate provisioning networks (Hirth 1996:224-225). However, the potentially conflicting ideology associated with a corporate institution must also be limited to that institution, and made to work in the context of an over-arching authority.
While the matrix control principle allows for coordination of different economies under the same social structure, Hirth's context principle addresses the need for spatial separation of economic bases by suggesting elite supervised spatial contexts for the production and storage of wealth. The principle does not require same-site economic activity, since mobilization networks are likely also included in the system for accumulation, but co-location of activities would be advantageous to supervision and management (Hirth 1996:223-224). Similarly, ritual activities for incompatible networking and corporate ideologies may utilize separate built space contexts, as suggested by Beekman (in press a).
39


Under Hearth's model, relative emphases on each economy under elite control largely depend on elite choices, although other factors, such as fluctuating availability of staples and changing demand for wealth items, remain important considerations. This changes the view of the economic balance that we may expect to be represented within the material culture of a site, relative to expectations of the models of Earle and Blanton, et al. Rather than a categorical view of wealth or staple dominance, the data may reflect more subtle differences of material culture represented by each economy. Hirth also identified several other economic categories that elites may attempt to control, although most (such as service economies) would not be represented in the material culture, especially when confined to lithic data, so Earle's wealth and staple categories will suffice as the more manageable division.
Structure, Economy and the Natural Environment
With all of the discussion on economic and related social dynamics in the literature, it often appears as if the natural environments of each respective social group are little more than backdrops in which these dynamics are played out. Environmental factors actually appear to have substantial effects on economic and strategic aspects of cultures (Blanton, et al. 1996:7, Earle 1991:13-15). However, opinions about the environment's specific role in shaping cultures have historically varied quite widely on a continuum of cultural vs. environmental dominance. Classic processual environmental determinists have commonly viewed cultures as "spatially delimited bodies of individuals living within and adapting to a specific physical environment" (Schortman and Urban 1987:63). On the
40


opposite end of the continuum, social theorists often see cultures as primarily driven by internally generated constructs of meaning which often actually instigate environmental change (Hodder 1982). Schortman and Urban have argued that the old processualist view ignores cultures which live and interact throughout areas of highly variable environments (1987:63), and Chase-Dunn and Hall have also suggested that a purely environmental distinction may work for some groups, but not others (1991:15-16). Others argue that the opposing concept of cultures as strictly innate, internally generated phenomena ignores evidence that environmental features combined with major climatic changes appear to have taken a large role, if not an exclusive one, in the formation and reconstruction of social structures (e.g. Chatters and Prentiss 2005; Shaw 2003). These few arguments are just a small sample of a very large debate which extends well beyond the intended scope of this paper.
Many archaeologists have chosen to either seek a middle ground or bypass arguments related to the primacy of nature vs. culture, and instead concentrate on the nature of the interplay between culture and environment. Maya lithicist Jason Barrett asserts that "neither natural nor anthropogenic inputs into the system necessarily take precedence in the metamorphic processes that affected realized landscapes. There is instead a perpetual balance of actions and reactions that inhibit stasis in either nature or culture" (2004:53), and Townsend states that the need to acquire resources required for survival is completely interrelated with the need for explanation and meaning (1998a:23).
Others still place culture and environment on somewhat less equal grounds: Otto Schondube has conceded to at least some internally generated cultural influence in stating that geography is very highly influential, although not entirely deterministic (1998:205),
41


and John Jackson (1984) assigns a very fundamental role to the natural landscape in determining the degree of social organization. Some researchers state that corporate strategies tend to be pursued in environments where the most potential exists for agriculture (Blanton, et al. 1996:7; Hayden 1995: 62,63), and Blanton, et al. (1996:7) have also claimed that network strategies tend to occur where environments are marginal for staple goods, and also geographically in semi-peripheral or peripheral areas.
However, they also state that political reorganization may create networking opportunities in the core as well. Here again, the authors take a view where the environment plays a dominant role in determining the economic base, which in turn determines the strategic direction of elites, but with some exception made for political upheaval in creating economic change.
Jackson (1984:150-151) makes a distinction which may also parallel core and periphery zones, between fragmented and irregular physical landscapes which he terms vernacular landscapes; and orderly, well connected areas which he calls political landscapes. Vernacular landscapes are also reflected in a social and political absence of structure, and are "usually small, irregular in shape, subject to rapid change in use, in ownership and in dimensions". Political landscapes offer a sense of centrality and structure, which is reflected in a structured landscape that includes "such things as walls and boundaries and highways and monuments and public places" (Jackson 1984:12).
Although the direction of this thesis does not directly address views regarding cultural or environmental primacy, it recognizes the interplay between culture and environment as highly influential in the formation of a culture's organization and economic base, and utilizes environmental data alongside material cultural data to determine initial
42


expectations for the questions being addressed. The supply potential is certainly a limiting factor in making decisions on an economic base in prehistoric contexts, but given multiple choices for viable economic pursuits (which may or may not be the case), elites will select and emphasize certain economic bases, based on multiple factors, including the demand for certain goods and in certain forms both within the core of a polity and among potential trading partners.
Examining Semi-periphery Economies
The theoretical question addressed in this thesis is whether prehistoric sites in economic semi-peripheral zones would have operated under a different emphases on economic bases, when the semi-peripheral zones differ in the availability of natural resources and access to internal groups. More specifically, it asks whether the semiperipheral area site that diverges the most from the core given the immediate presence of obsidian, lower access to arable land, and internal group access, emphasizes an economy based on wealth trade over one based on staple production.
According to Blanton, et al., "to understand social change of this type in marginal environmental cases the analyst must take a 'top- down' view, placing the local system within its larger macroregional context to examine its role in the control and manipulation of intergroup exchanges" (1996:7). Studies of peripheral and semiperipheral area relations in general should also begin with detailed information on social and economical organization at all scales before we can interpret comparative patterns between sites (Schortman and Urban 1987:80). Therefore, the remaining chapters serve
43


to tie background patterns discussed at the regional level, both inside and outside of the polity, to competing local strategies reflected in the material culture evidence of the social arenas formed within public spaces. It is hoped that this approach proves more useful than the usual research strategy of picking a scale of operation to work from, with no implications drawn from above or below that scale (Beekman 2000:386).
44


CHAPTER III
REGIONAL BACKGROUND
This chapter discusses environmental and cultural factors which may have influenced a semi-periphery site's degree of interaction with, or independence from, the core culture; and whether different availability of certain resources may have led one site to de-emphasize the core's staple economy and pursue wealth trade to a larger extent than a site associated with the core.
This chapter begins with a description of the natural environment within the Tequila Valleys, including the distribution of resources that would have influenced the economic and cultural trajectories of peripheral, semi-peripheral and core areas. The public architecture of the area (guachimontones and shaft tombs) is then described in relation to associated material culture evidence of economic strategies that appear to be employed within the region. The three political-economic zones outlined in Chapter II are then described for the region occupied by the Teuchitlan, and finally the sites under study (Llano Grande and Navajas) as well as the primary site of Los Guachimontones and the peripheral area of the Sayula Basin, are described in greater detail to determine more specific cultural and environmental characteristics related to each site's economic base.
45


The Natural Environment and Subsistence
Geology and Climate
West Mexico's Highlands region is situated at the intersection of two ranges: The Trans-Mexican Neovolcanic Belt that runs east-west across the continent from the Pacific to the Atlantic at the Gulf of Mexico, and the Sierra Madre Occidental which follows the Pacific Coast (Beekman 2010:44; Darling 1993:figure 1). The area is characterized by topographic complexity, extreme changes in altitude, a wide variety and abundance of inland fresh water sources and very fertile agricultural land that together produce a wide range of available resources (Beekman 2010:44; Schondube 1998:205; Stuart 2003:3; Weigand 1985:55; Weigand and Beekman 1998:37). Land and climate reliably capable of producing a substantial agricultural surplus is necessary for systems which rely on competitive displays of abundance for social positioning and political control (Hayden 1995: 62-63). The geography of the Tequila Valleys provides numerous pockets of arable land fully capable of overproducing staple crops. The lake basins of the Late Formative and Early Classic periods primarily consisted of large connected marshlands in the valleys surrounding the Tequila volcano (Weigand and Beekman 1998:37). One exception existed within the Magdalena Basin on the west side of the Tequila Valleys, where recent geomorphological surveys show that the majority of the area within the Magdalena basin was still a lake rather than marshland (Anderson et al. 2013:25). Still, large enough tracts of land were made available for agriculture in the area as the lake waters receded, and the surrounding mountains offered ecological diversity (Weigand
46


and Beekman 1998:37). However, rainfall is extremely seasonal in the area. 85 90 percent of annual rainfall occurs during the 5 month rainy season from June through October. The season produces very heavy rains, with rainfall accumulations of 900 -1600 mm per year.
The seasonality of the rain was likely not as much of an issue within the valleys in areas near the lake basins. The marshy nature of these areas and surrounding Lake Magdalena was likely similar to that around the remnants of the lakes that exist in the area today. Some items retrieved during Adela Breton's 1895 visit reflect the prehistoric landscape of their day as well: bracelets acquired from a purported grave context were decorated with carved frogs (Townsend 1998a: 15-16). Such images reflect the importance of marshland to the Teuchitlan culture. The marshes acted as a stabilizing factor between the extreme wet and dry seasons by storing moisture for agriculture, which likely greatly extended the crop season.
Geographic Area Relations to Political-Economic Zones
Schondube has stated that the microclimates formed across the extremely varied landscape of Jalisco made for close access to a large variety of available resources, but there was not enough resource capacity within each microclimate to support large settlements beyond chiefdoms (1998 207,215; see also Fernandez and Deraga 1988), thus the West Mexico Highland area was more suited to small, scattered settlements.
However, Schondube's assessment covers a very broad region, and the area of the Tequila
47


Valleys presents one exception as a relatively large, continuous ecosystem connected by the long series of lakes with abundant agricultural soils and other resources.
Economic Zones of the Teuchitlan Culture
Socio-economic roles do not automatically coincide with geographic areas, but the distribution of Teuchitlan architecture suggests a social pattern which closely relates to the geography of the Tequila Valleys. Weigand (1985) defined a cultural core zone which covers the entire Tequila Valleys region, with a periphery that describes any Teuchitlan culture outside of that region. Weigand also described an especially large and dense "habitation zone" within his core, to the south of the Tequila Volcano which in practice has been considered a sort of "core of the core". Following Beekman (in press a), the core zone within the Tequila Valleys is defined by a marked increase in architectural density across the landscape where distances between building and patio groups average about 100 m, and rarely exceed 250 m (Weigand 1985:82). The zone forms an arc of settlement that covers an area of 240 250 km (Weigand and Beekman 1998:39, Beekman 1996b: 136). This is a smaller area than defined by Weigand and corresponds to his "southern habitation zone" (Weigand 1985).
Also following Beekman (in press a), The semi-periphery corresponds to the remainder of Weigand's core, which include the areas to the north of the volcano, the Magdalena basin to the west, and the area within the Valencia Lake Basin to the Southeast. Semi-peripheral sites form a roughly defined outer core ring, still encompassed by the Tequila Valleys but much closer to the surrounding hills which mark the outer edge of the valleys (figure 1.2).
48


The peripheral zone matches both Beekman's and Weigand's definitions, and includes all Teuchitlan culture sites outside of the Tequila Valleys. Peripheral sites are spread over a wide range within West Mexico, including the states of Jalisco, Colima and Guanajuato (figure 1.1).
The natural landscape of the Tequila Valleys provides the first clue that semiperiphery areas may have varied from the core in political and economic organization. It is perhaps easiest to visualize the Tequila Valleys landscape starting from the prominent central cinder cone known as the Tequila Volcano, and moving outward through what are now very dry lake basins which, at the time of Teuchitlan occupation, were still mottled with receding lakes (Anderson et al. 2013:25). Across these basins from the volcano in all directions is the irregular "ring" of hills and valleys, which contain several passes with limited access to the basins from the periphery.
The more level and expansive area of the core allowed for easier travel between sites. This facilitated regionalized polities to mobilize people from larger areas, and allowed for greater corporate management of the labor-intensive staple economy, whereas the fragmented hills and gorges of the semi-periphery areas were more conducive to one-on-one or small group elite networking for wealth trade.
Weigand and Beekman (1998:47) have suggested that the Tequila Valleys' topographical features tend to foster a form of Jackson's political landscape (1984) with a strong tendency towards central control. The core landscape sharply contrasts that of the semi-periphery sites in the surrounding hills, which describe a more disorganized and fragmented vernacular landscape also defined by Jackson. Even though the architectural landscape within the valleys has been viewed as multinuclear, with several relatively
49


independent city centers controlling or vying for separate areas within the territory (Weigand and Beekman 1998:51; cf. Martin and Grube 2000:18-21), areas within the core were likely more centralized than the populations within the topographically fragmented semi-periphery areas.
Subsistence Resources
Maize was an important crop throughout the area, but nearly all subsistence studies for West Mexico within the last 20 years contain a recurring theme of variety in food sources rather than extreme reliance on domesticated maize. Pre-Spanish contact flora resources were extremely varied and most are still found there today. At the time of Teuchitlan occupation, the area likely saw an abundance of yams, avocados, soursops, tomatoes, chiles, beans, various squashes, maize, amaranth, chia (sage seeds), papayas, Maguey (agave), nopal cactus (tunas and paddles), cacao and several other crops. Fauna, however, were limited to dogs, turkeys and very localized groups of bees and ducks. The native animal species did not include the now-familiar work and pack animals of the area, and no animals capable of functioning in such a capacity are known to have existed in West Mexican prehistory. People therefore primarily travelled and transported resources on foot, over and around very steep and difficult terrain (Schondube 1998: 205- 206). Even place names in the area reflect the abundance of resources: Mazatlan, the place of deer; Michoacan, the place of fish; Zapotlan, the place of zapotes. Archaeological evidence for the utilization of Maize, beans, chiles and squash goes back thousands of years and
50


midden evidence has been located for apparent food use of over a 16 faunal species, including deer, small game, crustaceans and birds (Schondube 1998:208).
The diversity of food resources within the Tequila Valleys was well exploited during the Teuchitlan occupation of the area, but maize remained the most common crop. One environmental impact study of ancient maize fields in the area has estimated the degree of reliance on the common Mesoamerican staple relative to other crops at around 50 percent, although somewhat lower near the lakes and wetlands where resources more directly associated with the lake ecosystem were plentiful (Beekman and Baden 2011:359). Agave (maguey) flourishes in the high country within and around the semiperipheral zones where it is generally grown commercially in modern times (Chadwick 2011; Heredia 2008), and was likely heavily utilized in that area during the Teuchitlan occupation as well.
Within the Sayula basin, prehistoric oven shapes and sizes most resemble those described in ethnographic data for the roasting of maguey pinas (hearts), a practice which today is most closely associated with Tequila production from the blue agave variety most abundant within the Tequila Valleys, (cf. Francisco Valdez 1998). Vessel forms often depict various food sources, sometimes prepared for meal service, such as roasted maguey leaf sections, squash, pitayas and organ cacti (Schondube:209-210, figures 13, 14, 18-20, 22-24).
51


Minerals
Abundant mineral resources in West Mexico include obsidian, various forms of greenstone, various crystals, copper, silver, quartz, opal and salt (Weigand 1985:56, Weigand and Beekman 1998:37). Obsidian was one of the most utilized minerals throughout Mesoamerica during the time of the Teuchitlan, and Jalisco has the largest number of obsidian sources in Mesoamerica (Clark and Weigand 2009:79). The Tequila Valleys are especially rich in obsidian deposits (Beekman 2010; Esparza 2003:74,86; Stuart 2003:3), and contains over 30 obsidian outcrops, at least 12 of which appear to have been exploited before European contact (Spence et al. 2002:65).
The Economic Impact of a Changing Climate
Despite the apparent climate stability expressed by Schondube (1998:205-206), relatively slow and subtle climate changes over time may have profoundly affected cultural dynamics within the area. One question of interest is the cause of the expansion of the Teuchitlan culture from the core areas to the hills and beyond, and what part a changing climate may have played in the decision to expand and establish peripheral and remote sites. Understanding the reason for the move may determine something about the purpose of the remote sites. Were people forced to seek alternate resources, or was the expansion an enterprising "power move" on the part of Teuchitlan elites to gain better access to remote resources?
52


Overall, the Late Holocene has been quite variable in climate in the Highland Lakes area, with multiple wet and dry periods (Schondube 1998:205 ). The drying trend which began after the Late Formative Period may have profoundly impacted people living in the region from the Early Classic to the end of the Teuchitlan culture at the beginning of the Middle Classic (Anderson, et al. 2013; Beekman 2010:61).
Whether or not drought was motivating factor for culture change is unclear. However, Early Classic settlements in the hills of the semi-peripheral zone would have required different subsistence strategies than settlements in the valleys, although hills would not have been entirely agriculturally impoverished. Maguey thrives in hot, sunny and arid climates and is quite tolerant of poor soils and drought (Chadwick 2011; Knox 2013).
The succulent also tolerates occasional moderate frosts down to about -3C (25 F) (Chadwick 2011). The Jalisco high country, and in particular, the hills north of the Tequila Valleys, are considered an ideal climate for maguey, and are currently one of the main regions for growing the blue agave variety commercially (Chadwick 2011; Heredia 2008). The agricultural fields in the valleys south of the volcano (including the former Teuchitlan core), in contrast, are primarily populated with maize and sugarcane.
Economic Strategies Within the Teuchitlan Culture
The two prominent forms of ritual spaces among the Teuchitlan, shaft tombs and guachimontones, appear to support the two economic strategies outlined by Blanton, et al. (1996). The shaft tombs reflected a network or exclusionary strategy manifest by descent groups which is associated with a wealth finance economy. In contrast, the
53


guachimontones were the setting for more group-oriented activities associated with staple finance (Beekman 2000:403-404; in press a:57-60). These economic bases and their relationships with the ritual architecture are described below.
Staple Finance
Guachimontones: Corporate Management of a Staple Economy.
The ritual architecture of the Teuchitlan culture, the guachimonton, is composed of a circular patio with rectangular buildings arranged around the circle's circumference, and usually a circular stepped pyramid altar in the center (Beekman 2008:419, 2010:62; Weigand 1985:66-69; Witmore 1998:138; figures 3.1, 3.2). Ceramic dioramas depicting Guachimontones, as well as wall remains on the actual patios, show the presence of buildings surrounding the patio. Dimple-shaped holes were discovered on the top surfaces of the altars within Los Guachimontones (Townsend 1998b: 110) and in the center of the patio at Llano Grande, and some clay dioramas show the presence of a vertical pole in the same positions. The pattern has been the subject of much discussion on the rituals it hosted (e.g. Beekman 2000, 2008; Weigand and Beekman 1998; Whitmore 1998).
The earliest evidence of occupation at guachimonton sites is from around 300 B.C., and construction of guachimontones started some time later, by 100 B.C. (Beekman 2010:63). The level of effort required for these circular constructions is substantial. Even the moderately sized 8-building circles would have involved labor efforts in the range of hundreds of thousands of person-hours. Construction of the circles at the site of Los
54


Guachimontones is estimated to have required about 1.25 million person-hours of labor (Beekman 2000:395, 397). Despite the variation in architectural details, guachimontones follow several patterns relatively consistently across the landscape throughout the region. The number of buildings around the patio varies, but most have 8 satellite platforms and other variations are limited to 4, 10,12 or 16 platforms, all arranged in opposing pairs (Beekman 2008:419). Different construction techniques were used for each building in
Figure 3.1: Circle 2 guachimonton from the Los Guachimontones site.
55


Circulo 5, Navajas
Figure 3.2: Navajas Circle 5 guachimonton plan view.
several documented cases, so the buildings were most likely constructed by different groups, with different learned techniques (Beekman 2008). Beekman has suggested that the groups representing each building on the circle were likely competing lineages (2000:414, 415).
Material culture within the guachimontones tends to be less ornate compared to artifacts within the shaft tombs described below, possibly due to suppression of aggrandizing behavior in a space dedicated to cross-institutional unity (Beekman in press a:63). For example, Catherine Johns' ceramic ware study (2014) from one core guachimonton complex revealed that the majority of the sherds by weight were of a plain utilitarian ware related to cooking and storage (Colorines), or a very uniformly painted
56


ware associated with communal feasting (,Arroyo Seco). Relatively few of the sherds in each building were determined to be from more elaborately decorated forms (Tabachines and Estolanosj typically associated with aggrandizing ritual behavior (Johns 2014:figures V-12, V-13).
The social use of guachimontones appears quite varied from ceramic dioramas found within tomb contexts, which depict different scenes of dancing, drinking, playing music, apparent marriage ceremonies and rituals (Beekman 2003b: 12, in press a:63,64). dioramas also depict apparent funerary processions crossing a guachimonton (Beekman 1996b: 136, in press a:62). Clay dioramas found in some tombs depict scenes of dancers around the pole (Townsend 1998b: 111).
Several archaeologists have proposed that the guachimonton architectural form symbolizes agriculture or fertility (Beekman 2003b, in press b), and that elites were involved in the staple economy through frequent agriculture related rituals held within and around the guachimontones (Beekman 2000; Butterwick 1998; Lopez and Ramos 2006a,b; Schondube 1998). These proposals suggest symbolism of economic aspects that extend beyond the guachimontones themselves, and are described in the following sections.
Concentrations of obsidian often appear in ceremonial centers, and visual sourcing most often points to nearby area mines as the sources for collections within the centers. Therefore, any elite control of obsidian was likely not centralized. Spence, et al.
(2002:68) have suggested that elites controlled obsidian on a per-area basis from secondary guachimonton sites that oversee each area. Control of obsidian may have been an important component in the political dynamics of the area (Spence et al. 2002:68), and
57


Weigand (1985:89; 1996:96; and others) has stated that guachimontones are also often associated with obsidian workshops, although Weigand's evidence is very preliminary since it is based on surface survey reports. If further excavation verifies Weigand's observation, the association reflects elite interest and likely control of the obsidian production industry.
Agriculture Production
Beekman (2003b: 14,17) has suggested that the guachimonton architectural form may symbolize maize, since the pattern is quite similar to a cross sectional view of an ear of corn. In this view, the circular patio represents the cob and the surrounding buildings represent kernels. For the Harinoso de Ocho variety of maize, a likely common variety for the Teuchitlan culture, ears most commonly carried eight rows of kernels but ten and twelve kernels are also found on occasion. Likewise, most guachimontones in the core and semi-peripheral areas have eight surrounding buildings, although ten and twelve building guachimontones are occasionally found. Maize is a very high yield crop on which Mesoamerican people heavily depended, but also highly vulnerable to climate and soil conditions. It is therefore considered a high-risk crop. Attempts to mitigate the risk to the amount of surplus required for elites to maintain their political status probably relied heavily on cosmological solutions (Beekman 2003b: 18).
Others have suggested that agricultural fields themselves show evidence of production intensification. Weigand (1993:228) has interpreted cross-hatch patterns in the Tequila Valley basins as artificially raised agricultural fields called chinampas. Conversely,
Stuart (2003) has suggested that the pattern is actually a series of water management
58


channels, based on a lack of sufficient sediment for chinampas. Stuart also warns that his uncertain dating of the fields to the Terminal Classic period should be treated "with extreme caution" (Stuart 2003:241), and a geomorphological study of the lake basins in 2013 demonstrated that Stuart's canal system would have dried up long before his proposed termination dates (Anderson et al. 2013). Stuart's uncertainty of the dates, however, opens up a possibility that agricultural channels could have been utilized much earlier.
Beekman (2010:61) also suggests that The Early Classic Period expansion of the Teuchitlan culture spread ideas of agriculture based ritual and social heterarchy into the semi-peripheral and peripheral areas, which might have fundamentally transformed economies throughout West Mexican into a regional agrarian system if the gradual drying trend leading to Epiclassic drought had not already begun to affect the local environment. The drying trend also may have influenced the semi-peripheral zone sites to at least partially replace or augment their staple finance with a greater emphasis on the wealth economy.
Feasting
Butterwick has proposed that ritual feasting, an integral part of a staple economy, was pervasive throughout the Teuchitlan culture. The archaeological evidence suggests that the practice goes back at least as far as 100 B.C. (Butterwick 1998:99). Numerous depictions of food in ceramic art suggest that feasting was a deeply ingrained aspect of the West Mexican way of life since the Late Formative period, and some figurines suggest the ritual consumption of aguamiel {maguey nectar) or it's fermented form pulque
59


out of a container (Butterwick 1998). The contents of shaft tombs, particularly at the semi-peripheral site of Huitzilapa (Lopez and Ramos 2006a,b), have also suggest a related component of large food offerings for the dead, likely related to food for the deceased and funerary displays (Beekman 2000:393).
Wealth Finance
Shaft Tombs: Networking and Wealth Trade
Shaft tombs are often associated with Guachimontones, but are more frequently distributed outside the Tequila Valleys than Guachimontones (Weigand and Beekman 1998). They are nevertheless unique to West Mexico within Mesoamerican contexts, although tombs of similar construction also appear in South America (Anawalt: 1998:238). The tombs are composed of a vertical shaft which leads down several meters to one or more chambers (figure 3.3). These chambers house multiple burials and may include numerous grave goods. Shaft tombs have been of interest to both archaeologists and collectors for much of the 20th century, but until the last four decades, the interest has only been due to the fact that they are the most productive caches of figurines, ceramic vessels, obsidian objects and other artifacts for modern-day collection, trade and sale (Beekman 2000, 2008; Weigand 1985). Still, much of what we have learned thus far about the culture and economic relationships within West Mexico have been derived from the contents of shaft tombs. Shaft tombs have offered some of the best preserved contexts due to their locations several feet underground, and in relatively stable soils that avoid early collapse. But looted shaft tombs still far outnumber
60


0 12 3 mtrs.
r on,
Figure 3.3: Profile of a shaft tomb from the Atitlan Las Cuevas site.
From Beekman 2000:391 figure 4.
those excavated by professional archaeologists, and Guachimontones have suffered the same fate. Published artifact data from actual tomb contexts is therefore very thin (Beekman 2000:386). But when found in context, some details of grave goods can be used as clues regarding ideologies and social relationships, and a differential distribution of goods between interments can provide clues about the organization and degree of social stratification.
Shaft tombs appear to have been used only for a certain portion of the elite population; overall, only about 10 percent of the population appear to have been buried in the tombs, with the rest interred in much simpler pit graves (Beekman 2000:391). Lineage appears to
61


have been emphasized over individuality in shaft tomb interments. Multiple individuals are sometimes buried within the same tomb, and genetic testing on human skeletons interred within tomb chambers has shown that individuals within each chamber are biologically related (Beekman 2008:418).
Middle- to Late-Formative shaft tombs are quite variable in construction with three ranks defined by Weigand according to size, elaboration and depth, which varies from 4 to 17 meters. Shaft tombs are often found closely associated with Guachimontones and in some cases underneath guachimonton buildings. They are also often interspersed among common pit graves (Weigand 1985: 64,66,128). Shaft tombs become smaller and contain less wealth in offerings in the Early Classic (Galvan 1991 as found in Beekman 2000:396).
Funerary processions are depicted by some ceramic dioramas found within the tombs. Processions appear to function as displays of the wealth offered to the interred, which provides a way to advertise power as elites negotiate their position in the social structure (Beekman 1996b: 136; Beekman in press:62). Shaft tombs instilled Chase-Dunn and Hall's idea of "differentiation" (1991:19), a social difference of prestige, the ability to mobilize people, and economy rather than one of overt territorial control. The presence of prestige goods within the tombs also illustrates Blanton, et al.'s (1996) networking/exclusionary strategy operating through a network of kin relations.
Obsidian eccentrics and other status items within the Teuchitlan culture are much more concentrated in elite locations, such as shaft tombs and guachimontones, in Late Formative and Classic period contexts than in earlier contexts; elites appear to have exercised a much greater degree of control over objects of authority during these periods
62


(Beekman 2010:63). During the later periods, elite grave offerings included imported goods from areas across the Mesoamerican continent, and include seashells from both coasts, jade beads and cotton (Beekman 2000:394, 2010:63, Lopez and Ramos 1998:61).
Differential access to wealth goods is evident in the different distributions of goods between tombs, and even chambers within the same tomb, which shows clear evidence of a fully established social stratification (Lopez and Ramos 2006b), and an exclusionary strategy. Obsidian artifacts in shaft tomb contexts include cruciforms, lunates, thin circles, pendants, large double pointed knives, mirror backs, beads and ear spools (Weigand 2000:49).
Ceremonial ceramics and other wealth goods are also extremely variable in style within each chamber collection, and Butterwick has suggested that the diffused ceramic distribution pattern is likely due to frequent trade and/or gift giving, possibly during funerary and mortuary events (1998:104-105). The practice of staple-financed feasting may therefore have actually encouraged internal wealth trade by simply providing a social context for the interaction of elites between groups, even if the corporate ideal officially discouraged networking practice.
Long-distance Trade
Evidence of foreign goods within shaft tombs indicates that long distance trade intensified through time to support the rise of an elite class. Some have suggested specific long distance trade contacts from South America based on material culture similarities and travel feasibility studies (Anawalt 1998; Callaghan 2003), and Weigand (2000, 2008) has proposed trade with the American Southwest based on artifact similarities and the
63


presence of high quality turquoise. However, the Southwest connection has been considered doubtful, since the Southwest had closer access to obsidian and the cultural similarities to West Mexico do not extend as far as the Tequila Valleys. Also, Weigand's turquoise was judged on visual aspects. Since existing compositional sourcing techniques for turquoise have not been successful, Weigand's assertion cannot yet be confirmed.
Obsidian's Relation to the Teuchitlan Wealth Economy
Jay Johnson (1996:171) has suggested that obsidian in general may occupy an intermediate status between a staple good and an elite item throughout Mesoamerica, since obsidian objects have been consistently found in highest concentrations within elite household and ritual contexts, yet are also present in commoner households. Obsidian is also made into both utilitarian and ritual or eccentric objects. Within the Tequila Valleys, utilitarian forms of obsidian as well as raw or prepared cores can be considered staple goods because of the material's ubiquitous presence throughout the area. More finely worked eccentric or ritual forms, however, were more likely produced by specialists and therefore easier to control by elites. We must also distinguish between obsidian designated for internal use, and that which is prepared for export to areas for which the material is in much greater demand. The production of finished goods from fine quality obsidian within consumer areas in other parts of Mesoamerica appears to most frequently occur within elite contexts (Johnson 1996; Spence 1981). Traders seeking the material would need to be connected with remote elite groups. Obsidian bound for export may then be considered a wealth good, since it can be used to obtain elite imports. Therefore,
64


obsidian items considered wealth economy goods are restricted to external trade items, and all eccentric items whether internally or externally circulated.
Economies and Ritual Space
The material culture evidence demonstrates that Teuchitlan elites relied heavily on both staple and wealth economies to maintain their power. The evidence for corporate crop related ritual and feasting is abundant, as is the evidence for the exchange of wealth goods throughout, and possibly even beyond, the Mesoamerican subcontinent.
Considering the distribution of wealth and staple related material culture as it relates to the shaft tombs and the Guachimontones, it appears that the primary purpose of the Guachimontones is for corporate festivities and rituals mainly related to the staple economy, and the primary purpose of the shaft tombs is the display of wealth in the context of the social network. However, unlike the degree of spatial separation predicted by Earle (1991) and Blanton, et al. (1996), both wealth and staple economies appear to not only coexist within the same sites, but in some ways can even complement each other, even within the same ritual spaces. The configuration of the Guachimontones reflects the formation of corporate groups depicted by the central patio and altar, out of the institutions that form the wealth economy; the kin groups which build and occupy the surrounding buildings. As Butterwick (1998:104-105) has noted, corporate rituals and festivals in the spatial contexts of the Guachimontones provide a potential catalyst for intergroup wealth trade among elites of different subgroups. The result is an internal site network formed from the elites of each kin group which resulted in the wide dispersal of
65


different forms of wealth items, but still only among elite members of each group. Guachimonton patios may then act as arenas for trade and other interactions, such as inter-group marriage relations and political maneuvering. Also, the use of ritual items in the corporate ritual contexts, including elite attire, jewelry and symbols of power, is demonstrated in some of the clay dioramas (Anawalt 1998:figure l;Weigand 1996:figure 9). Conversely, shaft tombs display large quantities of food items alongside the various wealth goods, as the perceived need for staple supplies (and the elites' perceived ability to provide them) extends well beyond the earthly plane (Lopez and Ramos 1998, 2006a,b).
Timeline
Carbon dating combined with ceramic seriation has determined that the guachimontones were first constructed as large centers in and near the core during the Late Formative period, starting around 100 BC. Much smaller guachimonton construction later spread into the surrounding hills and more distant outlying areas, as far as Colima near the Pacific Coast and into all neighboring states, around AD 200. Around the same time, new construction within the core ceased, and only evidence of structure modifications appears for the core area during the Early Classic period. Based on a lack of evidence for any later maintenance of the structures, the culture is currently believed to have collapsed around AD 500. (Beekman 2010:64; Beekman and Weigand 2008:315). The shaft tomb culture also saw its peak in the scale and elaboration of construction, as well as the quantity of wealth offerings, roughly at the time of the earlier
66


guachimontones. Much like the guachimontones, the size of shaft tombs decreased in the Early Classic period, as did the quantity, size and wealth of tomb offerings.
Several timelines with different phase definitions have been created for different areas within the Tequila Valleys. Recent phase definitions for the sites and periods of interest is included here for general reference (Table 3.1). The periods of interest here are Tequila III (100 BC AD 200) and Tequila IV (AD 200 500). Although timelines between different sites are not well synchronized over long time periods, an AD 200 transition is common to several locations and aligns well with the transition from the Late Formative Period to the Early Classic used throughout most of Mesoamerica. This thesis most often references the most universally understood Late Formative and Early Classic Mesoamerican periods, following Beekman (2010).
Distribution of Architecture
Even though the main features of guachimontones and shaft tombs are evident throughout the Tequila Valleys, more specific architectural differences exist between the core, semi-peripheral and peripheral areas. These differences may show different relative emphases on different economies by reflecting differing ideologies promoted by aggrandizing wealth-oriented political organizations rather than corporate ones.
Ideologies which stress community are directly tied to Blanton, et al.'s (1996) corporate/inclusion economic strategy and therefore more likely target a staple economy. Aggrandizing systems which elevate one group above others at the highest structural level may be more likely to pursue a network/wealth trade economy. Although both
67


Table 3.1: Site occupation spans and related periods relevant to the current study
Date Mesoamerica (Beekman 1996b:figure 3) Tequila Valleys (Beekman 1996b:figure 3) Los Guachimontones (Beekman and Weigand 2008:figure 9) Sites Under Study (Beekman and Weigand 2008:figure 9)
400 500 A.D. Middle Classic Tequila IV Ahualulco
300-400 A.D. Early Classic
200 300 A.D. Llano Grande
100 200 A.D. Late Formative Tequila III El Arenal Navajas Circle 5
0 100 A.D.

100 B.C. 0
200 100 B.C. Tequila II
300 -200 B.C.
400 -300 B.C. Late Tequila I San Felipe
economies are evident in the architecture of all three zones, sites appear to become relatively less corporate and more wealth oriented with increasing distance from the core.
Core Sites
The Teuchitlan culture's peak population period within the core settlement zone has
been very tentatively, although conservatively, estimated at approximately 40,000 people
2 2
(Beekman 2008:416). The overall settled area covers 24 km with 30 km of proposed
68


2
Chinampas (or canals, according to Stuart [2003]) and 300 km of terraced fields (Beekman and Weigand 2008). Many of these claims and figures have not been substantiated by excavation and proper dating, so these data must be viewed with caution. At the latest published count, 38 ceremonial centers have been verified in the core population zone (Beekman in press a:69).
The forms and sizes of buildings surrounding the guachimontones in the core are quite uniform. No buildings stand out as being larger or more elaborate around or near the circle (Beekman 2008), which points to the representation of a likely corporate power sharing system between groups represented by each building (Beekman 2008). Shaft tombs within core and semi-periphery areas can be much more elaborate and labor-intensive, contain more grave goods than those on the periphery, and vary more in elaboration and wealth than more distant tombs (Beekman 2000:389-390).
Semi-peripheral Sites
Beekman notes that the more distant sites within the semi-periphery include a few guachimontones with unusually large residential buildings or a single disproportionately large building on the edge of a guachimonton. Semi-peripheral sites also more frequently include shaft tombs underneath guachimonton buildings, all of which reflect self aggrandizing behavior and an increase in emphasis on descent groups (Beekman in press a:70,71).
Six especially distinctive sites in passes overlooking the valleys have the greatest exposure to outside groups (Beekman 1996a,b ; Weigand and Beekman 1998). Four of
69


these sites are concentrated in the La Venta corridor to the east, and include Estolanos Mesa, Mesa El Zacate, Cerro Tepopote and Penol Tepopote. The remaining two sites, Cerro Pipiole and Llano Grande, are in Magdalena basin valley passes to the west (Beekman 1996a:984-985, 1996b). Beekman suggests that all of these sites were established to monitor their corridors as trade and communication routes. Based on the scale, substantial nature of the architecture and amount of artifacts and debris at each location, the sites appear to be full-time settlements which range in size from just 5-6 architectural features at Estolanos Mesa, to 120 features for the largest site, Penol Tepopote. All four of the La Venta sites contain wall features that either cross the pass or surround the site (Beekman 1996b: 139-140), as does Llano Grande (Beekman 2001:4-5). No wall structures are mentioned for Cerro Pipiole, but the site has yet to be examined in detail.
The appearance of walls in semi-peripheral areas is one of a few archaeological clues that can be found of a group's attempt at defining social boundaries (Chase-Dunn and Hall 1991:18), and Beekman asserts that these semi-peripheral sites form a controlled boundary surrounding the core area (Beekman 2000:404).
Beekman has also noted that the architectural and functional redundancy of guachimontones throughout the Teuchitlan culture suggests more of a segmentary state than a unitary one, as each site shows emulation of the largest core sites and functional independence, rather than a functional specialization that requires centralized control expected from a unitary state (Beekman 1996b: 136). Yet, if the walled semi-peripheral sites do indeed represent boundaries, then core elites may have attempted to increase their power by encompassing and managing the semi-peripheral areas as part of a conversion
70


of the political system to a more unitary state. If the area had become more centralized, then economic activity in semi-peripheral areas would have been directed towards the core, rather than external groups. As Kowalewski, et al. have pointed out, such transformations may not always be completely successful due to the required resources involved in controlling remote areas (1983:37), especially for remote locations which are less accessible across steep hills, marsh-ringed lakes and arroyos.
A few semi-peripheral sites also show differences relative the core in the proportions of guachimontones themselves. These circles include a single enlarged building which may reflect one descent group becoming dominant and gaining control over the ritual aspects of the entire group (Beekman in press a:64). This also reflects a move towards a vertical structure that favors aggrandizing elites, and likely accompanies an increase in wealth/trade aspects of the local economy. Weigand (1985:66) has stated that the most elaborate shaft tombs were located in the peripheral area of the Atemajac Valley east of the Tequila Valleys, and that the remaining tombs he had discovered by 1985 were "3rd rank", meaning less than 4 m deep and only one chamber Four of five sites known to contain shaft tombs beneath guachimonton buildings are in semi-peripheral areas (Beekman in press a:70). The guachimonton tombs are larger and more elaborate than those found in cemeteries. They also contain more variable and numerous grave offerings, and are more often reused for multiple interment. (Beekman in press a:62). Among the more elaborate semi-peripheral area examples is the tomb at Huitzilapa with an 8 m shaft, and a total of approximately 60,000 artifacts (Weigand and Beekman 1998:39). Beekman (in press a:62) suggests that kin relations were more emphasized at these sites, along with descent related claims to ceremonial positions related to the public
71


architecture, whereas cemetery tombs more consistently found in the core related elites to corporate control of labor and land capital.
Peripheral Sites
Peripheral guachimontones are underreported due to a lack of comprehensive, formal surveys in West Mexico, but so far two to three dozen sites have been found in Bolanos Canyon, several have been discovered in southern Jalisco and Colima, and one has been located in Puerto Vallarta (Beekman 1996a:88,89). These sites range from approximately 50 km to at least 200 km from the core. Guachimonton circles have also been found in the state of Guanajuato (Beekman 2000:figure 10, 2003b:5).
Peripheral sites tend to be located at strategic positions along transportation routes on the way to desired resources (Beekman 2003b:5). Peripheral and semi-peripheral areas were very sparsely populated, and Beekman has suggested that they likely emphasized exclusionary/networking strategies due to a lack of a labor pool to effectively implement a corporate strategy (in press a:73). Variation from the core architecture is most evident in peripheral sites. They have a much more narrow guachimonton size range, are not as well constructed and are the only Teuchitlan sites with ballcourts. The Tequila Valleys are also more agriculturally rich than any of the peripheral areas, and is the only area that shows evidence of intensive agriculture practices (Beekman 1996b: 143). Peripheral sites in Bolanos Canyon appear to have incorporated local architectural styles into the standard concentric circle pattern well established in the Teuchitlan core. Beekman interprets these remote sites as local elites who have incorporated the guachimonton architecture through
72


the regional trade system in order to create an alliance with the Teuchitlan core, and likely adopted some elements of the Teuchitlan culture (1996b: 143, 2000:400-401, 2010:64).
The remote sites to the northwest in Bolanos Canyon show the most variation from the core guachimontones. Hrdlicka described and illustrated the site of Totoate and other, similar sites as a circular layout with four square mounds surrounding a square central mound. Three of the four satellite mounds appear quite uniform, but one mound to the northwest is of an unusually oblong shape with a much smaller mound directly in front of it (1903:392-393,figure 9).
A few peripheral excavated guachimonton circles contained burials under the central altar (Cabrero 1989:149-161, 187-195; Hrdlicka 1903:392-395; Kelley 1971: 770-771), which strongly suggests that a single aggrandizing descent group may have claimed ritual authority over its entire community. Figurines from Ixtlan del Rio of the Nayarit highlands show apparent elite rulers and/or religious leaders, one male and one female, holding scepter-like objects which were probable symbols of authority. The top of the male figure's object depicts an apparent abstract model guachimonton that includes the center pyramid, circular patio and four surrounding temples (Anawalt 1998:Fig. 1; Beekman 2003a:313).
Although the various guachimonton activities would have been a binding force of the corporate system, they may simultaneously have also contributed to its undoing. The interactions take place in the context designed to bring separate and normally competing institutions together. Therefore, all corporate gatherings designed for unity likely included a layer of negotiation and political positioning as a manifestation of Turner's
73


arena (1974:133). Social manipulations may involve subtle, seemingly casual strategies within festive settings, or more formal means, such as the funeral processions which involve parading grave goods of a particular lineage in sight of all participants, as a display of the lineage's wealth and potential influence (Beekman in press:62). Over time, certain groups may have become more prominent, at least in part by utilizing better access to wealth economies to out-compete other groups, especially in areas with less potential for staple finance.
Jackson's (1984) description of vernacular architecture as "usually small" and "irregular in shape" certainly matches the observed pattern in semi-peripheral and peripheral sites of the Teuchitlan culture, especially when compared to those of the Teuchitlan core, and Jackson suggests that such sites are "subject to rapid change in use, in ownership and in dimensions" (1984:150-151). We do not know what changes may have occurred within the semi-peripheral sites following their establishment, but they were likely dynamic entities which underwent continual change in their political and economic identities from both core and external influences. Thus far, interpretations of the difference in use, ownership and dimensions of semi-peripheral and peripheral sites from those within the core have favored a greater emphasis on networked wealth economy of emerging dominant descent groups. The difference can still be seen as one of degree rather than type, since the relative influence of more powerful elites within a site may vary, along with the degree of emphasis on alternate economies to which they may be linked. Some semi-peripheral and peripheral guachimontones vary from those in the core, they are still very much recognizable as guachimontones and thus still represent some degree of adherence to a corporate ideal, where a slight modification to Orwell's
74


adage that "all descent groups are equal, but some are more equal than others" may well apply.
Beekman has recommended that "further research, utilizing extensive excavations, should focus on this variation [in core and periphery political strategies], as well as on the multiple roles of boundary settlements" (1996a:998), and the present research can be viewed as a step towards the goal of understanding the role and nature of the Teuchitlan semi-peripheral sites.
Descriptions of Compared Sites
West Mexican archaeologists have very generally described the region as rich in certain agricultural and mineral resources. Yet, very little specific data on geology, soils and land use have been published for West Mexican archaeological contexts. The availability and quality of natural resources can change drastically between distant sites within an area as complex as the Tequila Valleys, so economic opportunities at the site level need to be assessed with consideration of very localized natural resource details. Environment details were therefore examined for the primary core site (Los Guachimontones) and the sites under test (Llano Grande and Navajas), and the pertinent data are summarized below. Land resource descriptions are taken from official Carta maps of the specific areas in question with some observations made at each site (excluding any at the Sayula Basin) during the 2011 summer lab season.
Elevations are also compared between sites to determine whether differences are large enough to effect available resources. Since elevation can profoundly affect available
75


resources, both topographical maps and an interactive tool (http://www.al titude-maps.com/) were used to find elevations and sloping trends where they are not specified in the literature. The altitude-maps tool finds elevations at more precise points than topographical maps can provide, but uses low resolution SRTM data which can show error rates of up to 23m on steep slopes (Thomas, et al. in press; Tighe and Chamberlain 2009). The resolution is not as high as desired, but suffices for the current purpose.
Feature and layout details of the central site (Los Guachimontones), and also sites under test (Llano Grande and Navajas) are given here to better characterize the sites within the context of the Teuchitlan culture, and also to provide some context for the excavated features (Llano Grande Feature 14, and Navajas Circle 5) and the analysis of the lithic artifacts recovered from those features. The lithic analysis can then be used to further describe the economy of each site relative to the ecological and cultural conditions that the site represents.
Los Guachimontones
If some level of centralized control over semi-periphery sites such as Llano Grande did exist, it would likely have come from the primary site of Los Guachimontones just south of the Tequila Volcano, and near the center of the core (figure 1.2). The site likely influenced the entire core area at least culturally if not politically, and therefore is described in some detail here.
76


Site Environment
Los Guachimontones is on a hillside near the edge of marshland within the La Vega lake basin. Currently, Los Guachimontones contains mainly seasonal farmland to the immediate south, and mainly brush land to the north and west. During the Late Formative and Early Classic periods, much of the land immediaely below the Los Guachimontones settlement area to the south was marshland (Beekman personal communication, September 2013). Most of the soil in the area is of mixed quality for farming, except for an area about 1 km to the southwest, off the northwest shore of the lake, with deep, humus rich soil. On the opposite side of the public ritual center from the lake is a bank which runs southwest to northeast, directly behind the largest circle. Behind the bank is another area of very good soil, which may have been partly used by another Teuchitlan site on the opposite side of the field (figure A.7).
The elevation of the ritual center is 1,374 m, taken at the center of Circle 1. Over the bank above the large circle to the northeast, the field slopes downward away from the bank from southeast to northwest, and from an elevation of approximately 1,460m to 1,370m.
Description
The site's ceremonial center consists of ten circles and two ballcourts, with what has been described as a "miniature shaft and chamber tomb" in one of the older and more moderately sized circles (Beekman 2008:426-428). Los Guachimontones is by far the largest site in the region(Beekman 2008:427). Construction details are described as being similar to Navajas, although the difference in construction actually appears more similar
77


to what is found in Navajas Circle 5 with differences in construction material and method details for each building.
The public architecture is quite large in proportion to the size of the residential area, suggesting that it was a more central administrative center, possibly controlling a second tier of smaller major centers within the valleys (Ohnersorgen and Varien 1996:fig. 9), and the core population density also suggests that Los Guachimontones was part of a larger system, which Weigand suggests controlled six habitation zones via their respective administrative centers in the core area (Ohnersorgen and Varien 1996:103). Weigand's assertion again brings up the question as to how far Los Guachimontone's control actually extended beyond the south central core area.
Ohnersorgan and Varien (1996) borrowed a model successfully used by geographers to determine virtual "boundaries" for segmentary groups, and also tested by Alden (1979) on Aztec group data, to determine the potential political reach of Los Guachimontones. The model depends largely on the distance between sites, and the size of each site. Llano Grande wasn't included in Ohnersorgen and Varien's study, but closer and larger groups were determined to be "isolated clusters" in scenarios where the culture's willingness to travel long distances was similar to Alden's Aztec case. Therefore, Llano Grande would likely not have been included as a centrally controlled site according to the model. Navajas is half the distance to the cultural core area and on an easily accessed plateau so it is more likely to interact with, and therefore conform to, the core area.
Even though the concept has been used successfully to determine spatial limits of segmentary groups in geographic contexts, the model has not been subject to repeated testing for an archaeological context, the results should be treated with some caution.
78


Also, Ohnersorgen and Varien recognize the lack of any interpretation of the nature of the interaction between sites, and suggest that clarifying craft production/distribution systems, perhaps particularly in the especially prominent obsidian industry, can go beyond descriptions of the structure of the interactions to determine more detailed information about their nature (1996:119).
Navajas
Site Environment
Navajas is located in center of current farmland on a gently sloping, round, nearly circular plateau of bedrock about 1 km in size, surrounded by a high-altitude plain. The site is also just outside of the semi-peripheral area hills but with very close access to the core near a pass (Figure 1.2). The Navajas guachimontones are on shallow bedrock, but agriculturally rich fields of marginal to ample depth form an expanse similar in area to the arable fields around Los Guachimontones. Arable land extends north from the ceremonial center, and also exists west of the site. Some poorer, loose and granular soil is intermixed in the northern half of the field. The rough circle of bedrock is mainly andesite, an extrusive igneous volcanic rock. The elevation at Navajas' Circle 5 guachimonton is 1,538 m, with a gradual slope from 1,592 m down to 1,458 m across the 2 km bedrock circle diameter (approximately 1 ), from the southwest to the northeast.
79


Description
The site of Navajas was excavated during the 2003 field season. The site is located in an area just outside the Tequila Valleys (Beekman 2008:424), on a plateau in the small basin surrounding Lake Valencia.
The site complex contains at least 10 guachimonton circles, one ball court 85m in length, scattered apparent residence buildings arranged in clusters of three to five structures, terraces, cemeteries, and satellite sites with their own guachimontones. All of these features are within 2 km. of the main center at Circle 1.
The guachimonton used for this thesis, Circle 5, was built just after the estimated time of occupation for Circle 1 (ca. 50 B.C. to A.D.50 for Circle 1, and ca. A.D. 50 200 for Circle 5) (Beekman 2007, 2008:424; Beekman and Weigand 2008:308,310), (figure 3.4). Circle 5 can also be seen as an early contributor to the Early Classic trend of downscaled guachimonton architecture, whereas the scale of Circle 1 is typical of its time within the Late Formative. Circle 5 contains eight perimeter platforms, and shows uniform size and shape except for one smaller building. The buildings are also positioned somewhat irregularly around the circle forming a lack of exact symmetry, with distances between buildings that vary from about 1 to 4.5 m, and each building shows different construction design and materials (Beekman 2008:423-425). The Circle 5 patio also forms a somewhat irregular circle (Beekman 2008:424). The total diameter of the guachimonton circle measured to the back walls of each opposing building is 36.0 37.9 m (Beekman 2008:424), and the diameter of the altar is 7.35 7.55 m. Using median diameter values as an approximation, the total activity area (the total area excluding the area of the altar) is about 1031 m2
80


Figure 3.4: Navajas central group with areas of excavation circled. The guachimonton used in this thesis, Circle 5, is the northern-most circle.
Both Johns (2014) and Tyndall and Beekman (2007) conducted distributional studies of ceramic artifacts from Circle 5 to determine the nature of activities related to ceramics. Both studies found that relative proportions of the different types of vessels were essentially similar throughout the circle, suggesting a communal, corporate structure based form of ritual. However, both studies also note that overall counts of sherds varied
considerably between buildings, suggesting unequal participation between groups. Ceramics recovered from Circle 5 include a total sherd count of 9,475 (Johns 2014:83),
81


which "far surpassed original expectations in terms of quantity and quality" (Tyndall and Beekman 2007:154), along with several complete and partial vessels and bowls. Based on the previous evidence for feasting depicted in ceremonial ceramic forms throughout the Teuchitlan culture (Butterwick 1998) as well as an abundance of storing and serving vessel forms at the circle (Johns 2014: figures V-12, V-13; Tyndall and Beekman 2007:174), the ceramic containers are interpreted as implements for corporate feasting. Although feasting was apparently still practiced during the Circle 5 occupation, at least two groups may have begun to dominate the rest in the corporate power field, utilizing the guachimonton ritual setting as an arena for political maneuvering. The outcome may have ultimately led to increased power of aggrandizers within the dominant groups, and a related expansion of economic control to include areas beyond staple goods.
Circle 1 also possesses 8 perimeter platforms, and Circle 1 buildings also vary in size, but in a consistent alternating pattern between large and small buildings. Within each respective size group, the large and small buildings also approximately equal each other in size, and the buildings are also much more regularly spaced around the platform. The perimeter buildings are constructed using similar materials and methods, with only a noticeable difference in the shapes of each building.
Again following the theory that the spatial relationships between buildings reflect the social relationships between lineages that built and operated within them, Beekman suggests thatNavajas Circle 1 may represent a time of well-defined group relationships with major and minor groups uniformly and distinctly represented, whereas Circle 5 represents shifting relationships between groups. Just as variation in the Circle 5 architecture is limited to the basic guachimonton pattern, however, changing lineage
82


relationships are also seen as constrained within certain limits by the corporate code (Beekman 2008:426). The more uniform structures at Navajas are also very similar to those within the nearby internal core zone, and Navajas therefore appears to be both culturally and socially associated with the core.
Within the Navajas guachimonton, two plant species suspected to represent contemporaneous sources of subsistence were located in soil samples from activity floor surfaces, maize and huauzontle. The latter is a plant often cultivated for the use of its seeds and leaves in Mesoamerican cooking and medicine (Benz 2007:248). The huauzontle seeds are of some interest because they are confirmed to be of a domesticated variety, and the only other Classic Period incidence of domesticated huauzontle known to Benz is in Teotihuacan, the closest major Mesoamerican polity to the Tequila Valleys. It is not clear whether the huauzontle represents any sort of link with Teotihuacan, but it could be related to trade or some other social tie with the polity. Four obsidian tools registered positive tests for the presence of proteins which may have been food related. Three scrapers were positive for small animal proteins (species related to rabbits and guinea pigs), and one projectile point registered a grass protein which may have been maize, although no particular species were resolved from the tests (Parr 2007:225).
Llano Grande
Site Environment
Llano Grande's location is currently in a narrow tract of semi-open pastureland sparsely populated by wild Holm oak groves, surrounded by more dense Holm oak forest
83


land. Holm oak is an especially hardy tree variety which consists of several species of the genus quercus, which tolerate sandy soils and semi-arid environments (Botanicals Online 2014). The pass opens up approximately 1 km to the east into a narrow valley. The east end of the valley connects with the Magdalena Basin. The west end of the valley, closest to the site, is sloped pastureland highly subject to erosion which continues for about 200m as one travels east from the pass. Beyond this area in the valley below the pass is primarily part-time or seasonal farmland. The site is thus less than 1 1/4 km from arable land, although the pass drops down a hill into the valley. The closest water source in the valley may be 3/4 km from the west end, where a spring extends further east into the remnant Lake Magdalena. Several other springs provide water to the valley, mainly near the east end of the pass. A recent area geological study has determined that the northeast end of the valley would have been a marsh sourced by the current spring (Beekman personal communication August 2013). Thus, during the rainy season, Llano Grande would have had a 1.2 km distance to cropland. If the land was farmed during the dry season, water from the spring marsh would need to be utilized about 2 km from the site.
Land differences between the valley and the pass at Llano Grande are quite pronounced, and even assuming an adequate water source would have been available, there would have been no chance of growing any type of crop in the immediate vicinity of the pass. Although the valley contains entirely alluvial soil, the canyon is composed entirely of acidic (high-silicon) extrusive igneous rock. The canyon's soil type is considered Regosol, a sparsely distributed, residual soil which does not completely cover bedrock, and therefore has no appreciable depth. The valley soil type is a humus-rich, highly arable pheozim soil, in all areas except the southwest section which contains a
84


high clay soil type considered quite difficult to farm. Throughout the valley, the soil is more than adequately deep for agricultural purposes, although the soil in the sloping and eroding west end is much more shallow than the lower areas of the valley, at around 10-50 cm. The slope and degree of apparent erosion may have precluded farming in that section of the valley. The rest of the valley gradually slopes from 1,370 to 1,420 m from the Magdalena basin to the western end of the alluvial soil area. The Llano Grande guachimonton sits at 1,487 m in the center of the pass, which is essentially flat and the areas towards the margins only vary in elevation from 1,480 to 1,500m.
Botanical remains were located within the Llano Grande guachimonton, which indicate agriculture. These include charred cobs and pollen from maize, starchy seeds which could have been corn or beans, along with bottle gourd, cotton, and possibly tomato plant pollen (Schoenwetter 2004:4). Phytoliths analyzed from food residues of ceramics (four ollas and one bowl) primarily indicate maize, and also fruit from the genera Prunus and Celtis (Schoenwetter 2004:4,5). Schoenwetter states that these results indicate agricultural production at Llano Grande, even though the site "was not embedded in an agricultural landscape" (Schoenwetter 2004:1).
What Llano Grande lacked in agricultural potential, however, may have been well compensated for by the site's mineral rich environment. The site overlapped a large obsidian quarry (Beekman 2001:3; Spence et al. 2002) with a wide variety of color patterns (Spence, et al. 2002). Obsidian is ubiquitous throughout the site (Beekman 2001). The most abundant variety at the site is a dark green obsidian with very few inclusions and excellent flaking characteristics for the manufacture of stone products. Llano Grande's mineral resources also include red ochre and opals (Weigand 1985: 90),
85


although there is no definitive evidence that the latter two minerals were exploited by the Teuchitlan culture.
Like Navajas, Llano Grande has a direct route to the core directly across the valley floor, but Llano Grande is on the other side of the Lake Magdalena basin and is twice as far from the core than Navajas, at 20 km (figure 1.2). To a large extent, the lake and marsh levels would have further buffered the already geographically secluded Llano Grande against direct interaction with the core, and therefore also against centralized control. Although average lake levels during the Late Formative and through Epiclassic periods were steadily receding, The shoreline of Lake Magdalena reached the eastern mouth of the valley, which would have required either a longer route on foot or travel via watercraft to access the core (cf. Anderson et al. 2013:25). Thus far, no evidence of watercraft has been located within the valleys. If canoes were used the situation would have been alleviated to a large extent. However, ethnohistoric data regarding contact period canoe travel suggest that prehistoric transportation over water was 33 percent slower than land travel, and therefore increased the required travel time by an additional half of the time to travel the equivalent distance by land (Alden 1979:175). Also, the presence of a spring-fed marsh at the northeast end of the valley would have hampered travel between Llano Grande and all sites north of the core (Beekman personal communication 2013). Llano Grande was not completely cut off from the core, but a much longer route including travel through the hills would have been required. The increased travel difficulty would also increase the resources required to maintain regular interaction between the core and Llano Grande. The core may therefore have been unable
86


or unwilling to fully assimilate Llano Grande into the Teuchitlan social system (cf. Kowalewski et al. 1983:37).
Given Llano Grande's location in a mountain pass, one may expect a higher elevation than Navajas' location in a lake basin outside the semi-peripheral hill area, but the plateau on which Navajas sits is actually 51m higher in elevation than Llano Grande. Los Guachimontones, however, sits 113 m below Llano Grande. Elevation is therefore considered negligible as an environment factor that could affect natural resource limitations between the two sites. Both areas also have access to good soils. However, Navajas has more expansive land resources, but with somewhat lower quality soil about 2.5 km away from the site. The topography and geology surrounding each site are categorically different, as is each site's specific location relative to staple resources. Navajas Circle 5 is adjacent to its highest quality farmland, whereas Llano Grande's guachimonton is approximately 2 km from any viable cropland.
Los Guachimontones' and Navajas' positions in open areas and adjacent to fertile soil may have been related to similar staple oriented economies, whereas Llano Grande's contrasting distance from agricultural resources suggests that the site may have diverged economically from the two core sites. However, both Llano Grande and Navajas show evidence of some degree of staple economy within their guachimontones, in the form of macrobotanical remains and residue analyses which reveal staple processing or storage.
Description
Llano Grande is situated in the center of a pass which opens into a valley off the Magdalena lake basin, northwest of the core (figure 1.2). Initial architecture surveys
87


Full Text
Alban. The study confirmed that although the trajectory of the three variables within the region is inconsistent in the short term, the long-term trend is towards decentralization of the polity. The correspondence between size, centralization and permeability only aligns with the Kowalewski hypothesis for 5 of 11 organizational phases. A positive correspondence between size and permeability, however, is indicated for 9 phases. Therefore, the level of centralization does not appear to be a strong indicator of boundary permeability at Oaxaca, but the relation between permeability and the number of internal groups is consistent for over 80 percent of Kowalski et al.'s cases.
Exceptional cases noted by Kowalewski occur at two extremes, which reveal patterns as notable as the overall results. In one case, the system size shrinks to a point where the distance between groups is too great and the amount of inter-group interaction drops too low due to declining populations. As a result, so the price of integration can no longer be met. The polity then becomes fragmented and really no longer exists as a unified system. The remaining, newly-independent groups seek additional resources across what has essentially become a defunct boundary (Kowalewski et al. 1983:49). The situation also allows for the possibility of a redefinition of boundaries around multiple smaller polities, which may then also begin to interact. The second situation occurs when the increase in size is so great that it requires a complete reorganization, and elites once again turn inward to re-establish groups via staple resources until a new system is well established (Kowalewski et al. 1983:49).
Although the Valley of Oaxaca case reveals local trends that may be important to future studies, the authors stress that the patterns found in Oaxaca have not been demonstrated elsewhere. They suggest expanding investigation along the same lines to
32


other areas to determine how far beyond Oaxaca similar results might continue to hold (Kowalewski et al. 1983:49).
Network and Corporate Strategies
Differences in economic bases between different groups within a culture may lead to differences in the underlying power strategies employed by elites to build and maintain power. Richard Blanton and colleagues (1996) have recognized a distinction between basic strategic directions which he terms networking/exclusionary and corporate strategies. Elites engaged in a networking/exclusionary strategy establish relationships with specific contacts within a network of other elites, with whom they share and exchange resources which differentiate them from non-elite groups. Network affiliates may be formed along pre-determined parameters, such as kin or ethnic group, or purposely selected based on some criteria, such as access to specific resources. The network strategy involves a large degree of control of access to specific prestige goods requiring skill specialization, especially in peripheral areas, which may see a large jump in specialized production.
Network strategies also tend to foster innovation of product types, although an element of secrecy may slow the spread of some innovations. Production techniques which allow for less specialized skillsets to produce prestige items are not commonly spread, as the effect would be to "banalize" the product by making it common and universally available, and therefore lose all value as a prestige item. Given enough time, however, production innovations, "secrets" and/or source connections may slip to the
33


CHAPTER IV
LITHICS BACKGROUND
This chapter discusses lithic forms and relevant lithic studies within West Mexico, and primarily within the Tequila Valleys. Lithics in other selected parts of Mesoamerica are also discussed, and are utilized along with the West Mexico data to form the basis for expectations regarding wealth and staple industry related products further described in Chapter V. As Jay Johnson asserted in his study of the Mayan obsidian economy, "a clear understanding of what actually is being made is necessary before hypotheses about state level economics can be tested" (1996:171). In order to determine whether, and to what extent, lithic assemblages within semi-peripheral sites were likely a result of wealth trade with external groups rather than agricultural, food preparation and/or other common functional activities, it is necessary to determine which forms tend to be utilized for each of these purposes. Some quantities of prestige forms produced locally were also likely used for ritual purposes, so context becomes especially critical to interpretation. Distribution studies can then describe not only what was made, but what portion of the process may have been included in source, production and use contexts.
This chapter is divided into two main sections: Descriptions of lithic forms related to both elite prestige and general purpose utilization; and distributions of various forms related to local, regional and long distance trade. The primary site under test is in a semiperipheral zone which may have maintained economic relations with core populations, external groups or both. Therefore, these two sections are further divided into subsections on internal Teuchitlan data, and external group data. Some external groups might
98


Benz, Bruce F.
2007 Preliminary Analysis of Archaeological Material from Bulk Sediment Samples. In Tequila Valley Regional Archaeological Project Season Report 2003, edited by Christopher S. Beekman, pp. 248-255. Manuscript on file,
Anthropology Department, University of Colorado Denver.
Blanton, Richard E. and Gary M. Feinman
1984 The Mesoamerican World System. American Anthropologist 86(3): 673-682.
Blanton, Richard E., Gary M. Feinman, Stephen A. Kowalewski and Peter N.Peregrine 1996 A Dual-Processual Theory for the Evolution of Mesoamerican Civilization. Current Anthropology 37(1): 1-14.
Bleed, Peter
2001 Trees or Chains, Links or Branches: Conceptual Alternatives for Consideration of Stone Tool Production and Other Sequential Activities. Journal of Archaeological Method and Theory 8(1): 101-127.
Botanicals Online
2014 Holm Oak (Holm oak characteristics). Electronic source, accessed online from http://www.botanical-online.com/english/holmoak.htm on October 4, 2014
Boeda, E.
1995 Levallois: A volumetric construction, methods, a technique. In The Definition and Interpretation of Levallois Technology, edited by H. Dibble and O. Bar-Yosef, pp. 41-68. Prehistory Press, Madison.
Bourdieu, Pierre
2002 Social Space and Symbolic Space. In Contemporary Sociological Theory, edited by Craig Calhoun, Joseph Gerteis, James Moody, Steven Pfaff, and Indermohan Virk, pp. 267-275. Blackwell Publishing, Oxford.
Braswell, Geoffrey E., E. Wyllys Andrews V, and Michael D. Glascock.
1994 The Obsidian Artifacts of Quelepa, El Salvador. AncientMesoamerica 5: 173-192.
Braswell, Geoffrey E., Iken Paap, and Michael D. Glascock
2011 The Obsidian and Ceramics of the Puuc Region: Chronology, Lithic Procurement, and Production at Xkipche, Yucatan, Mexico. Ancient Mesoamerica 22(1): 135-154.
Breton, Adela C.
1903 Some Mexican Portrait Clay Figures. Man Volume 3, pp. 130-133. The Anthropological Institute, London.
252


Clark, John E.
1981 Towards a definition of workshops. Paper presented at the Obsidian in Mesoamerica Symposium, January 26 to 31, edited by Henry Pratt. Fairchild, Pachuca, Hidalgo, Mexico.
1982 Manufacture of Mesoamerican Prismatic Blades: An Alternative Technique. American Antiquity 47(2): 355-376.
1987 Politics, Prismatic Blades, and Mesoamerican Civilization. In The Organization of Core Technology, edited by Jay K. Johnson and Carol A.
Morrow, pp. 259-284. Westview Press, Boulder.
1988 The Lithic Artifacts of La Libertad, Chiapas, Mexico: An Economic Perspective. Papers No. 52, The New World Archaeological Foundation, Brigham Young University, Provo, UT.
1997 Prismatic Blademaking, Craftmanship, and Production: An analysis of Obsidian Refuse from Ojo de Agua, Chiapas, Mexico. Ancient Mesoamerica 8(1): 137-159.
Clark, John E. and Douglas Donne Bryant
1997 A Technological Typology of Prismatic Blades and Debitage from Ojo de Agua, Chiapas, Mexico. Ancient Mesoamerica 8(1): 111-136.
Clark, John E. and Phil C. Weigand
2009 Obsidian lapidary without polishing. In Investigaciones Recientes sobre la Litica Arqueologica en Mexico, edited by L. Mirambell and L. Gonzalez Arratia, pp. 79-93. Coleccion Cientifica, no. 561. INAH, Mexico City.
Clarkson, Chris
2008 Changing Reduction Intensity, Settlement, and Subsistence in Wardaman Country, Northern Australia. In Lithic Technology: Measures of Production, Use, and Curation, edited by William Andrefsky, Jr., pp. 286-316. Cambridge University Press.
Cowgill, George
2000 The Central Mexican Highlands from the Rise of Teotihuacan to
the Decline of Tula. In The Cambridge History of the Native Peoples of the
Americas. Volume II: Mesoamerica, Part 1, edited by Richard
E.VI. Adams and Murdo J. MacLeod, pp, 250-317. Cambridge University Press.
Curtin, Philip D.
2000 The World and the West: The European Challenge and the Overseas Response in the Age of Empire. Cambridge University Press.
254



PAGE 1

CIRCLES OF GLASS AND GRAIN: ECONOMIC D IFFERENCES BETWEEN C ORE AND SEMI PERIPHERAL ZONES A STUDY OF PUBLIC CENTER L ITHICS FROM THE TEQUILA VALLEY S OF WEST MEXICO. By JOHN P. WAGNER B.A.S. Lake Superior State University, 1989 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in p artial fulfillment o f the requirements for the degree of Master s of Arts Anthropology 2014

PAGE 2

ii This thesis for the Masters of Arts degree by John P. Wagner has been approved for the An thropology Program by Christopher Beekman, Chair Julien Riel Salvatore Tammy Stone December 5 2014

PAGE 3

iii Wagner, John P (MA, Anthropology) Circles of Glass a nd Grain: Economic D ifferences Between Core a nd Semi Peripheral Zones, a Study of Public Ce nter L ithics f rom t he Tequila Valley s of West Mexico Thesis directed by Associate Professor Christopher Beekman ABSTRACT The dynamics of expanding polities and relationships between cultural core groups peripheral populations and sites in semi peri pheral areas between the two groups are important topics in studies of complex societies. One area where these distinctions are clearly identified within the settlement pattern formed by the relationship between the cultural and the natural landscape is th e Tequila Valley s of Western Mexico T he Teuchitln culture of the Late Formative and Early Classic period s formed distincti ve settlements around the edge s of the v alley s which were also marginally bound to most complex social developments within the cult ural core region near the center of the valley s Semi peripheral sites between cultural traditions are of particular interest as focal points for economic, political, and social relationships. This thesis focuses on two sites w hich occupied very different environments, namely Llano Grande and Las Navajas I ask whether these sites show different degrees of emphas is on two basic economic strategies in ways which capitalized on the advantages of each site's respective environment S pecifically did Llano Grande's relative physical isolation from the cultural core area, more distant location and differences in available resources reflect a greater reliance on trade via exported obsidian ? Alternatively, did Navajas' closer relation to the core al low a continuance of the core 's degree of emphasis on the staple oriented economy with less emphasis on obsidian production and trade than Llano Grande ? This thesis draws upon the work of Earle (19 91 ) to structure the analysis particularly his

PAGE 4

iv contrast b etween wealth and staple finance. Past research is reviewed to develop expectations for each model, which are tested using the a nalysis of obsidian debitage and products within the ritual centers of each site The form and content of this abstract are app roved. I recommend its publication. Approved: Christopher Beekman

PAGE 5

v ACKNOWLEDGEMENTS This thesis required m u c h time and effort, and I am very grateful for the support, patience and persistence of the faculty and everyone else involved in all aspects of its creation. I would first like to thank my advisor, Dr. Christopher Beekman for the opportunity to work in his research area, as well as his tremendous attention to detail, added information, reviews and advice which have been invaluable in the process of refining the composition of this thesis. I also wish to thank Dr. Julien Riel Salvatore for his advice on the lithic analysis, and in particular his input on research design for datasets which have proven quite challenging to compare. I w ant to thank Dr. Tammy Stone as well for her advice on the statistical portions of analysis, her overall advice and perspective on the purpose and focus of the thesis. S ome colleagues have also been tremendous assets, and I especially want to thank Lu cas Hoedl, Catherine Johns and Nichole Abbott for their sharing of information and companionship over two lab seasons in Mexico, and also Tony DeLuca who also shared valuable new information on the latest surveys in the region of study. I want to also than k Dr. Verenice Heredia of the Colegio de Michoacn for providing the use of lab facilities, as well as taking care of the logistics for extended analysis during the 2012 lab season. Finally, I wish to extend thanks to Camilo Mireles for his insight into so me of the more curious artifacts in the area of study for this thesis. Several people outside of the academic setting have strongly encouraged me to pursue this very different path from my technical career background, and a few have encouraged me thro ugh the process of writing this thesis. I wish to thank my siblings, Deborah

PAGE 6

vi Hentschel and Kristopher Wagner, for encouraging me to see this process through to completion. I would also like to extend thanks to two additional co laborers in the thesis writi ng process who have also been sources of encouragement: my niece "in law" Andrea Hentschel and Kathleen Chambers. I would also like to thank my good friend and natural sciences professional Mark Reichel for his interest, occasional inquiries, informative i nsight, discussions and further encouragement. Finally, to my parents and all others in my circle whom I have not already mentioned, thank you for your support and patience with my work on this thesis, which has absorbed so much of my time and energy over the last few years.

PAGE 7

vii TABLE OF CONTENTS CHAPTER I. INTRODUCTION ........................................................................ ... ..........................1 The Tequila Valleys and the Teuchitl n Culture ......... ........... ..... ... ... ....... ..............1 Overview of Chapters .............. ............................................................ ... ................... 8 II. THEORETICAL DIRECTION ......... ....................... .............................. ... ... .. .........1 2 Core Periphery Relations ......... ....................... ........................................... ..... .........1 2 Outposts and Remote Groups .. ........................................................... ... .............1 6 Boundaries ..................................................................... ....... ................ ... .......... 21 Unitary States and Segmentary States ............................. ........................ ... ....... 2 3 Political S trategies ......................................................... ......... ......................... ... .... 25 Wealth vs. Staple Economies ............................................. ........................... ... .2 5 Wealth Relations and Bo undary Permeability ....................... ...................... ... .. 30 Network and Corporate Strategies ........................................... .. ........ ......... .... .. 33 Connecting Economy and Strategy........................................ . ........ ............. ... ..3 5 Institutional Competition .................................................. ...... . ..... ...... ..... ......3 5 Combining Economies: Context and Matrix Control ........ ....... . . ......... .... ......3 8 Struct ure, Economy and the Natural Environment ............ ....... .... ...... ..... ..... ........ 40 Examining Semi periphery Economies ... .................... ... ... ......................... .. ....... 43 III. REGIONAL BACKGROUND ............ ......... ....... ...... .. .. . ... ... .. .... .... ... .... ..... .. ..... 4 5 The Natural Environment and Subsistence ........... .......... .. ... .......... ..... .......... .... ..4 6 Geology and Climate ....................................... .. ............... ... ............. ..... ............4 6 Geographic Area Relations to Political Economic Zones ......... .. . .. .. .. ... ...... ...4 7

PAGE 8

viii Subsistence Resources ............................................. ........ ......... ...... ... .. ........... 50 Minerals ...................................................................... ...... ............. .... .. ............ 52 The Economic Impact of a Changing Climate................ .. ......... . .... ..... .. ......... 52 Economic Strategies with in the Teuchitln Culture ..... ...... .......... .... . ...... ... .. ..... . 53 Staple Finance ............. .......................................... ..... ...... .......... .. . ......... .. ... 54 Wealth Finance ............................ ......................... .... .. ............... ........ ........ .. .. 60 Economies and Ritual Space .... ........ ..... ............. ...... ............ ....... . .. .......... .. .. 65 T imeline ..................................... .......... ................... .................. ......... ..... .. .......... 66 Distribution of Architecture ............. ... ................ . .................. .... . .. ........... .. ........ .6 7 Core Sites ................................... ....................... ............... ............. ......... .. .......6 8 Semi Peripheral Sites ....................................... ................ ........... ... ........... .. .. ..6 9 Peripheral Sites ............................................ ......... ........... .... ....................... ... 72 Descriptions of Compared Sites ..................................... .... .... ......... ............. .. ....... 75 Los Guachimontones .................................................. . ........... ....... ........... .. ..... 76 Navajas ............................. ......................................... .................................. .. ...7 9 Llano Grande .................... ..................................... ............... . ....................... .. 83 Sayula Basin .......... ............................................. ..................................... .. ....... 92 Conclusion ..................... ............................. ..................................... .......... .. ..... 94 IV. LITHICS BACKGROUND. ................................................ ... ................. .......... .. ... 98 Overview of Mesoamerican Forms ..................... ........................ ........... .. ............ 9 9 Polyhedral Core Blades .. ......... . ................... .......................... ...... ....... .. ........ 10 2 Eccentrics ............................. .. ........... ................ .................................... .. ...... 105

PAGE 9

ix Overview of West Mexican forms .......... .................. .......... .......................... ..... 105 Pikes and Rod Ornaments ........ .............................. . .............. . ...................... 107 Laminar Flake Jewelry ........ ....... ... . .......................... ... ..... . ... .. . .................. 111 Distribution.............................. ..... ....................................... . .... . ...... .. ... ..............1 1 6 Workshops and Points of Exchange ....... ........................... . .... . ..... .. .. .............1 18 Challenges to Interpreting Inter polity Exchange ......... ............ ....... .. ..............1 19 Exchange in Mesoamerican contexts ..................... ................. . ......... .. .... ........1 20 Te uchitln: Internal Distribution.............................. ........ ....... . .......... .. ..........1 38 Area and Site Studies ............. .. ................ ........... . ........ ... . .... ................ .... .. .......1 47 Lo s Guachimontones ................................................... . . ....... ... .............. .. ..... 1 49 The La Venta Corridor .............. ...... ............................. . ........ .................... .. ...1 55 The Sayula Basin .............. ........... ... ........................... .. . .... ..............................1 58 Inter Area Comparisons ....................... ...................... .. ............ .. .....................1 61 Conclusion ........................................ ................................... .. ......... . ............... ....1 65 V. METHODS ..................... ...................................................... .. ........ . ...... ............ 1 68 The Llano Grande and Navajas Datasets ............ ...................... .. ........ ................... 1 68 Attributes ............................................................................... .. ... ..... ...............1 72 General Approach for Analysis ........................................ ............ .. . .....................1 81 Criteria for Comparing Staple and Wealth Lithic Economies ... .... .. .... ...............1 82 The Nature of the Deposits ............................ .............................. .. ..... ..... .......1 82 E xpectations and Tests for Economic Distinctions ......... .............. .. ... ..... ..... ..1 85 Import Considerations ....................................................... .............. ... ..... ......1 99 Outcome A n alysis .................. ............... ... ........................... ............ .. ............. 200

PAGE 10

x Conclusion ....................................................................................... . .. .. .. ........... 201 V I RESULTS ................................ ........................................... ....... ..... .. ............ ..... 202 Organization and Review of the Test Criteria .................................... . .... ............. 202 Lithic Type Proportion Comparison to Background Contex ts .......... .. . .. ............ 204 General observations on the Collections Under Test ................... ........... . ...... .... 207 Overall Quantities and Site Coverage Differences ................ .............. . ......... 208 The Natur e of the Deposits ......................................................... .......... ...... ........ 210 Lithic Industry Tests ...................................... ..................,................ ......... .. ....... 212 Test 1: Formal Products vs. Debitage and Expedient Tools ........... .... . ..... ...... 212 Test 2: Debitage to Ceramic Sherd Ratios ........................... ................ . ..... ..... 213 Test 3 : Core Totals ............................................................ ...... ............ .. . . .. .. .. 215 Test 4: Amounts of Dorsal and Platform Cortex on Flakes ................. . . ..... ... 216 Test 5 : The P ercentage of S pecialized T ools for S taple P rocessing ....... . ....... 2 20 Test 6 : The P ercentage of D ebitage A ssociated with P olyhedral B la d es .. . . ... 2 2 1 Test 7: The P ercentage of U nfinished or F ailed B ifaces .. .. ........ . .. . . ....... .....2 2 2 Test 8 : Emphasis on P roduction R elated to Laminar Flakes ..... .. .... ......... .... 2 2 3 VI I DISCUSSION & CONCLUSION ....... .......... ... ....................... .. .......... ....... .. . .. 2 29 The Origin and Purpose of Guachimontn Obsidian ................ . ........... .......... .. 2 33 Conclusion .................................................................. ................ .. ... ...... . .... ........ 2 35 Challenges L imitations and Lessons Learned ................................ .. .... ....... ...... 2 41 Future Opportunities ............................................................ ...... .... . . .. ........ .. .... 2 43 Continued Exploration at Llano Grande ......... ............ ..... .... ...... . ..... ..........2 4 4 Surveys and Additional Excavations ............ ..... ........ .... .... ............ .... ..........2 45

PAGE 11

xi New Questions with Expand ing Exploration............. ..... ................. ... .............246 Other Paradigms.................................... ................ .................. .. ....... ...............247 REFERENCES .......................................................... .. .. ................................. ............... 2 49 APPENDI X A. Carta Land Use and Soil Map sections ........ .......... ................................... . .......... 2 69 B Scatterplots of scar counts and size measurements ............ ........... ... .... .... ..........2 78

PAGE 12

xii LIST OF TABLES Table 3. 1: Site occupation spans and related periods relevant to the current study ...... . .. ...... 6 8 4.1 : Platform preparation percentages, center vs. workshop ................ ....... ... .. ...... ... ..1 51 4.2 : Lith ics from the Los Guachimontones public center ................ ..... ...... .. ... ........ ... 1 52 4 .3 : Lithics from the Feature 83 workshop, with used item percentages ... ... .. ............ 1 54 4.4 : Lithics from the La Venta Corridor (nodules excluded) ...... ...... ........... ...... ...... ... 1 57 4.5 : Percentages of each lithic type per location ................................................ ...... ... .. 162 5.1: Llano Grande size frequency distribution comparison for different flake types . .1 78 5.1: Navajas siz e frequency distribution comparison for different flake types ...... ......178 6.1: Summary of compared background l ocations . ............ . ...... .................................. 204 6. 2 : Intersite comparison of lithic proportions by percentage relative t o total item counts . . . . . . ............. ............. ........................................................ ....................... ....... . 204 6. 3 : Percentages of formal product counts relative to total analyzed sample . .. . . . ...... 21 3 6. 4 : Debitage and Sherd Total Comparison ........................ . .................................. ...... 213 6. 5 : Core quantities as a percentage of debitage at each site ............ . ......... .............. ... 215 6. 6 : Site percentage s of flakes and blades which possess full or no cortex . ..... . . ... ... 216 6. 7 : Means and standard deviations for flakes and blades with partial c ortex ... . . . .. .. 2 17 6. 8 : Percentage Comparisons of Blade Types by Count ......... ....... ........ ...... .. .. . . .. .. . 2 2 1 6. 9 : Bifacial Blank Percentages by Count ................................................... . ........... ..... 2 2 2 6. 10 : Correlations of flake scar density with dimension/size variables for all flake types . . ..................................... ........................................................... ....... .. ................. 2 25 6.1 1 : Reduction flake correlations of flake scar density with dimension / size variables . .. ................................................................ ........................ ..... . . ...................... .. ... 2 2 5

PAGE 13

xiii 6.1 2 : Shaping/finishing flake correlations of flake scar density with dim /size variables ... .. ....... ..... . ....................................................................... ........... ......................... 2 2 6

PAGE 14

xiv LIST OF FIGURES Figure 1.1 Map of West Mexico ............................................................................. .......... . .... .. . 2 1.2 Map of the Tequila Valley s showing core and semi periphery . ..................... .... ... .. . 3 1.3 The Teuchitlan Periphery (West Mexico) relative to the core and semi periphery . . .4 2 .1 A scaled representation of different forms of exchange ...... ...... ....... .... . .. . .............. 1 9 3.1 Circle 2 guachimont n from the Los Guachimontones site ................ .... ... .. ............. 55 3.2 Navajas Circle 5 guachimontn plan view ........................... ................. . ...... .... .... ... 5 6 3.3 Profile of a shaft tomb from the Atitlan Las Cuevas si te ... ... ............. .. ..... .. ........... . 61 3.4 Navajas central group with a reas of excavation circled ................ ...... . ...... ... ......... .. 81 3.5 Site layout of Llano Grande ..... .......................................................... .. ........... .. .... .. 8 8 3. 6 Llano Grande central group, with 5 meter increment c ontours ................ ... ........... . 90 3. 7 Plan of the Llano Grande guachimontn (with excavation units) ........ ....... . ............ 9 1 4.1 Profile of a rhyolite dome from a lava flow, showing the obsidian layer and spines ...................................................................................... ......... .................................. 100 4.2 Polyhedral blade production sequence ............................ .... ... ...... ....... .............. .. .... 103 4.3 Prismatic blade production sequence defined .................... ........ .... ......... .............. ... 104 4.4 Pike collection .................................................................. ............ ............................ 107 4. 5 Rough pike from Navajas Circle 5, lateral and cross sectional views .... ........ . . ... 109 4. 6 Refined pike fragment from Navajas Circle 5, lateral and cross sectional views . .10 9 4. 7 Anthropomorphic laminar jewelry from Navajas .............................................. ......1 1 2 4. 8 Major areas and sites studied within the Tequila Valleys ........................................1 40 4. 9 Tequila Valley obsidian source locations mentioned by Spence, et al ....... .............1 41 5.1 Llano Grande plan, with excavation units ................................................ ....... .... ..1 69

PAGE 15

xv 5.2 Navajas plan, with excavation units .................... ................... . .. ................... .......1 70 5.3 Redu ction flake ........................................................................ .. ............................ .1 75 5.4 Small shaping/finishing flake ................................................... ....... ........................1 75 5.5 Large shap ing/finishing flake ..................................................... .............................1 75 6.1 Llano Grande flake size distribution ................................................. .......... .. . ....... 211 6.2 Navajas flake size distribut ion ............................................................. ..... .. ... . ...... 211 6. 3 Flake and blade comparisons of c ortex c overage d istributions p er site .. ..... . .... .. 219 6.4 Core comparisons of Cortex Coverage Distributions at Llano Grande ........ .. ......... 2 19

PAGE 16

1 CHAPTER I INTRODUCTION The Tequila Valleys and the Teuchitln Culture The Tequila Valley s area of West Mexico is a portion of a chain of lake basins at the base of the Tequila Volcano, which form an open area within t he forested hills of the Sierra Madre Occidental mountain range that runs parallel to the Pacific Coast (Beekman 2010:43 44, 63; Fig. 1.1). During the Mesoamerican Late Formative and Early Classic periods ( ca. 3 00 B.C. 500 A.D), the Magdalena B asin on we st side of the valleys contained a large lake and the remaining basins were marshlands The areas immediately surrounding the lake were interspersed with steep hills and valleys of marshes and fertile alluvial soils, which formed a secluded setting for th e development of a complex culture known as Teuchitln. Aside from a general attraction to clay figurines and dioramas that have been quite popular with collectors since the turn of the 20th century (Townsend 1998a:17 22; Weigand 1975:186, 223 1985:4 7 54; cf B Lumholtz 190 3 ), the region ha s generally been marginalized by earlier archaeologists and explorers as an area populated only with nomadic groups and simple chiefdoms (Beekman 1996b:136; cf. Fernandez and Dera ga 1988; Schndube 1998). However, more recent investigations have better recognized the area's true cultural significance. A round 1970 investigators began to recognize a pattern of ceremonial architecture that points to

PAGE 17

2 a much larger and more complex soc iety than previously assumed ( Kelley 1974; Ohnersorgen and Varien 1996:103; Weigand and Beekman 1998). Today, the culture is Figure 1.1: Map of West Mexico. Numbers indicate lake basins. Of particular interest t o this thesis are the two main basins within the Tequila Valleys : Magdalena (2), and La Vega (3). Also of interest are more distantly associated sites in Sayula (7). (Beekman 2010:42). known to a small but growing number of investigators as one unique to Mesoamerica with a social structure that has been described as something akin to either a complex chiefdom or an early state ( Beekman 1996a,b ; Schndube 1998 ; Weigand and Beekman 1998) O ne area for potential research is site specialization. Although the majority of sites associated with th e Teuchitln culture are found in a core population zone south of the

PAGE 18

3 Tequila Volcano (figure 1.2) o thers are scattered throughout the rest of the Tequila Valleys. Six of t he sites outside the core zone, namely Estolanos Mesa, Mesa El Figure 1.2: Map of t he Tequila Valley s showing core and semi periphery. Circled sites are within the core, and all s ites outside the core population zone are considered semi peripheral Labeled sites are those under study. Navajas' location in a small exterior basin assoc iates it more closely with the core zone. Adapted from Beekman in press a :Fig. 3.2. Zacate, Cerro Tepopote, Peol Tepopote, Cerro Pipiole and Llano Grande, were built within or near passes in the surrounding hills T he exact purpose and nature of these surrounding sites has been a major focus of study ( Beekman 1996a,b ; Weigand and Beekman 1998 ). Within this paper, all sites within the Tequila Valleys but outside of the core zone are referred to as semi peripheral sites because of their location between t he

PAGE 19

4 densely populated core habitation zones and populations outside the Tequila Valleys. P eripher al sites are outside the Tequila Valley s yet within West Mexico (Figure 1. 3 ) Figure 1.3: The Teuchitln periphery (West Mexico) relative to the core a nd semi periphery. The line of questioning pursued for this thesis is a continuation and extension of these core periphery discussions, and utilizes evidence found in local obsidian production for additional interpretation of their economic dimension s. Substantial evidence from archaeological, ethnographic and ethno historic sources shows that elites of West Mexico relied heavily on agricultural production to finance their positions in the social structure (Beekman 2003a ; Butterwick 1998 ; Lopez and Ra mos 2006a ; Lumholtz 1903 ; Schndube 1998 ; among several others) Contrastingly, the position of Las Navajas, a Tequila Valleys (Teuchitln core and semi periphery) Teuchitln periphery Teuchitln periphery

PAGE 20

5 southeastern semi peripheral site on a high plateau with an expanse of fertile farmland, provided ample opportunity to continue the agricultural tradition of excess production needed for staple finance (Beekman 2007a ; Appendix A:Fig ure s 4 6). In contrast the micro environment of Llano Grande 's chosen location in a northwestern pass overlooking the Magdalena lake basin was not conducive to agric ultural production of any sort (Beekman 2001 ; Appendix A:Fig ure s 1 3) It did however, provide immediate access to an adjacent high quality obsidian source 1 and much greater potential exposure to external groups. Did Llano Grande operate under a different economic strategy relative to the core population zones of the Teuchitln culture? And, given the presence of obsidian did Llano Grande represent a relative shift within the Teuchitln culture from a surplus producing, internally competitive agrarian eco nomy to one based on trade? Studies and discussions of semi peripheral site functions in the Tequila Valleys have thus far concentrated primarily on architecture and area wide surface surveys with minimal excavation (cf. Beekman 1996 a,b; Weigand and Beekman 1998), and therefore could not yet address some of the economic dimensions of site function. However, more recently a large sample area has been excavated in the Llano Grande ceremonial space in the hills just outside the valleys to the northwest, as well as the entirety of one other example of public architecture at the site of Navajas to the s outheast. Ceramic and obsidian artifacts recovered from both sites have been the subjects of stud y over the last few years (Hoedl 2013 ; Johns 2014). F or t he current study, the above questions are addressed through a comparative analysis of the lithic assemblages recovered from the excavated ritual centers within 1 The Llano Grande and Navajas obsidian sources mentioned in this thesis refer to thos e near the sites within the Tequila Valley s by the same name, and should not be confused with the more northern Llano Grande and Cerro de las Navajas lithic sources in Durango (cf. Darling 1993).

PAGE 21

6 Llano Grande and Navajas Wealth and staple based economic patterns discussed by Timothy Ear le (1991) and related political strategies proposed by Richard Blanton and colleagues (1996) are used to frame the geographic and environmental implications of each site's economic potential in order to determine general economic expectations for the sites under test. P reviously pu blished lithic distribution analyses both within the valleys and in additional areas within Mesoamerica are used here to determine common lithic distribution patterns which relate to normative distribution strategies, and further refine specific expectations for each site's lithic industry in relation to Earle's economic categories Site contexts are discussed by local area, and include the primary core area near Los Guachimont o nes ( Esparza 2003 ; Soto 1982, 1990) as well as three semi peripheral and peripheral areas The latter areas include the La Venta Corridor (Beekman 1996a, 1996b), the Magdalena Basin (Spence et al. 2002) and the Sayula Basin (Reveles 2005) Because of the high volume and spatial breadth of comparative data, d istribution patterns of outside groups necessarily include a limited number of studies that have the best potential to identify common patterns which may have influenced semi peripheral groups External groups discussed include Teotihuacan (Spence 1981, 19 87 ), the Maya (focusing primarily on Nohmul and Colha as discussed by Johnson [1 996 ] ) and earlier groups in the Valley of Oaxaca ( De Len et al. 2009 ). The specific research questions addressed in this thesis are as follows: 1 D o the lithic assem blages at Llano Grande show a greater emphasis on production over use contexts than the assemblages at Navajas and other sites more closely affiliated with the core population zone ?

PAGE 22

7 2. Does Llano Grande show evidence for a greater emphasis on the product ion of potential trade than Navajas, includ ing commonly exported forms known elite items and/or items which require specialized production techniques ? A positive answer to the first question supports the notion of a specialized area of production t hat may be defined as a workshop (Clark 1981 ; Spence 1981 ) at Llano Grande either w ithin or nearby the public architecture, which also suggests a possibility of systematic overproduction for some form of exchange H owever, production could still be used f or the general trade of common goods or of minimally prepared raw materials. The second question deals strictly with the forms of items produced, and a positive answer to the question may show increased production of potential ritual or trade items a t Llano Grande However, based on the second question alone, the use of these items is ambiguous in that they may be intended for elite trade, or strictly for internal use. Positive answers to both questions would support an increased emphasis on both over production for trade and the development of wealth items in the same context, and therefore also an i ncrease in the likelihood of a greater emphasis on the wealth trade economy. The null hypothesis for this study is a lack of support for an increase i n wealth trade at Llano Grande over Navajas T herefore both questions must be positively answered for the null hypothesis to be rejected.

PAGE 23

8 Lithicist George O'Dell's lament about his own specialty eighteen years ago is still essentially true today: "Alt hough advances in lithic studies have been made in both the theoretical and methodological realms, yawning gaps exist in their articulation" (1996:4). This thesis is theory driven by design, and will serve as an attempt to bridge the "yawning gap" between the theoretical approach and methods through an explicit description of the relevant theory, the middle range interpretive rules and assumptions, and the implications of the results on the theoretical foundation. Overview of Chapters Chapter II is d evoted to describing the social theory which addresses and guides the questions under study T he chapter first covers theoretical discussions on the nature of core periphery relations, and then turns to economic dichotomies that may explain how core areas may differ from peripheral or isolated semi peripheral areas in economic strategies: Earle's wealth vs. staple elite financing (1991) and related n etworking/exclusion vs. corporate/inclusion power strategies (Blanton et al. 1996 ) The internal dynamics of these strategies as they are pursued in the same cultural groups and engage each other within designated community spaces, or arenas (Turner 1974) is also discussed T he chapter then discusses the reciprocal relationship between culture and environment, a nd how environmental factors may play a major role in elite decisions to emphasize certain economic strategies over others. Chapter III covers Tequila Valley s background information. The chapter begins with the history of archaeology in the area and t hen turns to details of the area 's prehistory,

PAGE 24

9 which begins with past and current understanding of the timeline focusing on the transition from the Late Formative to the Early Classic perio ds. The prehistory section first briefly describes the distinctive elements of the culture ( guachimontones shaft tombs, ceramic dioramas/figurines, and unique forms of obsidian), then describes the demographics and other physical properties of core, semi peripheral and peripheral sites. General descriptions and interpret ations of apparent cultural uses of the shaft tombs and the guachimontones are then described in separate sub sections. The chapter then discusses the current evidence for staple and wealth finance strategies. Feature details of the primary sites under stu dy, Llano Grande and Navajas, are then discussed and compared to the primary core site, Los Guachimontones The chapter then turns to environmental data, including resource and climate details. General area wide details are presented before concentrating s pecifically on geography and climate differences between Llano Grande, Navajas and Los Guachimontones. Chapter IV discusses the area lithic data, starting with an overview of Mesoamerican lithic forms, including details of West Mexic an forms and disc ussions of proposed production methods which may offer ways to identify the manufacture of items which have been exported off site from the remaining debitage. Since semi peripheral area strategies may be as much or more influenced by external groups as th ey are by the core lithic distribution strategies including both products and debitage, three other Mesoamerican groups (Teotihuacn, Oaxaca and the Maya) are addressed before discussing the literature on lithic distribution patterns within the Tequila V alleys Other r elevant lithic a ttributes at the site and area level are then discussed and compared between Los Guachimontones, the La Vent a Corridor and unaffiliated sites within the

PAGE 25

10 neighboring Sayula Basin to find any detectible patterns of similarities and differences between core, semi peripheral and peripheral sites Chapter V discusses the methods used to test the hypotheses outlined in th is chapter. The chapter begins with a description of the datasets at both Navajas and Llano Grande, includi ng the prior excavation process and associated issues encountered during excavation and artifact retrieval the method of storage, and later retrieval and recording of the data. Expectations from the data for a wealth trade or an internal use (staple) stra tegy are then listed and described with the implications of all potential outcomes, and a series of tests used to determine whether these expectations are met are described. Chapter VI reports on the outcomes of the tests outlined in Chapter V T he im plications of each test's outcome on the research questions are also discussed in relation to the site expectations. Chapter VII concludes the thesis with a syn thesis and summary of all outcomes and provides an overall interpretation of the sites und er test in relation to the original questions and the implications that the results have to the theoretical foundations of the tests. A section on future work in pursuit of the same or related questions is also provided. T his project is not designed o r intended to completely resolve the economic nature of all semi peripheral sites associated with the Teuchitln culture or even completely resolve the issue specifically for Llano Grande and Navajas, as it is more of a theoretically guided probing study involving only two sites in detail, and general information from a few others. I t is anticipated that, when employed on a much larger

PAGE 26

11 scale, this form of theoretically informed technical analysis will bring West Mexico lithic research much closer to resolv ing these kinds of questions.

PAGE 27

12 CHAPTER II THEORETICAL DIRECTION At the most fundamental level, this thesis seeks to determine whether economic differences can be seen in the material culture between sites associated with the same polity, but at dif ferent locations and positions relative to the core area, particularly when the chosen locations are within very different environments; and whether these economic differences represent differing emphases on wealth and staple economies (Earle 1991) F ollow ing Hirth (1996), economic differences between sites are seen here as relative differences in emphasis on multiple specific economies that may cross cut staple and wealth categories, rather than a predictable dominance of one economic category over the oth er at the site level as originally proposed by Earle, and also Blanton et al. (1996). This study further tests whether observable elite choices in emphasis on different economies are influenced by the immediate area's economic resource options, as well as the distance and access to other internal and external groups Core Periphery Relations D ifferences in resource availability within different areas relate to patterns of settlement and social relationships originally described by Wallerstein's (197 4) world systems theory. Wallerstein describes a core group which is comprised of concentrated area s of consumption and a peripher y which contain s the resources necessary for maintenance of the core's economy I n between the core and periphery are semi

PAGE 28

13 pe riphery groups, which interact with both the core and periphery and create relations of exchange between the two areas (Wallerstein 1974:401 405) W allerstein applied his theory exclusively to industrial capitalist societies with the technological capabili ty to extend their reach globally and create dependent and unequal exchange relationships with less powerful groups, for the acquisition of agricultural and industrial bulk goods (1974:398 399) T hese relationships create a single system of labor division which cuts across cultural groups and polities (1974:390). The primary feature of Wallerstein's theory is the formation of a dependent and dominated periphery in an asymmetrical exchange relationship. Schneider has suggested that Wallerstein's world s ystem "suffers from too narrow an application of its own theory" in its strict application to capitalist systems (1977:47), and relies too heavily on bulk goods while dismissing luxury items as non systematic trade items that had little effect on the overa ll economy (1977:52) S chneider (1977:52 53) contends that several luxuries, such as sugar, wine and precious metals, were also used quite effectively to manipulate peripheral groups before the world system began to form in the mid seventeenth century. Sch neider also referenced archaeologist Robert Adams' discussion of inter polity social influence of long distance luxury trade as far back as the earliest complex polities within Mesopotamia. Adams suggested that luxury trade was a "formidable socioeconomic force" (1974:247) which was often tumultuous, and used by the more powerful polities to coerce and dominate less powerful groups within Mesopotamia, given the presence of slave labor and very dynamic changes in exchange relationships revealed by the cuneif orm records for luxury item trade (Adams 1974:247 249)

PAGE 29

14 Blanton and Feinman (1984:674) suggested that the world system theory may be a more productive framework for the study of polity interaction within Mesoamerica, rather than the common approach t o the region at the time, which treats the entire subcontinent as one massive culture area, which is yet multicultural, but held together by common ideologies and technologies that spread through various interpolity relations. These relations were identifi ed through inter polity commonality of some aspects of material culture, but essentially left unstudied regarding the nature of how and why they form and change over time. Blanton and Feinman do not advocate the application of all details of the theory to prehistoric Mesoamerica, and point out that no polity within Mesoamerican prehistory has ever extended its reach throughout the subcontinent, and some institutions associated with the capitalist system do not appear in prehistoric context. Blanton and Fein man's idea of a Mesoamerican "world system" utiliz es Wallerstein's basic framework of cross cultural labor division in a single multi polity, cross cultural system within the Mesoamerican context (Blanton and Feinman 1984:674; Hirth 1996:19) However Blan ton and Feinman assign a role for luxury goods which is as important as the role Wal lerstein assigns to bulk goods ( 1984:675 676) "through the calculated distribution of symbols of status that the elite controlled" (Blanton and Feinman 1984:676) T hey al so suggested that multiple cores likely existed beyond their example of the Postclassic Central Highlands (areas exploited by the Aztecs), including Late Formative and Classic Period Teotihuacn, but "at present, we know less about them" (1984:679) The principles of world systems theory have since been utilized and modified by archaeologists to describe the economics of early prehistoric states, and theorists developed varying adaptations which included different dynamics between the core and

PAGE 30

15 periphe ry. Algaze (1993) closely follows Wallerstein's definition for modern capitalist societies A lgaze interprets peripheral sites as "outposts" intentionally established within areas controlled by external groups with a very singular, intentional purpose of b ringing foreign resources into an expanding core area, as the core's resource requirements i ncrease due to its own growth. He further states that "by definition, marked developmental asymmetries always exist between pristine states and communities in the p eriphery." Informal trade relationships would then be established with a high probability of success, since peripheral groups would, at least initially, perceive some benefit from political relationship with the a more powerful core However, Algaze descri bes outposts as "dendritic" entities entirely controlled by delegates of the core, therefore the trade off to local elites is the cost of any real political control over the relationship. These outposts would then serve as remote collection points of goods from their respective locales (Algaze 1993:304). Algaze further suggests that these outposts are, by nature, exploitive as he states that "early outposts reflect a system of economic hegemony whereby early emergent states attempted to exploit less complex polities located well beyond the boundaries of their direct political control and this system may be construed as imperialistic in both its extent and nature (1993:305) Gil Stein (1999:36 ; 2014:55 56, among others ) has criticized the view of a strong, dominant core exploiting a week, subaltern periphery in world system theory as over simplified and monolithic. Stein views the nature of the relationships between groups as quite variable according to individual political, economic ideologic al and logistical factors (such as distance and transportation economics ) within each group I n many

PAGE 31

16 cases, according to Stein, peripheral groups are active agents in the direction of trade relationships, and in some cases managed fully interdependent re lationships rather than asymmetrical ones Chase Dunn and Hall (1991:19) also see a wider range of relationships involved in core and periphery relations, and offer a more inclusive definition in their model which distinguishes between core peripher y differentiation, or groups of different sizes and social structures interacting within the same system, and core periphery hierarchy the domination of one society over one or more others by means of political, economic and/or ideological control; differ entiation does not assume dominance, and can involve favorable exchange for both sides E quality" of exchange is also difficult to measure, largely because different groups value the same objects of exchange differently within their own cultural contexts, especially if they utilize exchange items for different purposes ( Chase Dunn and Hall 1991:31). For the purpose of the current study, no initial assumption is made about the specific nature of the relationships between core and peripheral groups, and ther efore Chase Dunn and Hall's more flexible definition of core and peripheral sites are used. Outposts and R emote G roups Outpost settlements often occur at or beyond the farthest reaches of direct control, and are often strategically placed on natura l transportation routes between the polity and external groups to facilitate both control and efficient transportation, collection and redistribution of traded goods, especially where transport is particularly difficult (Algaze

PAGE 32

17 1993:321) P rehistoric Mesoa merica provides an excellent example of a region where transportation would have been especially problematic in many areas, due to a lack of pack animals and very rugged, hilly terrain in many places, especially along the Sierra Madre Occidental and Trans Continental mountain ranges (Schndube 1998:206). According to Algaze, collection and redistribution operate most efficiently when handled in outposts somewhere near the outer edge of a group's area of political control, in corridors which provide li mited access O utposts are also frequently placed very close to a desired resource in a remote area O utposts vary greatly in size from small apparent way stations positioned along long distance trade routes, to large remote city settlements set up on the way to other major polities that provide high volume trade (1993:310 311) The establishment of peripheral outpost trade is not generally considered an end point to core expansion. European colonial trade enclaves could be initially benign and even w elcome by peripheral groups, but they often became entry points for later expansion of core powers throughout the are a. Expanding trade eventually radically transformed cultural landscape s around the world, including the America n continent (Curtin 2000:4 5 ). Although the arrival of the Spanish empire was a chief concern for the natives i n Central Mexico, the Spanish trade economy was likely not initially considered a threat compared to other colonial ventures in agriculture, ranching and the mining of gold and copper H owever, eventually "trade empires" became the single most important factor driving European expansion as they s uccessfully out competed all ventures Enclaves i n the centers of the native populations ultimately became the Spanish political cen ters for the area (Curtin 2000:79).

PAGE 33

18 Alternative models of economic redistribution have been proposed which do not emphasize Wallerstein's principle of core dominance Stein 's (2014:59) concept of a trade diaspora is quite similar to descriptions of ou tpost s and enclave s, except that they may experience a wide range of power relationships with local polities D iasporas can be very dominant as in World Systems models, but may also be marginalized by remote groups or experience neutral positions of prote cted autonomy among their hosts if the local elites find them useful A lso, the distribution of goods may be relatively uncontrolled, or only distantly managed within open market systems K enneth Hirth's (1998:453) proposed approach to identifying centrali zed market activity notes several consistent spatial patterns used by some archaeologists as clues for identification, derived from Spanish contact period observations. Large markets were often set up in Mesoamerica along major trade routes and either near or adjacent to "administrative centers," and served as points of collection and redistribution of goods. These observations closely match many aspects of Algaze's definition of large trade route outposts, and some areas interpreted as "outposts" may have actually operated as centralized marketplaces Hirth describes a centralized marketplace a s one strategically placed for access by multiple groups, which features un restricted exchange (although likely sponsored and managed by elites), with equal access of all goods to all consumers regardless of social status Centralized markets tend to be quite large and typically operate on a regional scale (1998:454 455). In contrast, Algaze's interpretation of peripheral exchange is a hegemonic one highly favoring the core polity T he observed ethnohistoric markets were typically regional exchanges set up in distinctive forms of permanent architecture surrounded by walls which also surround buildings directly associated with market administration, but as

PAGE 34

19 Hirth ( 1998:45 5) cautioned, prehistoric exchange s may have ranged widely in scale from central exchanges much like those in historic accounts, to locations designated for occasional isolated partner trades which do not involve direct architectural evidence of the point of exchange. A few forms of trade also occur in venues in between these two extremes in scale, including trade activity conducted directly at workshops locations ( Figure 2.1). Smaller scaled exchanges tend to be less centralized, and unequally accessed by certain segments of the population with access to trade networks of specific groups and an unequally high or even exclusive distribution of specific products and resources among network affiliates (Hirth 1998:455) One likely example of a large outp ost marketplace exchange area may be within the 6 km 2 core city area of Matacapan, perhaps at or near the central plaza. Matacapan is situated about half way between Teotihuacan and Mayan territory, yet shows very strong material cultural continuity with T eotihuacn, including Teotihuacn's trademark apartment building like residential structures (Algaze 1993:304, 310 312) Figure 2.1: A scaled representation of different forms of exchange. ( Hirth 1998:455 )

PAGE 35

20 Although Alg aze asserted that especially remote and/or isolated outposts must maintain a fragile connection to the core due to transportation and communication limitations (1993:325), Phil Weigand has argued that direct control of outposts is n o t necessary, since thei r function and reason for existence is defined by economics and dedication to external trade and interaction with outside economies, rather than political relations ( Weigand 2000:52; Weigand and Beekman 1998:44,45). However, Algaze refers to more distant outposts than Weigand, and the form of core control to which he refers is not political, but rather hegemonic and economic Elite groups maintain their dominance through economic influence rather than institutionalized positions of power and authority. Chase Dunn and Hall have also argued for the use of world system theory on smaller scales in prehistoric contexts, on the basis that the range of interaction is actually part of the whole world of influence known to the system's participants, and is larg er than any individual polity within the system. In prehistoric contexts, the authors also hold, along with Schneider (1977) and Blanton and Feinman (1984), that the prehistoric core/periphery connections tend to emphasize prestige goods; bulk exchanges in staple goods generally range over much smaller areas than exchanges in prestige items (Chase Dunn and Hall 1991:12; Blanton and Feinman 1984:677) Chase Dunn and Hall also make use of Wallerstein's semiperiphery category of economic divisions in bet ween core and periphery sites T he semiperiphery is an economic and cultural intermediate point between the core and periphery, and is considered "both exploited and exploiter" in the economic process T he semi periphery thus acts as a political buffer zon e (and in this case, according to Wallerstein, the function

PAGE 36

21 is indeed mainly political rather than economic) between core and peripheral areas, to avoid direct confrontation between starkly polarized regions, and thus promotes system stability (Wallerstein 1974:405). Semiperiphery sites may have hybridized core and peripheral organization strategies, and institutional features may show intermediate forms between those in the core and peripheral locations (Chase Dunn and Hall 1991:21) S ites in these areas o f transition are especially suited to social innovations since they can combine core and peripheral elements to create new cultural elements, and are less bound to the core due to their physical distance (Chase Dunn and Hall:1991:31). Blanton, et al. (1996 :7) also note that such areas are among the most likely areas to establish trade networks, and also often occur in areas of low agricultural potential. The Aztecs were well aware of the advantages of semi peripheries, and chose to transform the Postc lassic Ce ntral Highlands city of Tepeaca into a properly functioning semi peripheral group. Tepeaca was a city on an Aztec trade route which at one time discouraged trade by killing and robbing merchants passing through the area. Upon assimilating the city the Aztecs insisted that Tepeacans cease their usual practice with outsiders and welcome merchants from all areas. They were also instructed to increase their number of tlamemes (burden carriers), and the Aztecs established a marketplace for luxury goods within the city (Blanton and Feinman 1984:678). Boundaries Semi peripheral sites can be difficult to identify in the archaeological record, and we must rely on multiple lines of evidence beyond spatial organization to do so C ore,

PAGE 37

22 peripheral and s emi peripheral zones are defined by the division of labor across polities, which are independent of political territories H owever, elites utilize inter polity trade flows for their own political gain through the accumulation of goods for the building of s ocio political hierarchy, and through the economic manipulation of groups beyond their direct political control (Hirth 1996:219) T he functions of physical boundaries may therefore include attempts to control economic activity in semi peripheral areas. Of course, boundaries may also represent a number of other functions, including physical defense, service as an observation post, message relay point, symbolic representation or embodiment of the state, administrative center, strategic activity center, and di plomatic center (Southall 1988:56) A lso, boundaries are often invisible in archaeological data, although heavily controlled sections of a boundary are sometimes revealed by walls (Chase Dunn and Hall 1991:18) Cultural borders are also often ill def ined, and even when they are made explicit, lines drawn on a map in ink as "borders" as though indelible are actually quite transient, elusive entities which may move back and forth frequently with contention over territory T he existence of boundaries als o belies the great deal of interaction that tends to occur across them; no society can be assumed to be completely isolated, regardless of the level of complexity (Schortman and Urban 1987:81 ; Wolf 1982). A boundary's functions can also vary both spatially and over time (Wolf 1982). A boundary may be very porous to economic goods, while the movement of people is closely monitored and heavily guarded (cf. Beekman 1996a:742). Yet at another time, or even concurrently at a different location along the boundary it may be completely open and any physical manifestation may serve only as a historic symbol of the state. For these reasons, other lines of evidence should be

PAGE 38

23 used to confirm the existence of a boundary. Ultimately, although boundaries can be complex h eterogeneous and therefore challenging to interpret understanding the nature of a boundary means understanding something of the internal structure and strategy of the polity which created it (Southall 1988:56). Unitary S tates and S egmentary S tates Some early states may not even conceptualize a physical boundary as part of their self identity. Southall has noted an organizational difference based on the degree of intentional definition of a polity's boundaries, between what he calls unitary and segm entary states. Unitary states are much more formally defined, centralized and territorial by nature, and therefore strive towards "stable and controlled boundaries" as a management tool S egmentary states maintain their power structures through hegemonic c ontrol based on economic power and prestige, rather than formalized institutions of authority (1988:55). Segmentary states may be identified in material culture through secondary site emulation of primary site architecture and symbolism, as an indicator of hegemony. The developmental trajectory of a segmentary state may eventually transform it to a unitary one as control of more remote locations is consolidated under a centralized government, where previously only a hegemonic form of control had existed. Su ch a transition may not often be easily accomplished, since direct administration of remote sites in territorial polities is much more costly, and has only been completely demonstrated by very powerful, well centralized polities such as the Roman Empire (B eekman 1996b:136). The limitations of transportation only by foot, hilly terrain, and a

PAGE 39

24 very localized economy may make unitary state formation practically impossible in some areas, and control strategies may also vary within the same polity (Southall 1988 :63,72) Mesoamerican polities include a range of forms between unitary and segmentary states. A few of the most notable examples include the Aztecs along the Aztec Tarascan border, the remainder of the Aztec empire, and the Classic Maya. The Aztecs near the border with their most powerful adversary were highly territorial and centrally controlled, with several fortresses along the border ( Beekman 1996b:136 ; Beekman and Houston 1993: 3). Yet, throughout the rest of their area of control, the Aztecs pa id no attention to territories and only imposed hegemonic control with only a return of tribute from subordinate groups (Smith 1997:76) T he Aztecs at the Tarascan border may have been heavily influenced by the Tarascans themselves, since the Tarascan empi re appears to have been very highly centralized and organized as territories managed by a hierarchical administration (Pollard 2003:80 82) T he Maya gained power through ceremonial and symbolic hegemony to form various city states spread out through a larg e portion of the Yucatan Peninsula. Several especially powerful cities such as Tikal did reach some level of centralization as they subordinated several smaller groups under the rulership of "overkings", but there was no overall centralization, spatial org anization, or territorial boundary that would ultimately unify the Maya, and although some walls do appear, territories still do not appear to be a defining element of affiliation with the Maya as a whole (Martin and Grube 2000:18 21).

PAGE 40

25 Political Strate gies Wealth vs. Staple Economies The same division between luxury and bulk goods discussed above for the economic bases of world systems also has been described by Timothy Earle as a fundamental separation of economies utilized by elites in differen t groups to finance the maintenance of their social positions (1991) Earle uses this division to characterize economies of entire polities although he does acknowledge that economies may occur "as admixtures in individual cases" (Earle 1991:3) which sug gests that his economic division may occur at some component level within at least some polities Although more complex views of how different economic bases and their related strategies are utilized have been adopted more recently (e.g. Blanton, et al. 19 96 ; Hirth 1996), the basic definitions of wealth and staple economies remain important elements of more recent views. Wealth economies involve long distance exchange and/or specialized craft production for symbols of status which elites use to legitim ize their social positions, whereas staple economies include mainly food related resources utilized by elites as leverage for power S taple economies include the hosting of feasts to attract followers and create a level of obligation to the hosts, and poss ibly other forms of food related payment (Earle 1991:3) In wealth economies, elites incorporate exotic items from remote areas into group ideology as a means of creating exclusive control over objects that symbolically convey legitimacy of authority T hese prestige items are either made from locally rare or unavailable materials, or they are difficult or expensive to produce, which makes them

PAGE 41

26 more difficult to access and therefore easier for elites to control and monopolize their distribution (Beekma n 2000:386, Earle 1991:7). Such objects are often utilized in ritual contexts to reinforce the elite ideology and validate elite authority (Reilly 1996:30) A common example of a wealth economy is the Classic Maya, who extensively traded and utilized eccen trics, ritual items and rare, distantly acquired commodities (e.g. Martin and Grube 2000:16; Johnson 1996). Mayan rulers legitimize their positions by "impersonating" (or perhaps, transform into, or join with, in their interpretation) the gods they represe nt using masks, ornaments and jewelry (Houston and Stuart 1996:291), and used very fine, specialized bloodletters in other ritual contexts (e.g. Carballo 2009:494 ; Johnson 1996:172 ; Joyce 2004:195) In staple economies, feasting activity is used as a competitive display of power through abundance in order to gain political standing and an upper hand in economic and political negotiation I t is also very expensive as expectations of diverse and plentiful provision run high. Feasting is therefore also an indication of political competition among groups within a cultural tradition (Butterwick 1998:89 90,104 105). The extent of the sudden drain on staple resources required for feasting activity requires staple economies to rely heavily on a land and climate capable of consistent agricultural surplus (Hayden 1995: 62,63). Numerous ethnographic and archaeological accounts of feasting cultures have been identified through evidence of regular feasting behavior Many favor livestock or wild mammals as the more pr ized fare over agricultural products, although a wide array of agricultural products are still included, and some extensive land resources and human labor are still ne cessary to support horticulture for highly populated areas Some of the most common and e laborate feasts occur at funerals, which provide opportunities

PAGE 42

27 for interaction when a void has been created in a social position, possibly requiring a shift in the social structure (Hayden 2009). Among the best documented staple groups are those within the Tana Toraja regency of Sulawesi, Indonesia, who are very highly stratified and hold six levels of funeral feasts, each with socially understood degrees of elaboration, based on the social status of the deceased E xpected expenditures (which hosts will go beyond if possible) range from 1 2 nights with a single pig and possibly a water buffalo for poor people, slaves and infants; to at least 27 nights, over 36 pigs and 16 water buffalos for only the very wealthiest individuals. Although the latter case may o nly occur once or twice in a lifetime, the former tends to occur approximately 20 times per year, leading to a consistently high expense in livestock (Adams 2004:64 65). According to Kristiansen ( 1991:22 ) wealth economies tend to be individual orient ed and horizontally structured since they involve individual peer level contacts and networks for exchange, whereas staple economies are more collective and vertically structured (such as the above example of Tana Toraja), as agricultural production involv es the coordination of large numbers of laborers. Kristiansen and Earle both propose that although the two strategies differ by nature and thrive within different social structures, they can exist within the same polity. O ne economy will become dominant, a nd the "alternate" economy will become a dependent variable subordinate to the other (Kristiansen 1991:22; Earle 1991:8) I n Kristiansen's view, overall trends in culture trajectories follow a usual progression towards increased complexity from wealth fina nce towards staple finance, but each of the two economies can follow its own trajectory into different patterns of increased complexity (Kristiansen 1991: fig. 2.2). However, individual trajectories have proven quite difficult to predict with any certainty Past

PAGE 43

28 attempts at creating any universal rules applicable to cultural trajectories have been thwarted by differences in environments, related technological developments and individual culture histories F or example, some groups considered to be members of larger state like organizations may exhibit greater autonomy because they are cut off in some way from their parent culture E arle treats these situations as exceptional cases, and termed such groups devolved societies (Earle 1991:14 15,25). A trajectory may also be altered by political reorganization, which may increase the demand for rare and exotic items and/or create a period of resource intensification as factions vie for control in a power vacuum (Beekman 2010:71, c.f. Beekman and Christensen 2003:14 5 149, Pollard and Cahue 1999). Therefore, the particular history and environment of a group must be considered before attempting to interpret their economic base Despite Earle's and Kristiansen's separation of the two economies, actual examples con sistently show a trend towards a strong presence of both wealth and staple economies within the same social contexts T he Mayan elites are well known for their use and trade of a broad array of eccentrics, ritual items and luxury goods, as shown in grave g oods public ceremonial spaces and other ritual and elite contexts ( Johnson 1996; Martin and Grube 2000:16; Moholy Nagy 1999:310 ; Spence 1996 ; and several others) K nown wealth items for the Maya include jade jewelry, elaborately decorated ceramics jaguar p elts, cacao, shells and other marine items from the coasts, exotic feathers from other regions, clay and carved wood figures of gods ( Barrett 2004:60; Martin and Grube 2000:16) as well as chipped stone eccentrics and ritual items Elite Mayan stone items a re typically made from both local chert and rare obsidian from the Guatemalan Highlands and areas as far as Teotihuacan (1200 km) which controlled the nearby Pachuca source

PAGE 44

29 of an unusually pure green obsidian ( Johnson 1996 ; Moholy Nagy 1999:300 303 ; Spe nce 1981, 1996 ). But the Classic Maya also heavily emphasized maize and fertility, and elite connections to elements related to the staple yield, including a maize god which actually represented all life cycles, including those of people To the Clas sic Maya, maize was a central component of life and therefore all life cycles are related to the pattern represented in the life, death and regeneration of maize (Martin and Grube 2000:16). The Maya also emphasized regularly scheduled feasts. T he Mesoamer ican calendar's 18 month cycle recognizes monthly feasts (Townsend 1992:212 215) A lso, in the ancient Mayan language, the phrase for an esteemed Across polities, it do es not appear that an increase in staple reliance necessarily relates to a decrease in emphasis on wealth goods, or vice versa. In fact, quite the opposite appears to be the case for Blue Creek in the Early Classic to Middle Classic periods, which exhibits an unusually high quantity of various luxury goods (Guderjan 2007:91), and especially of jade (Barrett 2004:109,155), considered one of the most valued luxury items based on its rarity and consistency of association with elite contexts (Barrett 2004:23 24 ). Yet Blue Creek's main economy appears to be based on the overproduction and trade of various agricultural items, including staples such as corn, beans, squash and other foods evident in phytolith evidence examined from the very fertile bajos and highlan d fields in the area; and an elaborate system of canals and ditches which would have served to regulate the flow of water between the rainy and dry seasons (Guderjan 2007:91 101). This dual economy coincides with Blue Creek's fortunate combination of a

PAGE 45

30 str ategic position at the end of a long trade route to encourage its wealth economy, as well as a location among pockets of very fertile land to feed its staple production (Barrett 2004:155) A group's location along the cultural and physical landscapes, as w ell as the natural resources which surround it, may therefore have more to do with a group's chosen economic emphasis than its current organizational state within a structural trajectory. Wealth Relations and Boundary Permeability Kowalewski et al. (1983:39) have proposed that decreased centralization of a social structure and increasing group population size are both directly proportional to a boundary's permeability where permeability is the relative amount of energy flow across the boundary I n m ore concrete terms, permeability can be viewed as the amount of cross boundary interaction, which is most visible archaeologically as economic flow of goods evident in the quantity of similar materials and artifacts deposited on either side of the boundary Centralization is defined as the amount of interaction between the most active node or settlement group and other internal social groups, relative to the overall amount of interaction within the internal social system; and size is defined as the number o f interacting social groups ( Kowalewski, et al. 1983:35) Since Kowalewski et al. 's measure of size does not include group populations or areas, but only a group count, it may be best viewed as more of a measure of population dispersion, rather than overa ll population count or density G roup identity, according to Kowalewski and colleagues, is based on the degree of integration between component groups, where their most

PAGE 46

31 actionable definition of integration is "the total amount of flow of matter, energy, or information between system components" (Kowalewski, et al. 1983:35). As noted by Beekman (1996a:8), Kowalewski et al.'s theory is a proposed abstract relationship between social parameters and is not tied to any particular social model However, th e connection to semi peripheral area economics, especially where a clear boundary is present, does find some relevance here because it points to factors that may influence the degree to which economics of semi peripheral sites may be directed to external g roups; greater boundary permeability means that semi peripheral sites are more heavily engaged in externally directed interaction Increased interaction also increases the potential for trade relationships which would most likely include remote items invol ved in wealth trade. Elites may attempt to increase their power by artificially increasing the size of the system (i.e., adding to the number of social groups), and thus the required level of integration between groups. This strategy, however, may not succeed every time simply because the available resources are n o t enough to increase the system size and maintain an adequate level of interaction to avoid fragmentation (Kowalewski, et al. 1983:37) S ince elite sponsored attempts at expansion often take the form of outposts and semi peripheral area sites (Algaze 1993, Chase Dunn and Hall 1991), expansion also may maximize exposure to outside groups, which in turn increases elite opportunity for acquiring prestige goods. Kowalewski, et al. tested thei r hypothesis on count and location data for both sites and artifacts from the valley of Oaxaca, from the Middle Formative Period to the Contact Period, which include the temporally sequential groups of San Jose Mogote and Monte

PAGE 47

32 Alban. The study confirmed t hat although the trajectory of the three variables within the region is inconsistent in the short term, the long term trend is towards decentralization of the polity T he correspondence between size, centralization and permeability only aligns with the Kow alewski hypothesis for 5 of 11 organizational phases. A positive correspondence between size and permeability, however, is indicated for 9 phases T herefore, the level of centralization does not appear to be a strong indicator of boundary permeability at O axaca, but the relation between permeability and the number of internal groups is consistent for over 80 percent of Kowalski et al 's cases. Exceptional cases noted by Kowalewski occur at two extremes, which reveal patterns as notable as the overall r esults In one case, the system size shrinks to a point where the distance between groups is too great and the amount of inter group interaction drop s too low due to declining populations As a result so the price of integration can no longer be met. The polity then becomes fragmented and really no longer exists as a unified system. The remaining, newly independent groups seek additional resources across what has essentially become a defunct boundary (Kowalewski et al. 1983:49) The situation also allows f or the possibility of a redefinition of boundaries a round multiple smaller polities, which may then also begin to interact. The second situation occurs when the increase in size is so great that it requires a complete reorganization, and elites once again turn inward to re establish groups via staple resources until a new system is well established (Kowalewski et al. 1983:49). Although the Valley of Oaxaca case reveals local trends that may be important to future studies, the authors stress that the p atterns found in Oaxaca have not been demonstrated elsewhere They suggest expanding investigation along the same lines to

PAGE 48

33 other areas to determine how far beyond Oaxaca similar results might continue to hold (Kowalewski et al. 1983:49) Network and Corp orate Strategie s Differences in economic bases between different groups within a culture may lead to differences in the underlying power strategies employed by elites to build and maintain power R ichard Blanton and colleagues ( 1996) have recognized a distinction between basic strategic directions which he terms networking/exclusionary and corporate strategies. Elites engaged in a networking/exclusionary strategy establish relationships with specific contacts within a network of other elites, with who m they share and exchange resources which differentiate them from non elite groups. Network affiliates may be formed along pre determined parameters, such as kin or ethnic group, or purposely selected based on some criteria, such as access to specific reso urces T he network strategy involves a large degree of control of access to specific prestige goods requiring skill specialization, especially in peripheral areas, which may see a large jump in specialized production Network strategies also tend to foster innovation of product types, al though an element of secrecy may slow the spread of some innovations. P roduction techniques which allow for less specialized skillsets to produce prestige items are not common ly spread as the effect would be to "banal ize" the product by making it common and universally available, and therefore lose all value as a prestige item Given enough time, however, production innovations, "secrets" and/or source connections may slip to the

PAGE 49

34 common public This occurred with the E arly Postclassic spread of prismatic blade core platform grinding, which made the production of prismatic blades easy enough for non specialists to regularly engage in the practice for their own uses (Healan 2009) I n such situations, elites may need to de fine new forms of prestige items to symbolize their status. The strategy is also labeled "exclusionary" because the act of selecting specific contacts for the exchange of goods also channels wealth and power away from other groups by taking advantage of resource monopolies A s a result, the practice limit s the number of people which have access to certain products (Blanton, et al. 1996:5). Groups engaged in a corporate strategy form a group identity through multiple peer institutions The corpor ate group is unified through an ideology that is communicated and reinforced by ceremonial and ritual activity. Corporate power strategies limit or prevent domination of one group through a corporate code of behavior. Teotihuacn may be the largest and bes t known Mesoamerican exa mple of a corporate strategy. Blanton et al. (1996) suggest that the material culture close to the city center reflects a pattern of collective identity and action rather than sovereign rulership of a particular lineage (see also D eLucia 2008, DeMarrais et al. 1996). This is not to say that hierarchy is replaced by egalitarianism within each component institution, but rather that elite power images are suppressed in corporate level contexts such as ceremonies, collaborations on labo r projects, and defense, in favor of symbols and images which support corporate group identity ( Beekman 2008:414 415). Since corporate strategies are inclusive, they require large public spaces, whereas space requirements for networking strategies are very small and private due to the need for exclusion ( Beekman in press a:57 60).

PAGE 50

35 Connecting Economy and Strategy Blanton, et al (1996, Table 3) specifically associate their networking/exclusionary e their corporate/inclusion strategy with Earle's staple finance A ccording to Blanton, et al., wealth economies are controlled via trade relationships which form networks to obtain, transport and exchange prestige goods so that their distribution can be t ightly controlled. Prestige goods can then be used to legitimize the networked group's position of authority. Conversely, intensive agriculture practices required to maintain a staple economy (see also Hayden 1995: 62,63) require vast territorial assets an d very large labor pools. Elites may form inter group alliances to better manage and coordinate agricultural efforts, and the cooperative arrangements between several groups create a competitive advantage over any remaining smaller groups, as well as colle ctive control over a much larger population. Such arrangements m ay also reduce or eliminate conflict over resources between groups within the alliance. Cooperative arrangements may then be solidified over time as long term corporate entities through an ide ology that promotes the corporate bond. Institutional Competition Blanton, et al. present the two strategies as incompatible, even though they are commonly both used, at least to some extent, within the same groups (Beekman in press a ; Blanton, et a l. 1996:2), but usually among separate institutions (Beekman in press a:57). Blanton and colleagues further state that concurrent use of both inclusive and

PAGE 51

36 exclusive strategies may lead to conflict, and suggest that they may need to be separated at differe nt sites with very different functions for stability within the polity (Blanton, et al. 1996:7). A lthough these strategies may well co exist with some tension and conflict in their competition for site dominance, they do not tend to contend for the sa me resources As a result, their co location is possible and appears likely in areas where both staple and wealth trade resources are abundant, such as the above example for the Maya at Blue Creek At present we don't know the details of how institutions utilizing each of these strategies were coordinated within Blue Creek's political and economic systems Bourdieu views institutions as groups which occupy certain positions in social space which is "a space of differences, in which classes exist in some sense in a state of virtuality, not as something given but as something to be done (Bourdieu 2002:275, emphasis Bourdieu's). Institutional actions cross multiple social contexts that Bourdieu refers to as fields (Wacquant 1989:38 40) as they vie for dominance in the power structure Social actions also are played out in physical space, and are reflected in the patterns of spatial distribution of material culture in the archaeological record. Bourdieu's fields are not directly tied to physical space, but different architectural contexts are designed around different performative strategies that may include ritual, ceremony and other forms of social interaction (Beekman in press a), and therefore a designated physical space becomes an aspect of one or m ore fields. A multi arena, a built space where competing institutions engage. An arena can be as obvious as a literal battlefield, or as subtle as a discussion between members of opp osed institutions

PAGE 52

37 (1974:133). Community spaces are often regularly used for open festivals and other public rituals, and bring multiple institutions together to interact in one place. These spaces may regularly become ad hoc arenas for various combinations of institutions and fields. Arenas also are expected to eventually produce some ultimate decision over power, "even if it is the decision to leave things temporarily undecided" (Turner 1974:135). Despite what expectations may be present, the shifting bala nce of power may become an ongoing struggle, as has been the case with most Mesoamerican cultures between the Middle Formative and Postclassic periods which, according to Blanton et al. (1996:8 12) have oscillated between corporate and network strategies o ver time S ince institutions which employ different power strategies tend to utilize different kinds of assets, political competition becomes, in effect, a struggle over which form of capital in press a 57 60). The Nyari of West Mexico have demonstrated co location of networking institutions in the form of descent groups and a community administration, both of which have their own spatially separated ritual expressions of ideals intended to instill loyalty to the institution T he relationship has been far from stable, however, as the lineages and the community administration have been in constant conflict, and emphasis has shifted back and forth between the two strategies throughout their history, much like the situation described by Blanton et al. (1996) for much of Mesoamerican prehistory I nstability in the Nyari case, however, is mired in direct interference from the Spanish during the colonial period, and the dominant American governments through much of t he 20th century, which overturned and altered much of what may have been a pre existing institutional balance. Beekman ( in press b:7,8) has suggested that two factors may need

PAGE 53

38 to be in place for both institutions to coexist in a stable relationship: Explic it benefits of belonging to each institution need to be understood by its members, and some form of cross legitimization of institutions pursuing both strategies may need to be embedded in the group's ideology S uch cross legitimization, which may perhaps bind networking institutions, rather than individuals, more directly to the community institution may have been lost after Spanish contact Another problem in the Nyari case is that the traditional community ritual does not show any associated resou rces or capital which may benefit participants. T he only potential economic benefit is the elder responsibility of insuring the rainy season cycle but land and agriculture appear to be the sole property of the lineages (Beekman in press b:5, 7) Combini ng Economies: Context and Matrix Control Earle and Blanton each proposed a form of social separation between economies. Earle proposed a separation of wealth and staple economies by polity, and Blanton, et al. suggested the coexistence of the two eco nomies within the same polity, but within separate institutions and probably also separate sites. Kenneth Hirth (1996) has criticized this tendency towards compartmentalization of prehistoric economies: "The dichotomization between food and nonperishable commodities obscures the fact that (1) both food and luxury items play complementary roles in the development of political economies, and (2) elites may

PAGE 54

39 reorganize agricultural production to generate surpluses to trade for luxury items or control the prod uction of craft goods or luxury items to procure food." [Hirth 1996:208] Hirth developed a model which allows for the control and activities involved in both wealth and staple economies to co exist at the same site. Hirth's m atrix c ontrol p rinciple s uggests that elites may actively accumulate and diversify economies which they control even across wealth and staple boundaries as a risk management strategy. Elites accomplish this by including diverse strategies as separate lower level institutions withi n the social structure Production and accumulation of resources can then be managed under separate provisioning networks (Hirth 1996:224 225 ). However, the potentially conflicting ideolog y associated with a corporate institution must also be limited to th at institution, and made to work in the context of an over arching authority While the matrix control principle allows for coordination of different economies under the same social structure Hirth's context principle addresses the need for spatial separation of economic bases by suggesting elite supervised spatial contexts for the production and storage of wealth The principle does not require same site economic activity, since mobilization networks are likely also included in the system for accumu lation, but co location of activities would be advantageous to supervision and management (Hirth 1996:223 224) S imilarly, ritual activities for incompatible networking and corporate ideologies may utilize separate built space contexts, as suggested by Bee kman ( in press a)

PAGE 55

40 Under Hearth's model, relative emphases on each economy under elite control largely depend on elite choices, although other factors, such as fluctuating availability of staples and changing demand for wealth items, remain important considerations. This change s the view of the economic balance that we may expect to be represented within the material culture of a site, relative to expectations of the models of Earle and Blanton, et al. Rather than a categorical view of wealth or stapl e dominance, the data may reflect more subtle differences of material culture represented by each economy Hirth also identified several other economic categories that elites may attempt to control, although most (such as service economies) would not be re presented in the material culture, especially when confined to lithic data so Earle's wealth and staple categories will suffice as the more manageable division. Structure, Economy and the Natural Environment With all of the discussion on economic and related social dynamics in the literature, it often appears as if the natural environments of each respective social group are little more than backdrops in which these dynamics are played out. Environmental factors actually appear to have substantial effects on economic and strategic aspects of cultures (Blanton et al. 1996:7, Earle 1991:13 15) However, opinions about the environment's specific role in shaping cultures have historically varied quite widely on a continuum of cultural vs. environmental dominance C lassic processual environmental determinists have commonly viewed cultures as "spatially delimited bodies of individuals living within and adapting to a specific physical environment" (Schortman and Urban 1987:63) On the

PAGE 56

41 opposite end of the c ontinuum social theorists often see cultures as primarily driven by internally generated constructs of meaning which often actually instigate environmental change (Hodder 1982) S chortman and Urban have argued that the old processualist view ignores cultu res which live and interact throughout areas of highly variable environments (1987:63), and Chase Dunn and Hall have also suggested that a purely environmental distinction may work for some groups, but not others (1991:15 16) O thers argue that the opposin g concept of cultures as strictly innate, internally generated phenomena ignores evidence that environmental features combined with major climatic changes appear to have taken a large role, if not an exclusive one, in the formation and reconstruction of so cial structures (e.g. Chatters and Prentiss 2005 ; Shaw 2003). These few arguments are just a small sample of a very large debate which extends well beyond the intended scope of this paper. Many archaeologists have chosen to either seek a middle ground or bypass arguments related to the primacy of nature vs. culture, and instead concentrate on the nature of the interplay between culture and environment. Maya lithicist Jason Barrett asserts that "neither natural nor anthropogenic inputs into the system n ecessarily take precedence in the metamorphic processes that affected realized landscapes. There is instead a perpetual balance of actions and reactions that inhibit stasis in either nature or culture" (2004:53), and Townsend states that the need to acquir e resources required for survival is completely interrelated with the need for explanation and meaning (1998a:23) Others still place culture and environment on somewhat less equal grounds: Otto Schndube has conceded to at least some internally gene rated cultural influence in stating that geography is very highly influential, although not entirely deterministic (1998:205),

PAGE 57

42 and John Jackson (1984) assigns a very fundamental role to the natural landscape in determining the degree of social organization Some researchers state that corporate strategies tend to be pursued in environments where the most potential exists for agriculture (Blanton, et al. 1996:7; Hayden 1995: 62,63), and Blanton, et al. (1996:7) have also claimed that network strategies tend to occur where environments are marginal for staple goods, and also geographically in semi peripheral or peripheral areas. However, they also state that political reorganization may create networking opportunities in the core as well H ere again, the autho rs take a view where the environment plays a dominant role in determining the economic base, which in turn determines the strategic direction of elites, but with some exception made for political upheaval in creating economic change. Jackson (1984:15 0 151) makes a distinction which may also parallel core and periphery zones, between fragmented and irregular physical landscapes which he terms vernacular landscapes ; and orderly, well connected areas which he calls political landscapes V ernacular landsc apes are also reflected in a social and political absence of structure, and are "usually small, irregular in shape, subject to rapid change in use, in ownership and in dimensions". Political landscapes offer a sense of centrality and structure, which is re flected in a structured landscape that includes "such things as walls and boundaries and highways and monuments and public places" (Jackson 1984:12). Although the direction of this thesis does not directly address views regarding cultural or environm ental primacy, it recognizes the interplay between culture and environment as highly influential in the formation of a culture's organization and economic base, and utilizes environmental data alongside material cultural data to determine initial

PAGE 58

43 expectati ons for the questions being addressed. The supply potential is certainly a limiting factor in making decisions on an economic base in prehistoric contexts, but given multiple choices for viable economic pursuits (which may or may not be the case), elites w ill select and emphasize certain economic bases, based on multiple factors, including the demand for certain goods and in certain forms both within the core of a polity and among potential trading partners. Examining Semi peripher y Economies The the oretical question addressed in this thesis is whether prehistoric site s in economic semi peripheral zone s would have operated under a different emphases on economic bases, when the semi peripheral zones differ in the availability of natural resources and a ccess to internal groups. More specifically, it asks whether the semi peripheral area site that diverges the most from the core given the immediate presence of obsidian, lower access to arable land, and internal group access, emphasizes an economy based on wealth trade over one based on staple production According to Blanton, et al., "to understand social change of this type in marginal environmental cases the analyst must take a 'top down' view, placing the local system within its larger macroregio nal context to examine its role in the control and manipulation of intergroup exchanges" (1996:7). Studies of peripheral and semi peripheral area relations in general should also begin with detailed information on social and economical organization at all scales before we can interpret comparative patterns between sites (Schortman and Urban 1987:80). Therefore, the remaining chapters serve

PAGE 59

44 to tie background patterns discussed at the regional level, both inside and outside of the polity, to competing local s trategies reflected in the material culture evidence of the social arenas formed within public spaces. I t is hoped that this approach proves more useful than the usual research strategy of picking a scale of operation to work from, with no implications dra wn from above or below that scale (Beekman 2000:386).

PAGE 60

45 CHAPTER III REGIONAL BACKGROUND This chapter discusses environmental and cultural factors which may have influenced a semi periphery site's degree of interaction with, or independence from, the core culture; and whether different availability of certain resources may have l ed one site to de emphasize the core's staple economy and pursue wealth trade to a larger extent than a site associated with the core. Th is chapter begins with a descript ion of the natural environment within the Te quila Valleys including the distribution of resources that would have influenced the economic and cultural trajectories of peripheral semi peripheral and core areas. The public architecture of the area (guachim ontones and shaft tombs) is then described in relation to associated material culture evidence of economic strategies that appear to be employed within the region T he three political economic zones outlined in Chapter II are then described for the region occupied by the Teuchitl n and finally the sites under study (Llano Grande and Navajas) as well as the primary site of Los Guachimontones and the peripheral area of the Sayula Basin, are described in greater detail to determine more specific cultural and environmental characteristics related to each site's economic base.

PAGE 61

46 The Natural Environment and Subsiste nce Geology and Climate West Mexico's Highlands region is situated at the intersection of two ranges: The Trans Mexican Neovolcanic Belt tha t runs east west across the continent from the Pacific to the Atlantic at the Gulf of Mexico, and the Sierra Madre Occidental which follows the Pacific Coast (Beekman 2010:44; Darling 1993:figure 1). The area is characterized by topographic complexity, ext reme changes in altitude, a wide variety and abundance of inland fresh water sources and very fertile agricultural land that together produce a wide range of available resources (Beekman 2010:44; Schndube 1998:205; Stuart 2003:3; Weigand 1985:55 ; Weigand and Beekman 1998:37) L and and climate reliably capable of producing a substantial agricultural surplus is necessary for systems which rely on competitive displays of abundance for social positioning and political control (Hayden 1995: 62 63) T he geograph y of the Tequila Valleys provides numerous pockets of arable land fully capable of overproducing staple crops. The lake basins of the Late Formative and Early Classic period s primarily consist ed of large connected marshlands in the valleys surrounding the Tequila volcano (Weigand and Beekman 199 8:37) One exception existed within the Magdalena Basin on the west side of the Tequila Valleys where recent geomorphological surveys show that the majority of the area within the Magdalena basin was still a lake ra ther than marshland (Anderson et al. 2013:25). Still, large enough tracts of land were made available for agriculture in the area as the lake waters receded, and the surrounding mountains offered ecological diversity (Weigand

PAGE 62

47 and Beekman 199 8:37) H owever, rainfall is extremely seasonal in the area. 85 90 percent of annual rainfall occurs during the 5 month rainy season from June through October The season produces very heavy rains, with rainfall accumulations of 900 1600 mm per year The seasonal ity of the rain was likely not as much of an issue within the valleys in areas near the lake basins. The marshy nature of these areas and surrounding Lake Magdalena was likely similar to that around the remnants of the lakes that exist in the area today. S ome items retrieved during Adela Breton's 1895 visit reflect the prehistoric landscape of their day as well: bracelets acquired from a purported grave context were decorated with carved frogs (Townsend 1998a:15 16) Such images reflect the importance of ma rshland to the Teuchitln culture. The marshes acted as a stabilizing factor between the extreme wet and dry seasons by storing moisture for agriculture, which likely greatly extended the crop season. Geographic Area Relations to Political Economic Zone s Schndube has stated that the microclimates formed across the extremely varied landscape of Jalisco made for close access to a large variety of available resources, but there was not enough resource capacity within each microclimate to support large settlements beyond chiefdoms (1998 207,215; see also Fernandez and Deraga 1988), thus the West Mexico Highland area was more suited to small, scattered settlements. However, Schndube's assessment covers a very broad region, and the area of the Tequila

PAGE 63

48 Val ley s presents one exception as a relatively large, continuous ecosystem connected by the long series of lakes with abundant agricultural soils and other resources Economic Z ones of the Teuchitln C ulture S ocio economic roles do not automatically co incide with g eographic areas but the distribution of Teuchitlan architecture suggests a social pattern which closely relates to the geography of the Tequila Valleys Weigand (1985) defined a cultural core zone which covers the entire Tequila Valleys regio n, with a periphery that describes any Teuchitln culture outside of that region. Weigand also described an especially large and dense "habitation zone" within his core, to the south of the Tequila Volcano which in practice has been considered a sort of "c ore of the core" Following Beekman ( in press a), t he core zone within the Tequila Valleys is defined by a marked increase in architectural density across the landscape where distances between building and patio groups average about 100 m, and rarely excee d 250 m (Weigand 1985:82). The zone forms an arc of settlement that covers an area of 240 250 km 2 (Weigand and Beekman 1998:39, Beekman 1996b:136) T his is a smaller area than defined by Weigand and corresponds to his "southern habitation zone" (Weigand 1985). Also following Beekman ( in press a), The semi periphery corresponds to the remainder of Weigand's core, which include the areas to the north of the volcano, the Magdalena basin to the west, and the area within the Valencia Lake Basin to the Sou theast. Semi peripheral sites form a roughly defined outer core ring, still encompassed by the Tequila Valleys but much closer to the surrounding hills which mark the outer edge of the valleys (figure 1.2).

PAGE 64

49 The peripheral zone matches both Beekman 's a nd Weigand's definitions, and includes all Teuchitln culture sites outside of the Tequila Valleys. Peripheral sites are spread over a wide range within West Mexico, including the states of Jalisco, Colima and Guanajuato (figure 1.1) The natural land scape of the Tequila Valleys provides the first clue that semi periphery areas may have varied from the core in political and economic organization. It is perhaps easiest to visualize the Tequila Valley s landscape starting from the prominent central cinder cone known as the Tequila Volcano, and moving outward through what are now very dry lake basins which, at the time of Teuchitln occupation, were still mottled with receding lakes (Anderson et al. 2013:25). Across these basins from the volcano in all dire ctions is the irregular "ring" of hills and valleys, which contain several passes with limited access to the basins from the periphery The more level and expansive area of the core allowed for easier travel between sites This facilitated regionaliz ed polities to mobilize people from larger areas and allowed for greater corporate management of the labor intensive staple economy, whereas the fragmented hills and gorges of the semi periphery areas were more conducive to one on one or small group elite networking for wealth trade. Weigand and Beekman (1998:47) have suggested that the Tequila Valleys topographical features tend to foster a form of Jackson's political landscape (1984) with a strong tendency towards central control. The core landsca pe sharply contrasts that of the semi periphery sites in the surrounding hills, which describe a more disorganized and fragmented vernacular landscape also defined by Jackson. Even though the architectural landscape within the valleys has been viewed as mu ltinuclear, with several relatively

PAGE 65

50 independent city centers controlling or vying for separate areas within the territory (Weigand and Beekman 1998:51; cf. Martin and Grube 2000:18 21), areas within the core were likely more centralized than the population s within the topographically fragmented semi periphery areas. Subsistence Resources Maize was an important crop throughout the area, but nearly all subsistence studies for West Mexico within the last 20 years contain a recurring theme of variety in food sources rather than extreme reliance on domesticated maize. Pre Spanish contact flora resources were extremely varied and most are still found there today. At the time of Teuchitln occupation, the area likely saw an abundance of yams, avocados, sour sops, tomatoes, chiles, beans, various squashes, maize, amaranth, chia (sage seeds), papayas, Maguey (agave), nopal cactus (tunas and paddles), cacao and several other crops. F auna, however, were limited to dogs, turkeys and very localized groups of bees a nd ducks. The native animal species did not include the now familiar work and pack animals of the area, and no animals capable of functioning in such a capacity are known to have existed in West Mexican prehistory P eople therefore primarily travelled and transported resources on foot, over and around very steep and difficult terrain (Schndube 1998: 205 206). Even place names in the area reflect the abundance of resources: Mazatln, the place of deer; Michoacn, the place of fish; Zapotln, the place of z apotes. Archaeological evidence for the utilization of Maize, beans, chiles and squash goes back thousands of years and

PAGE 66

51 midden evidence has been located for apparent food use of over a 16 faunal species, including deer, small game, crustaceans and birds (S chndube 1998:208). The diversity of food resources within the Tequila Valleys was well exploited during the Teuchitln occupation of the area, but maize remained the most common crop. One environmental impact study of ancient maize fields in the area has estimated the degree of reliance on the common Mesoamerican staple relative to other crops at around 50 percent although somewhat lower near the lakes and wetlands where resources more directly associated with the lake ecosystem were plentiful (Beekm an and Baden 2011:359). Agave (maguey) flourishes in the high country within and around the semi peripheral zones where it is generally grown commercially in modern times (Chadwick 2011; Heredia 2008), and was likely heavily utilized in that area during th e Teuchitln occupation as well. Within the Sayula basin, prehistoric oven shapes and sizes most resemble those described in ethnographic data for the roasting of maguey pias (hearts), a practice which today is most closely associated with Tequila p roduction from the blue agave variety most abundant within the Tequila Valleys ( cf. Francisco Valdez 1998). Vessel forms often depict various food sources, sometimes prepared for meal service, such as roasted maguey leaf sections, squash, pitayas and orga n cacti (Schndube:209 210, figures 13, 14, 18 20, 22 24).

PAGE 67

52 Minerals Abundant mineral resources in West Mexico include obsidian, various forms of greenstone, various crystals, copper, silver, quartz, opal and salt (Weigand 1985:56, Weigand and Beekman 1998:37). Obsidian was one of the most utilized minerals throughout Mesoamerica during the time of the Teuchitln, and Jalisco has the largest number of obsidian sources in Mesoamerica (Clark and Weigand 2009:79). The Tequila Valley s are especially rich i n obsidian deposits (Beekman 2010; Esparza 2003:74,86; Stuart 2003:3), and contains over 30 obsidian outcrops, at least 12 of which appear to have been exploited before European contact (Spence et al. 2002:65). The Economic Impact of a Changing Climate Despite the apparent climate stability expressed by Schndube (1998:205 206), relatively slow and subtle climate changes over time may have profoundly affected cultural dynamics within the area. One question of interest is the cause of the expansion of the Teuchitln culture from the core areas to the hills and beyond, and what part a changing climate may have played in the decision to expand and establish peripheral and remote sites. Understanding the reason for the move may determine something about t he purpose of the remote sites Were people forced to seek alternate resources, or was the expansion an enterprising "power move" on the part of Teuchitln elites to gain better access to remote resources?

PAGE 68

53 Overall, the Late Holocene has been quite va riable in climate in the Highland Lakes area, with multiple wet and dry periods (Schndube 1998:205 ) T he drying trend which began after the Late Formative Period may have profoundly impacted people living in the region from the Early Classic to the end o f the Teuchitlan culture at the beginning of the Middle Classic (Anderson, et al. 2013; Beekman 2010:61). Whether or not drought was motivating factor for culture change is unclear. However, Early Classic settlements in the hills of the semi peripher al zone would have required different subsistence strategies than settlements in the valleys, although hills would not have been entirely agriculturally impoverished. Maguey thrives in hot, sunny and arid climates and is quite tolerant of poor soils and dr ought (Chadwick 2011 ; Knox 2013). The succulent also tolerates occasional moderate frosts down to about (Chadwick 2011). The Jalisco high country, and in particular, the hills north of the Tequila Valleys, are considered an ideal climate for m aguey, and are currently one of the main regions for growing the blue agave variety commercially (Chadwick 2011 ; Heredia 2008). The agricultural fields in the valleys south of the volcano (including the former Teuchitln core), in contrast, are primarily p opulated with maize and sugarcane. Economic Strategies Within the Teuchitln Culture The two prominent forms of ritual spaces among the Teuchitln, shaft tombs and guachimontones, appear to support the two economic strategies outlined by Blanton, et al. (1996). T he shaft tombs reflect ed a network or exclusionary strategy manifest by descent groups which is associated with a wealth finance economy In contrast the

PAGE 69

54 guachimontones were the setting for more group oriented activities associated with stap le finance (Beekman 2000:403 404; in press a:57 60) T hese economic bases and their relationships with the ritual a rchitecture are described below. Staple Finance Guachimontones: Corporate Management of a Staple Economy. The ritual architecture of t he Teuchitln culture, the guachimontn, is composed of a circular patio with rectangular building s arranged around the circle's circumference, and usually a circular stepped pyramid altar in the center (Beekman 2008:419, 2010:62; Weigand 1985:66 69; Witmo re 1998:138; figures 3. 1 3. 2 ). Ceramic dioramas depicting Guachimontones, as well as wall remains on the actual patios, show the presence of buildings surrounding the patio. Dimple shaped holes were discovered on the top surfaces of the altars within Los Guachimontones (Townsend 1998b:110) and in the center of the patio at Llano Grande, and some clay dioramas show the presence of a vertical pole in the same positions. The pattern has been the subject of much discussion on the rituals it hosted (e.g. Beekma n 2000, 2008; Weigand and Beekman 1998; Whitmore 1998) The earliest evidence of occupation at guachimontn sites is from around 300 B.C., and construction of guachimontones started some time later, by 100 B.C. (Beekman 2010:63). The level of effort required for these circular constructions is substantial. Even the moderately sized 8 building circles would have involved labor efforts in the range of hundreds of thousands of person hours. Construction of the circles at the site of Los

PAGE 70

55 Guachimontones is estimated to have required about 1.25 million person hours of labor (Beekman 2000:395, 397). Despite the varia tion in architectural details, guachimontones follow several patterns relatively consistently across the landscape throughout the region T he num ber of buildings around the patio varies, but most have 8 satellite platforms and other variations are limited to 4, 10,12 or 16 platforms, all arranged in opposing pairs (Beekman 2008:419) D ifferent construction techniques were used for each building in Figure 3. 1 : Circle 2 guachimontn from the Los Guachimontones site.

PAGE 71

56 Figure 3. 2 : Navajas Circle 5 guachimontn plan view. several documented cases, so the buildings were most likely constructed by different groups, with different learned techniqu es (Beekman 2008). Beekman has suggested that the groups representing each building on the circle were likely competing lineages (2000:414, 415) Material culture within the guachimontones tends to be less ornate compared to artifacts within the shaf t tombs described below, possibly due to suppression of aggrandizing behavior in a space dedicated to cross institutional unity (Beekman in press a:63) F or example, Catherine Johns' ceramic ware study (2014) from one core guachimontn complex revealed tha t the majority of the sherds by weight were of a plain utilitarian ware related to cooking and storage ( Colorines ), or a very uniformly painted

PAGE 72

57 ware associated with communal feasting ( Arroyo Seco ). Relatively few of the sherds in each building were determi ned to be from more elaborately decorated forms ( Tabachines and Estolanos) typically associated with aggrandizing ritual behavior (Johns 2014:figures V 12, V 13) The social use of guachimontones appears quite varied from ceramic dioramas found withi n tomb contexts, which depict different scenes of dancing, drinking, playing music, apparent marriage ceremonies and rituals (Beekman 2003b:12, in press a :63,64). dioramas also depict apparent funerary processions crossing a guachimontn (Beekman 1996b:136 in press a :62). Clay dioramas found in some tombs depict scenes of dancers around the pole (Townsend 1998b:111) Several archaeologists have proposed that the guachimontn architectural form symbolizes agriculture or fertility (Beekman 2003b, in pr ess b), and that elites were involved in the staple economy through frequent agriculture related rituals held within and around the guachimontones (Beekman 2000; Butterwick 1998; Lopez and Ramos 200 6a,b; Schndube 1998). These proposals suggest symbolism o f economic aspects that extend beyond the guachimontones themselves, and are described in the following sections. C oncentrations of obsidian often appear in ceremonial centers, and visual sourcing most often points to nearby area mines as the sources for collections within the centers. Therefore, any elite control of obsidian was likely not centralized Spence et al. (2002:68) ha ve suggested that elites controlled obsidian on a per area basis from secondary guachimontn sites that oversee each area. C ontrol of obsidian may have been an important component in the political dynamics of the area (Spence et al. 2002:68), and

PAGE 73

58 Weigand ( 1985:89; 1996:96; and others) has stated that guachimontones are also often associated with obsidian workshops a lthough Wei gand's evidence is very preliminary since it is based on surface survey reports I f further excavation verifies Weigand's observation, the association reflects elite interest and likely control of the obsidian production industry. Agriculture P roduction Beekman (2003b:14,17) has suggested that the guachimontn architectural form may symbolize maize, since the pattern is quite similar to a cross sectional view of an ear of corn. In this view, the circular patio represents the cob and the surrounding bu ildings represent kernels F or the Harinoso de Ocho variety of maize, a likely common variety for the Teuchitln culture, ears most commonly carried eight rows of kernels but ten and twelve kernels are also found on occasion. Likewise, most guachimontones in the core and semi peripheral areas have eight surrounding buildings, although ten and twelve building guachimontones are occasionally found. Maize is a very high yield crop on which Mesoamerican people heavily depended, but also highly vulnerable to cli mate and soil conditions I t is therefore considered a high risk crop. Attempts to mitigate the risk to the amount of surplus required for elites to maintain their political status probably relied heavily on cosmological solutions (Beekman 2003b:18). Others have suggested that agricultural fields themselves show evidence of production intensification. Weigand (1993:228) has interpreted cross hatch patterns in the T equila V alley basins as artificially raised agricultural fields called chinampas Convers ely, Stuart (2003) has suggested that the pattern is actually a series of water management

PAGE 74

59 channels, based on a lack of sufficient sediment for chinampas. Stuart also warns that his uncertain dating of the fields to the Terminal Classic period should be tr eated "with extreme caution" (Stuart 2003:241) and a geomorphological study of the lake basins in 2013 demonstrated that Stuart's canal system would have dried up long before his proposed termination dates (Anderson et al. 2013). Stuart's uncertainty of t he dates, however, opens up a possibility that agricultural channels could have been utilized much earlier. Beekman ( 2010:61) also suggests that The Early Classic Period expansion of the Teuchitln culture spread ideas of agriculture based ritual and social heterarchy into the semi peripheral and peripheral areas, which might have fundamentally transformed economies throughout West Mexican into a regional agrarian system if the gradual drying trend leading to Epiclassic drought had not already begun to affect the local environment. The drying trend also may have influenced the semi peripheral zone sites to at least partially replace or augment their staple finance with a greater emphasis on the wealth economy. Feasting Butterwick has proposed that ritual feasting, an integral part of a staple economy, was pervasive throughout the Teuchitln culture. The archaeological evidence suggests that the practice goes back at least as far as 100 B.C. (Butterwick 1998:99). Numerous depictions of food in ceram ic art suggest that feasting was a deeply ingrained aspect of the West Mexican way of life since the Late Formative period, and some figurines suggest the ritual consumption of aguamiel ( maguey nectar) or it's fermented form pulque

PAGE 75

60 out of a container (Butt erwick 1998). The contents of shaft tombs, particularly at the semi peripheral site of Huitzilapa (Lopez and Ramos 2006a,b), have also suggest a related component of large food offerings for the dead, likely related to food for the deceased and funerary di splays (Beekman 2000:393). Wealth Finance Shaft Tombs: Networking and Wealth Trade Shaft tombs are often associated with Guachimontones, but are more frequently distrib uted outside the Tequila Valleys than Guachimontones (Weigand and Beekman 1998). They are nevertheless unique to West Mexico within Mesoamerican contexts, although tombs of similar construction also appear in South America (Anawalt:1998:238). The tombs are composed of a vertical shaft which leads down several meters to one or more cha mbers (figure 3. 3 ). These chambers house multiple burials and may include numerous grave goods S haft tombs have been of interest to both archaeologists and collectors for much of the 20th century, but until the last four decades, the interest has only bee n due to the fact that they are the most productive caches of figurines, ceramic vessels, obsidian objects and other artifacts for modern day collection, trade and sale (Beekman 2000, 2008; Weigand 1985) S till, much of what we have learned thus far about the culture and economic relationships within West Mexico have been derived from the contents of shaft tombs S haft tombs have offered some of the best preserved contexts due to their locations several feet underground, and in relatively stable soils that avoid early collapse. But looted shaft tombs still far outnumber

PAGE 76

61 Figure 3. 3 : Profile of a shaft tomb from the Atitlan Las Cuevas site. From Beekman 2000:391 figure 4. those excavated by professional archaeologists, and Guachimontones have suffered the same fate P ublished artifact data from actual tomb contexts is therefore very thin (Beekman 2000:386). But when found in context, some details of grave goods can be used as clues regarding ideologies and social relationships, and a differential distr ibution of goods between interments can provide clues about the organization and degree of social stratification. Shaft tombs appear to have been used only for a certain portion of the elite population; overall, only about 10 percent of the populatio n appear to have been buried in the tombs, with the rest interred in much simpler pit graves (Beekman 2000:391) L ineage appears to

PAGE 77

62 have been emphasized over individuality in shaft tomb interments. Mu ltiple individuals are sometimes buried within the same tomb, and genetic testing on human skeletons interred within tomb chambers has shown that individuals within each chamber are biologically related (Beekman 2008:418). Middle to Late Formative shaft tombs are quite variable in construction with three ranks defined by Weigand according to size, elaboration and depth, which varies from 4 to 17 meters. Shaft tombs are often found closely associated with Guachimontones and in some cases underneath guachimontn buildings. They are also often interspersed am ong common pit graves (Weigand 1985: 64,66,128) S haft tombs become smaller and contain less wealth in offerings in the Early Classic (Galvn 1991 as found in Beekman 2000:396). Funerary processions are depicted by some ceramic dioramas found within the tombs. Processions appear to function as displays of the wealth offered to the interred, which provides a way to advertise power as elites negotiate their position in the social structure (Beekman 1996b:136; Beekman in press :62). Shaft tombs instilled Chase Dunn and Hall's idea of "differentiation" (1991:19), a social difference of prestige, the ability to mobilize people, and economy rather than one of overt territorial control. The presence of prestige goods within the tombs also illustrates Blanton et al. 's (1996) networking/exclusionary strategy operating through a network of kin relations. Obsidian eccentrics and other status items within the Teuchitln culture are much more concentrated in elite locations, such as shaft tombs and guachimonto nes in Late Formative and Classic period contexts than in earlier contexts; elites appear to have exercised a much greater degree of control over objects of authority during these periods

PAGE 78

63 (Beekman 2010:63). During the later periods, e lite grave offerings included imported goods from areas across the Mesoamerican continent and include seashells from both coasts jade beads and cotton (Beekman 2000:394, 2010:63, Lpez and Ramos 1998:61). Differential access to wealth goods is evident in the different distributions of goods between tombs, and even chambers within the same tomb, which shows clear evidence of a fully established social stratification ( Lpez and Ramos 2006b), and an exclusionary strategy. Obsidian artifacts in shaft tomb contexts include cruciforms, lunates, thin circles, pendants, large double pointed knives, mirror backs, beads and ear spools (Weigand 2000:49). Ceremonial ceramics and other wealth goods are also extremely variable in style within each chamber collection, and Butterw ick has suggested that the diffused ceramic distribution pattern is likely due to frequent trade and/or gift giving, possibly during funerary and mortuary events (1998:104 105). The practice of staple financed feasting may therefore have actually encourage d internal wealth trade by simply providing a social context for the interaction of elites between groups, even if the corporate ideal officially discouraged networking practice. Long distance Trade Evidence of foreign goods within shaft tombs indica tes that long distance trade intensified through time to support the rise of an elite class S ome have suggested specific long distance trade contacts from South America based on material culture similarities and travel feasibility studies (Anawalt 1998; C allaghan 2003), and Weigand (200 0 2008) has proposed trade with the American Southwest based on artifact similarities and the

PAGE 79

64 presence of high quality turquoise. However, the Southwest connection has been considered doubtful, since the Southwest had close r access to obsidian and the cultur al similarities to West Mexico do not extend as far as the Tequila Valleys. Also, Weigand's turquoise was judged on visual aspects. Since existing compositional sourcing techniques for turquoise have not been successful, Weigand's assertion cannot yet be confirmed. Obsidian's Relation to the Teuchitln Wealth Economy Jay Johnson (1996:171) has suggested that obsidian in general may occupy an intermediate status between a staple good and an elite item throughout Mesoa merica, since obsidian objects have been consistently found in highest concentrations within elite household and ritual contexts, yet are also present in commoner households. Obsidian is also made into both utilitarian and ritual or eccentric objects. With in the Tequila Valleys, utilitarian forms of obsidian as well as raw or prepared cores can be considered staple goods because of the material's ubiquitous presence throughout the area. More finely worked eccentric or ritual forms, however, were more likely produced by specialists and therefore easier to control by elites. We must also distinguish between obsidian designated for internal use, and that which is prepared for export to areas for which the material is in much greater demand. The production of fi nished goods from fine quality obsidian within consumer areas in other parts of Mesoamerica appears to most frequently occur within elite contexts ( Johnson 1996 ; Spence 1981). Traders seeking the material would need to be connected with remote elite groups Obsidian bound for export may then be considered a wealth good, since it can be used to obtain elite imports. Therefore,

PAGE 80

65 obsidian items considered wealth economy goods are restricted to external trade items, and all eccentric items whether internally or externally circulated. Economies and Ritual Space The material culture evidence demonstrates that Teuchitln elites relied heavily on both staple and wealth economies to maintain their power. The evidence for corporate crop related ritual and feast ing is abundant, as is the evidence for the exchange of wealth goods throughout, and possibly even beyond, the Mesoamerican subcontinent. Considering the distribution of wealth and staple related material culture as it relates to the shaft tombs and t he Guachimontones, it appears that the primary purpose of the Guachimontones is for corporate festivities and rituals mainly related to the staple economy, and the primary purpose of the shaft tombs is the display of wealth in the context of the social net work. However, unlike the degree of spatial separation predicted by Earle (1991) and Blanton, et al. (1996), both wealth and staple economies appear to not only coexist within the same sites, but in some ways can even complement each other, even within the same ritual spaces. The configuration of the Guachimontones reflects the formation of corporate groups depicted by the central patio and altar, out of the institutions that form the wealth economy ; the kin groups which build and occupy the surrounding bui ldings. As Butterwick (1998:104 105) has noted, corporate rituals and festivals in the spatial contexts of the Guachimontones provide a potential catalyst for intergroup wealth trade among elites of different subgroups. The result is an internal site netwo rk formed from the elites of each kin group which resulted in the wide dispersal of

PAGE 81

66 different forms of wealth items, but still only among elite members of each group. Guachimontn patios may then act as arenas for trade and other interactions, such as inte r group marriage relations and political maneuvering. Also, the use of ritual items in the corporate ritual contexts, including elite attire jewelry and symbols of power, is demonstrated in some of the clay dioramas (Anawalt 1998:figure 1;Weigand 1996:fig ure 9). Conversely, shaft tombs display large quantities of food items alongside the various wealth goods, as the perceived need for staple supplies (and the elites' perceived ability to provide them) extends well beyond the earthly plane (Lopez and Ramos 1998, 2006a,b). Timeline Carbon dating combined with ceramic seriation has determined that the guachimontones were first constructed as large centers in and near the core during the Late Formative period, starting around 100 BC Much smaller guachi montn construction later spread into the surrounding hills and more distant outlying areas, as far as Colima near the Pacific Coast and into all neighboring states, around AD 200. Around the same time, new construction within the core ceased, and only evi dence of structure modifications appears for the core area during the Early Classic period. Based on a lack of evidence for any later maintenance of the structures, the culture is currently believed to have collapsed around AD 500. (Beekman 2010:64; Beekma n and Weigand 2008:315). The shaft tomb culture also saw its peak in the scale and elaboration of construction, as well as the quantity of wealth offerings, roughly at the time of the earlier

PAGE 82

67 guachimontones M uch like the g uachimontones, the size of shaft tombs decreased in the Early Classic period, as did the quantity, size and wealth of tomb offerings. Several timelines with different phase definitions have been created for different areas within the Tequila Valleys. Recent phase definitions for the sites and periods of interest is included here for general reference (Table 3.1). The periods of interest here are Tequila III (100 BC AD 200) and Tequila IV (AD 200 500). Although timelines between different sit es are not well synchronized over long time periods, an AD 200 transition is common to several locations and aligns well with the transition from the Late Formative Period to the Early Classic used throughout most of Mesoamerica. This thesis most often reference s the most universally understood Late Formative and Early Classic Mesoamerican periods, following Beekman (2010). Distribution of Architecture Even though the main features of guachimontones and shaft tombs are evident throughout the Tequila Valleys, more specific architectural di fferences exist between the core semi peripheral and peripheral areas These differences may show different relative emphases on different economies by reflecting d iffering ideologies promoted by aggrandizing wealth oriented political organizations rather than corporate ones. Ideologies which stress community are directly tied to Blanton, et al.'s (1996) corporate/inclusion economic strategy and therefore more likely target a staple economy. Aggrandizing systems which elevate one group above others at the highest structural level may be more likely to pursue a network/wealth trade economy. Although both

PAGE 83

68 Table 3.1 : Site occupation spans and related periods relevant to the current study Date Mesoamerica (Beekman 1996b:figure 3) Tequila Valleys (Beekman 199 6b:figure 3) Los Guachimontones (Beekman and Weigand 2008:figure 9) Sites Under Study (Beekman and Weigand 2008: f igure 9) 400 500 A.D. Middle Classic Tequila IV Ahualulco 300 400 A.D. Early Classic 200 300 A.D. Llano Grande 100 200 A.D. L ate Formative Tequila III El Arenal Navajas Circle 5 0 100 A.D. 100 B.C. 0 200 100 B.C. Tequila II 300 200 B.C. 400 300 B.C. Late Tequila I San Felipe economies are evident in the architecture of all three zones, s ites appear to become relatively less corporate and more wealth oriented with increasing distance from the core. Core Sites The Teuchitln culture 's peak population period within the core settlement zone has been very tentatively, although conservat ively, estimated at approximately 40,000 people (Beekman 2008:416). The overall settled area covers 24 km 2 with 30 km 2 of proposed

PAGE 84

69 Chinampas (or canals, according to Stuart [2003]) and 300 km 2 of terraced fields (Beekman and Weigand 2008). Many of these cl aims and figures have not been substantiated by excavation and proper dating, so these data must be viewed with caution. At the latest published count, 38 ceremonial centers have been verified in the core population zone (Beekman in press a:69) The forms and sizes of buildings surrounding the guachimontones in the core are quite uniform. No buildings stand out as being larger or more elaborate around or near the circle (Beekman 2008), which points to the representation of a likely corporate power sha ring system between groups represented by each building (Beekman 2008). Shaft tombs within core and semi periphery areas can be much more elaborate and labor intensive, contain more grave goods than those on the periphery, and vary more in elaboration and wealth than more distant tombs (Beekman 2000:389 390). Semi peripheral Sites Beekman notes that the more distant sites within the sem i periphery i nclude a few guachimontones with unusually large residential buildings or a single disproportionately l arge building on the edge of a guachimontn Semi peripheral sites also more frequently includ e shaft tombs underneath guachimontn buildings, all of which reflect self aggrandizing behavior and an increase in emphasis on descent groups (Beekman in press a :70,71). Six especially distinctive sites in passes overlooking the valleys have the greatest exposure to outside groups (Beekman 1996a,b ; Weigand and Beekman 1998) F our of

PAGE 85

70 these sites are concentrated in the La Venta corridor to the east, and incl ude Estolanos Mesa, Mesa El Zacate, Cerro Tepopote and Peol Tepopote. The remaining two sites, Cerro Pipiole and Llano Grande, are in Magdalena basin valley passes to the west (Beekman 1996a:984 985, 1996b) B eekman suggests that all of these sites were e stablished to monitor their corridors as trade and communication routes. Based on the scale, substantial nature of the architecture and amount of artifacts and debris at each location, the sites appear to be full time settlements which range in size from j ust 5 6 architectural features at Estolanos Mesa, to 120 features for the largest site, Peol Tepopote A ll four of the La Venta sites contain wall features that either cross the pass or surround the site (Beekman 1996b:139 140), as does Llano Grande (Beek man 2001:4 5). No wall structures are mentioned for Cerro Pipiole, but the site has yet to be examined in detail. The appearance of walls in semi peripheral areas is one of a few archaeological clues that can be found of a group's attempt at defining social boundaries (Chase Dunn and Hall 1991:18), and Beekman asserts that these semi peripheral sites form a controlled boundary surrounding the core area (Beekman 2000:404). Beekman has also noted that the architectural and functional redundancy of g uachimontones throughout the Teuchitlan culture suggests more of a segmentary state than a unitary one, as each site shows emulation of the largest core sites and functional independence, rather than a functional specialization that requires centralized co ntrol expected from a unitary state (Beekman 1996b:136) Y et, if the walled semi peripheral sites do indeed represent boundaries, then core elites may have attempted to increase their power by encompassing and managing the semi peripheral areas as part of a conversion

PAGE 86

71 of the political system to a more unitary state. If the area had become more centralized, then economic activity in semi peripheral areas would have been directed towards the core, rather than external groups. As Kowalewski, et al. have pointe d out, such transformations may not always be completely successful due to the required resources involved in controlling remote areas (1983:37), especially for remote locations which are less accessible across steep hills, marsh ringed lakes and arroyos. A few semi peripheral sites also show differences relative the core in the proportions of guachimontones themselves. These circles include a single enlarged building which may reflect one descent group becoming dominant and gaining control over the r itual aspects of the entire group (Beekman in press a:64) T his also reflects a move towards a vertical structure that favors aggrandizing elites, and likely accompanies an increase in wealth/trade aspects of the local economy W eigand ( 1985:66) has stated that the most elaborate shaft tombs were located in the peripheral area of the Atemajac Valley east of the Tequila Valleys, and that the remaining tombs he had discovered by 1985 were "3rd rank", meaning less than 4 m deep and only one chamber Four of fiv e sites known to contain shaft tombs beneath guachimontn buildings are in semi peripheral areas (Beekman in press a :70). The guachimontn tombs are larger and more elaborate than those found in cemeteries. They also contain more variable and numerous grav e offerings, and are more often reused for multiple interment. (Beekman in press a :62). Among the more elaborate semi peripheral area examples is the tomb at Huitzilapa with an 8 m shaft, and a total of approximately 60,000 artifacts (Weigand and Beekman 1 998:39). Beekman ( in press a :62) suggests that kin relations were more emphasized at these sites, along with descent related claims to ceremonial positions related to the public

PAGE 87

72 architecture, whereas cemetery tombs more consistently found in the core relat ed elites to corporate control of labor and land capital. Peripheral S ites Peripheral guachimontones are underreported due to a lack of comprehensive, formal surveys in West Mexico, but so far two to three dozen sites have been found in Bolaos Cany on several have been discovered in southern Jalisco and Colima and one has been located in Puerto Vallarta (Beekman 1996a:88,89) These sites range from approximately 50 km to at least 200 km from the core. Guachimont n circles have also been found in th e state of Guanajuato (Beekman 2000:figure 10, 2003b:5 ) Peripheral sites tend to be located at strategic positions along transportation routes on the way to desired resources (Beekman 2003b:5). P eripheral and semi peripheral areas were very sparsely populated, and Beekman has suggested that they likely emphasized exclusionary/networking strategies due to a lack of a labor pool to effectively implement a corporate strategy ( in press a:73). Varia tion from the core architecture is most evident in periph eral sites. They have a much more narrow guachimontn size range, are not as well constructed and are the only Teuchitln sites with ballcourts. The Tequila Valley s are also more agriculturally rich than any of the peripheral areas, and is the only area th at shows evidence of intensive agriculture practices (Beekman 1996b:143). Peripheral sites in Bolaos Canyon appear to have incorporated local architectural styles into the standard concentric circle pattern well established in the Teuchitln core. Beekman interprets these remote sites as local elites who have incorporated the guachimontn architecture through

PAGE 88

73 the regional trade system in order to create an alliance with the Teuchitln core, and likely adopted some elements of the Teuchitln culture (1996b: 143 2000:400 401 2010:64). The remote sites to the northwest in Bolaos Canyon show the most variation from the core g uachimontones similar sites as a circular layout with four square mounds surrounding a square central mound T hree of the four satellite mounds appear quite uniform, but one mound to the northwest is of an unusually oblong shape with a much smaller mound directly in front of it (1903:392 393,figure 9) A few peripheral excavated guachimontn circles contained burials under the central altar ( Cabrero 1989:149 161, 187 195 ; 395; Kel ley 1971: 770 771), which strongly suggests that a single aggrandizing descent group may have claimed ritual authority over its entire community. Figurines from Ixtln del Ro of the Nayarit highlands show apparent elite rulers and/or religious leaders, on e male and one female, holding scepter like objects which were probable symbols of authority. The t op of the m ale figure's object depicts an apparent abstract model guachimontn that includes the center pyramid, circular patio and four surrounding temples (Anawalt 1998:Fig. 1; Beekman 2003a:313). Although the various guachimontn activities would have been a binding force of the corporate system, they may simultaneously have also contributed to its undoing. The interactions take place in the context de signed to bring separate and normally competing institutions together. Therefore, all corporate gatherings designed for unity likely included a layer of negotiation and political positioning as a manifestation of Turner's

PAGE 89

74 arena (1974:133) S ocial manipulat ions may involve subtle, seemingly casual strategies within festive settings, or more formal means such as the funeral processions which involve parading grave goods of a particular lineage in sight of all participants, as a display of the lineage's wealt h and potential influence (Beekman in press :62). Over time, certain groups may have become more prominent, at least in part by utilizing better access to wealth economies to out compete other groups, especially in areas with less potential for staple finan ce. Jackson's (1984) description of vernacular architecture as "usually small" and "irregular in shape" certainly matches the observed pattern in semi peripheral and peripheral sites of the Teuchitln culture, especially when compared to those of the Teuchitln core, and Jackson suggests that such sites are "subject to rapid change in use, in ownership and in dimensions" (1984:150 151) W e do not know what changes may have occurred within the semi peripheral sites following their establishment, but the y were likely dynamic entities which underwent continual change in their political and economic identities from both core and external influences. Thus far, interpretations of the difference in use, ownership and dimensions of semi peripheral and periphera l sites from those within the core have favored a greater emphasis on networked wealth economy of emerging dominant descent groups. The difference can still be seen as one of degree rather than type, since the relative influence of more powerful elites wit hin a site may vary, along with the degree of emphasis on alternate economies to which they may be linked. Some semi peripheral and peripheral guachimontones vary from those in the core, they are still very much recognizable as guachimontones and thus stil l represent some degree of adherence to a corporate ideal, where a slight modification to Orwell's

PAGE 90

75 adage that "all descent groups are equal, but some are more equal than others" may well apply. Beekman has recommended that "further research, utilizing extensive excavations, should focus on this variation [in core and periphery political strategies], as well as on the multiple roles of boundary settlements" (1996a:998), and the present research can be viewed as a step towards the goal of understanding t he role and nature of the Teuchitln semi peripheral sites. Descriptions of Compared Sites West Mexican archaeologists have very generally described the region as rich in certain agricultural and mineral resources. Yet, very little specific data on geology, soils and land use have been published for West Mexican archaeological contexts. The availability and quality of natural resources can change drastically between distant sites within an area as complex as the Tequila Valleys, so economic opportuni ties at the site level need to be assessed with consideration of very localized natural resource details. Environment details were therefore examined for the primary core site (Los Guachimontones) and the sites under test (Llano Grande and Navajas), and th e pertinent data are summarized below L and resource descriptions are taken from official Carta maps of the specific areas in question with some observations made at each site (excluding any at the Sayula Basin) during the 2011 summer lab season Ele vations are also compared between sites to determine whether differences are large enough to effect available resources S ince elevation can profoundly affect available

PAGE 91

76 resources, both topographical maps and an interactive tool ( http://www.altitude maps.com/ ) were used to find elevations and sloping trends where they are not specified in the literature. The altitude maps tool finds elevations at more precise points than topographic al maps can provide, but uses low resolution SRTM data which can show error rates of up to 23m on steep slopes (Thomas, et al. in press ; Tighe and Chamberlain 2009). The resolution is not as high as desired, but suffices for the current purpose. Fea ture and layout details of the central site (Los Guachimontones), and also sites under test (Llano Grande and Navajas) are given here to better characterize the sites within the context of the Teuchitln culture, and also to provide some context for the ex cavated features (Llano Grande Feature 14, and Navajas Circle 5) and the analysis of the lithic artifacts recovered from those features. The lithic analysis can then be used to further describe the economy of each site relative to the ecological and cultur al conditions that the site represents. Los Guachimontones If some level of centralized control over semi periphery sites such as Llano Grande did exist, it would likely have come from the primary site of Los Guachimontones just south of the Tequil a Volcano, and near the center of the core (figure 1.2) T he site likely influenced the entire core area at least culturally if not politically, and therefore is described in some detail here.

PAGE 92

77 Site Environment Los Guachimontones is on a hillside nea r the edge of marshland within the La Vega lake basin C urrently, Los Guachimontones contains mainly seasonal farmland to the immediate south, and mainly brush land to the north and west D uring the Late Formative and Early Classic periods, much of the lan d immediaely below the Los Guachimontones settlement area to the south was marshland (Beekman personal communication, September 2013). Most of the soil in the area is of mixed quality for farming, except for an area about 1 km to the southwest, off the nor thwest shore of the lake, with deep, humus rich soil. On the opposite side of the public ritual center from the lake is a bank which runs southwest to northeast, directly behind the largest circle. Behind the bank is another area of very good soil, which m ay have been partly used by another Teuchitln site on the opposite side of the field (figure A.7). The elevation of the ritual center is 1,374 m, taken at the center of Circle 1. Over the bank above the large circle to the northeast, the field slopes downward away from the bank from southeast to northwest, and from an elevation of approximately 1,460m to 1,370m. Description The site's ceremonial center consists of ten circles and two ballcourts, with what has been described as a "miniature shaf t and chamber tomb" in one of the older and more moderately sized circles (Beekman 2008:426 428). Los Guachimontones is by far the largest site in the region(Beekman 2008:427) C onstruction details are described as being similar to Navajas, although the di fference in construction actually appears more similar

PAGE 93

78 to what is found in Navajas Circle 5 with differences in construction material and method details for each building. The public architecture is quite large in proportion to the size of the reside ntial area, suggesting that it was a more central administrative center, possibly controlling a second tier of smaller major centers within the valley s (Ohnersorgen and Varien 1996:fig. 9), and the core population density also suggests that Los Guachimonto nes was part of a larger system, which Weigand suggests controlled six habitation zones via their respective administrative centers in the core area (Ohnersorgen and Varien 1996:103) W eigand's assertion again brings up the question as to how far Los Guach imontone's control actually extended beyond the south central core area. Ohnersorgan and Varien (1996) borrowed a model successfully used by geographers to determine virtual "boundaries" for segmentary groups, and also tested by Alden (1979) on Aztec group data, to determine the potential political reach of Los Guachimontones. The model depends largely on the distance between sites, and the size of each site. Llano Grande wasn't included in Ohnersorgen and Varien's study, but closer and larger groups w ere determined to be "isolated clusters" in scenarios where the culture's willingness to travel long distances was similar to Alden's Aztec case Therefore Llano Grande would likely not have been included as a centrally controlled site according to the mo del N avajas is half the distance to the cultural core area and on an easily accessed plateau so it is more likely to interact with, and therefore conform to, the core area. Even though the concept has been used successfully to determine spatial limi ts of segmentary groups in geographic contexts, the model has not been subject to repeated testing for an archaeological context, the results should be treated with some caution.

PAGE 94

79 Also, Ohnersorgen and Varien recognize the lack of any interpretation of the nature of the interaction between sites, and suggest that clarifying craft production/distribution systems, perhaps particularly in the especially prominent obsidian industry, can go beyond descriptions of the structure of the interactions to determine mor e detailed information about their nature (1996:119) Navajas Site Environment Navajas is located in center of current farmland on a gently sloping, round, nearly circular plateau of bedrock about 1 km 2 in size surrounded by a high altitude plain. The site is also just outside of the semi peripheral area hills but with very close access to the core near a pass (Figure 1.2). The Navajas g uachimontones are on shallow bedrock, but agriculturally rich fields of marginal to ample depth form an expanse s imilar in area to the arable fields around Los Guachimontones. Arable land extends north from the ceremonial center, and also exists west of the site. Some poorer, loose and granular soil is intermixed in the northern half of the field. The rough circle of bedrock is mainly andesite, an extrusive igneous volcanic rock T he elevation at Navajas' Circle 5 guachimontn is 1,538 m, with a gradual slope from 1,592 m down to 1,458 m across the 2 km bedrock circle diameter (approxim

PAGE 95

80 Description The site of Navajas was excavated during the 2003 field season. The site is located in an area just outside the Tequila Valleys (Beekman 2008:424), on a plateau in the small basin surround ing Lake Valencia. The s ite complex contains at least 10 guachimontn circles, one ball court 85m in length, scattered apparent residence buildings arranged in clusters of three to five structures, terraces, cemeteries, and s atellite sites with their own g uachimontones. All of these features are within 2 km. o f the main center at Circle 1. The guachimontn used for this thesis, Circle 5 was built just after the estimated time of occupation for Circle 1 (ca. 50 B.C. to A.D.50 for Circle 1, and ca A.D. 50 200 for Circle 5) (Beekman 2007 2008:424; Beekman and Weigand 2008:308,310), (figure 3. 4 ). Circle 5 can also be seen as an early contributor to the Early Classic trend of downscaled guachimontn architecture, whereas the scale of Circle 1 is t ypical of its time within the Late Formative. Circle 5 contains eight perimeter platforms, and shows uniform size and shape except for one smaller building. The buildings are also positioned somewhat irregularly around the circle forming a lack of exact s ymmetry, with distances between buildings that vary from about 1 to 4.5 m, and each building shows different construction design and materials (Beekman 2008:423 425). The Circle 5 patio also forms a somewhat irregular circle (Beekman 2008:424). The total d iameter of the guachimontn circle measured to the back walls of each opposing building is 36.0 37.9 m (Beekman 2008:424), and the diameter of the altar is 7.35 7.55 m. Using median diameter values as an approximation, the total activity area (the tota l area excluding the area of the altar) is about 1031 m 2

PAGE 96

81 Figure 3. 4: Navajas central group with a reas of excavation circled. The guachimontn used in this thesis, Circle 5, is the northern most circle. Both Johns (2014) and Tyndall and Beekman (2007) conducted distributional studies of ceramic artifacts from Circle 5 to determine the nature of activities related to ceramics. Both studies found that relative proportions of the different types of vessels were essentially similar throughout the cir cle, suggesting a communal, corporate structure based form of ritual. However, both studies also note that overall counts of sherds varied considerably between buildings, suggesting unequal participation between groups. Ceramics recovered from Circle 5 inc lude a total sherd count of 9,475 (Johns 2014:83), Circle 5

PAGE 97

82 which "far surpassed original expectations in terms of quantity and quality" (Tyndall and Beekman 2007:154), along with several complete and partial vessels and bowls. Based on the previous evidence for fe asting depicted in ceremonial ceramic forms throughout the Teuchitln culture (Butterwick 1998) as well as an abundance of storing and serving vessel forms at the circle (Johns 2014: figures V 12, V 13; Tyndall and Beekman 2007:174), the ceramic containers are interpreted as implements for corporate feasting. Although feasting was apparently still practiced during the Circle 5 occupation, at least two groups may have begun to dominate the rest in the corporate power field, utilizing the guachimontn ritual setting as an arena for political maneuvering T he outcome may have ultimately led to increased power of aggrandizers within the dominant groups, and a related expansion of economic control to include areas beyond staple goods. Circle 1 also possesse s 8 perimeter platforms, and Circle 1 buildings also vary in size, but in a consistent alternating pattern between large and small buildings. Within each respective size group, the large and small buildings also approximately equal each other in size, and the buildings are also much more regularly spaced around the platform T he perimeter buildings are constructed using similar materials and methods, with only a noticeable difference in the shapes of each building. Again following the theory that the s patial relationships between buildings reflect the social relationships between lineages that built and operated within them, Beekman suggests that Navajas Circle 1 may represent a time of well defined group relationships with major and minor groups unifor mly and distinctly represented, whereas Circle 5 represents shifting relationships between groups J ust as varia tion in the Circle 5 architecture is limited to the basic guachimontn pattern, however, changing lineage

PAGE 98

83 relationships are also seen as constra ined within certain limits by the corporate code (Beekman 2008:426). The more uniform structures at Navajas are also very similar to those within the nearby internal core zone, and Navajas therefore appears to be both culturally and socially associated wit h the core. Within the Navajas guachimontn, two plant species suspected to represent contemporaneous sources of subsistence were located in soil samples from activity floor surfaces, maize and huauzontle. The latter is a plant often cultivated for t he use of its seeds and leaves in Mesoamerican cooking and medicine (Benz 2007:248). The huauzontle seeds are of some interest because they are confirmed to be of a domesticated variety, and the only other Classic Period incidence of domesticated huauzontl e known to Benz is in Teotihuacn, the closest major Mesoamerican polity to the Tequila Valleys. It is not clear whether the huauzontle represents any sort of link with Teotihuacn, but it could be related to trade or some other social tie with the polity. Four obsidian tools registered positive tests for the presence of proteins which may have been food related. Three scrapers were positive for small animal proteins (species related to rabbits and guinea pigs), and one projectile point registered a grass p rotein which may have been maize, although no particular species were resolved from the tests (Parr 2007:225). Llano Grande Site Environment Llano Grande's location is currently in a narrow tract of semi open pastureland sparsely populated by wild Holm oak groves, surrounded by more dense Holm oak forest

PAGE 99

84 land. Holm oak is an especially hardy tree variety which consists of several species of the genus quercus which tolerate sandy soils and semi arid environments (Botanical s Online 2014). The pass op ens up approximately 1 km to the east into a narrow valley. The east end of the valley connects with the Magdalena Basin. The west end of the valley, closest to the site, is sloped pastureland highly subject to erosion which continues for about 200m as one travels east from the pass. Beyond this area in the valley below the pass is primarily part time or seasonal farmland. The site is thus less than 1 1/4 km from arable land, although the pass drops down a hill into the valley. The closest water source in t he valley may be 3/4 km from the west end, where a spring extends further east into the remnant Lake Magdalena. Several other springs provide water to the valley, mainly near the east end of the pass. A recent area geological study has determined that the northeast end of the valley would have been a marsh sourced by the current spring (Beekman personal communication August 2013). Thus, during the rainy season, Llano Grande would have had a 1.2 km distance to cropland. If the land was farmed during the dry season, water from the spring marsh would need to be utilized about 2 km from the site Land differences between the valley and the pass at Llano Grande are quite pronounced, and even assuming an adequate water source would have been available, there would have been no chance of growing any type of crop in the immediate vicinity of the pass. Although the valley contains entirely alluvial soil, the canyon is composed entirely of acidic (high silicon) extrusive igneous rock. The canyon's soil type is co nsidered Regosol a sparsely distributed, residual soil which does not completely cover bedrock, and therefore has no appreciable depth. The valley soil type is a humus rich, highly arable pheozim soil, in all areas except the southwest section which conta ins a

PAGE 100

85 high clay soil type considered quite difficult to farm. Throughout the valley, the soil is more than adequately deep for agricultural purposes although the soil in the sloping and eroding west end is much more shallow than the lower areas of the val ley, at around 10 50 cm The slope and degree of apparent erosion may have preclude d farming in that section of the valley. The rest of the valley gradually slopes from 1,370 to 1,420 m from the Magdalena basin to the western end of the alluvial soil area. The Llano Grande guachimontn sits at 1,487 m in the center of the pass, which is essentially flat and the areas towards the margins only vary in elevation from 1,480 to 1,500m. B otanical remains were located w ithin the Llano Grande guachimontn, wh ich indicate agriculture. These include charred cobs and pollen from maize, starchy seeds which could have been corn or beans, along with bottle gourd, cotton, and possibly tomato plant pollen (Schoenwetter 2004:4). Phytoliths analyzed from food residues o f ceramics (four ollas and one bowl) primarily indicate maize, and also fruit from the genera Prunus and Celtis (Schoenwetter 2004:4,5). Schoenwetter states that these results indicate agricultural production at Llano Grande, even though the site "was not embedded in an agricultural landscape" (Schoenwetter 2004:1). What Llano Grande lacked in agricultural potential, however, may have been well compensated for by the site's mineral rich environment The site overlapped a large obsidian quarry ( Beekman 2001:3; Spence et al. 2002 ) with a wide variety of color patterns (Spence, et al. 2002) O bsidian is ubiquitous throughout the site (Beekman 2001). The most abundant variety at the site is a dark green obsidian with very few inclusions and excellent flaki ng characteristics for the manufacture of stone products Llano Grande's mineral resources also include red ochre and opals (Weigand 1985: 90),

PAGE 101

86 although there is no definitive evidence that the latter two minerals were exploited by the Teuchitln culture. Like Navajas, Llano Grande has a direct route to the core directly across the valley floor, but Llano Grande is on the other side of the Lake Magdalena basin and is twice as far from the core than Navajas at 20 km (figure 1.2) T o a large extent, the lake and marsh levels would have further buffered the already geographically secluded Llano Grande against direct interaction with the core, and therefore also against centralized control A lthough average lake levels during the Late Formative and through Epiclassic periods were steadily receding, The shoreline of Lake Magdalena reached the eastern mouth of the valley, which would have required either a longer route on foot or travel via watercraft to access the core (cf. Anderson et al. 2013:25). Thus far no evidence of watercraft has been located within the valleys. If canoes were used the situation would have been alleviated to a large extent. However, ethnohistoric data regarding contact period canoe travel suggest that prehistoric transportation over water was 33 percent slower than land travel, and therefore increased the required travel time by an additional half of the time to travel the equivalent distance by land (Alden 1979:175) A lso, the presence of a spring fed marsh at the northeast end of th e valley would have hampered travel between Llano Grande and all sites north of the core (Beekman personal communication 2013). Llano Grande was not completely cut off from the core, but a much longer route including travel through the hills would have bee n required. The increased travel difficulty would also increase the resources required to maintain regular interaction between the core and Llano Grande. The core may therefore have been unable

PAGE 102

87 or unwilling to fully assimilate Llano Grande into the Teuchit ln social system (cf. Kowalewski et al. 1983:37). Given Llano Grande's location in a mountain pass, one may expect a higher elevation than Navajas' location in a lake basin outside the semi peripheral hill area, but the plateau on which Navajas sits is actually 51 m higher in elevation than Llano Grande. Los Guachimontones, however, sits 113 m below Llano Grande. Elevation is therefore considered negligible as an environment factor that could affect natural resource limitations between the two sites. Both areas also have access to good soils. However, Navajas has more expansive land resources, but with somewhat lower quality soil about 2.5 km away from the site. The topography and geology surrounding each site are categorically different, as is each si te's specific location relative to staple resources. Navajas Circle 5 is adjacent to its highest quality farmland, whereas Llano Grande's guachimontn is approximately 2 km from any viable cropland. Los Guachimontones' and Navajas' position s in open areas and adjacent to fertile soil m a y h a v e b e e n r e l a t e d t o s i m i l a r s t a p l e o r i e n t e d e c o n o m i e s w h e r e a s L l a n o G r a n d e s c o n t r a s t i n g d i s t a n c e f r o m a g r i c u l t u r a l r e s o u r c e s s u g g e s t s t h a t t h e s i t e m a y h a v e d i v e r g e d e c o n o m i c a l l y f r o m t h e t w o c o r e s i t e s However, both L l a n o G r a n d e a n d N a v a j a s show evidence of some degree of stapl e economy within their guachimontones, in the form of macrobotanical remains and residue analyses which reveal staple processing or storage. Description Llano Grande is situated in the center of a pass which opens into a valley off the Magdalena lake basin, northwest of the core (figure 1.2). Initial architecture surveys

PAGE 103

88 which included Llano Grande recognized 43 structures and 3 walls to accompany the sole 8 building guachimontn (Beekman 2001:4 6) and the site only received passing mention in a couple of peer review articles, since it was only one of many included in early surface survey studies (Weigand 1985, 2000) C loser observations during the 2000 excavation brought the total number of recogn ized structures to 75, with 6 apparent walls totaling approximately 1000 meters in length. One is actually a double wall, composed of parallel walls approximately 3 meters apart, and is bounded at either end by steep arroyos (Figure 3. 5 ). The wall system, and particularly the double wall, appear quite defensive in nature, but the walls are positioned to direct people travelling into the area through the site, and thus the walls appear to serve a dual purpose of defense and access control F urther, although the labor investment has n o t yet been quantified, it appears quite substantial compared to the moderate size of the site, and may well have been commissioned and coordinated by a central authority (Beekman 2001:4 6). Figure 3. 5 : Site layout of Llano Grande. The largest wall in the foregroun d (Wall 3) is the double wall mentioned in the text

PAGE 104

89 The site is much smaller and less complex than Navajas, and the sole Llano Grande guachimontn, Feature Group 14 is dwarfed by Navajas' largest guachimontn However, the Llano Grande circle is approximately the same size as the Navajas guachimontn used in the current analysis, Circle 5 Field work on the Feature G roup 14 guachimontn, which also consists of 8 perimeter buildings around the central patio, was done in the year 2000 (Beekman 2008:419). Like Navajas Circle 5, the Llano Grande circle architecture is irregular, with different building materials and methods for each building T he outer diameter of the circle ranges from 36.9 to 41.1m. Using 40m as a rounded average figure, the area calculates to approximately 1257 m 2 and the patio's area is approximately 527m 2 A n old consolidated ash flow creates a slight rise in the center of the patio. A shallow hole was cut into the rise exactly in the c enter of the circle approx imately 25 cm in diameter and 20 cm deep. The ash rise app ears to have replaced the center altar (Beekman 2003a:302, 2008:421 423). Based on these differences and irregular sizes and shapes of the Llano Grande guachimontn buildi ngs (figure 3. 6 ), the Llano Grande circle has the least uniform overall geometry when compared to both of the Navajas circles (Beekman 2008:424). A patio group of several structures is established unusually close to the guachimontn and directly behind per imeter building 14 6 (figure 3. 6 ). The patio group may represent residences of more powerful elites associated with the lineage. An additional 50 structures grouped into 12 compounds, over an area estimated at about 20 ha, are interpreted as residences (Be ekman 2003a:301)

PAGE 105

90 Figure 3. 6 : Llano Grande central group, with 5 meter increment contours. Ceramics in the guachimontn are quite low in number, with notable artifacts located only in one building, 14 5, which included two jars and five or six bowls. Together, the collection appears to function for serving liquid within the g uachimontones (Beekman 2001:7,11). The sherd count is also sparse, at 1,525 (Tyndall and Beekman 2007:155). However, these quantities must take into account that only a sma ll proportion of the circle has been excavated so far, including only three of the eight buildings directly on the circle perimeter, and a trench across the diameter of the patio (figure 3. 7 ). Beekman has noted that the distribution of ceramics appears une qual, in that most sherds were found within the buildings, rather than on the patio (Tyndall and Beekman 2007:170).

PAGE 106

91 The lack of patio artifacts may simply point to sweeping of the public space, or as Beekman suggests, a non communal form of feasting where each lineage group holds feasting rituals independent of the others. Figure 3. 7 : Plan of the Llano Grande guachimontn (with excavation units). From Beekman 2008, figure 4. No shaft tombs have been published in relation to Llano Grande, but a cem etery has been located in the valley to the east (Beekman 2001:6 7), and shaft tombs were reportedly just discovered in that area during a 2013 survey, although they were thoroughly looted (Tony DeLuca, personal communication April 2014).

PAGE 107

92 Llano Grand e has been viewed as "a modest, specialized boundary fortification tied to an overarching political authority (Beekman 2008:20, emphasis added ), and indeed, the effort that was likely put forth for the substantial wall construction may represent resources beyond those available at Llano Grande, suggesting a cooperative effort involving core elite interests (Beekman 2001:5). However, the construction investment in the Llano Grande guachimontn appears relatively low, and its non standard, irregular style su ggest a local effort with little to no involvement from the core elite. Either the core elite were unsuccessful in fully conforming and controlling Llano Grande, or they never intended to, and kept a distant relationship with the remote site with an unders tanding of allegiance (however loosely defined) with the Teuchitln sociopolitical structure. Sayula Basin The Sayula Basin lies well outside the Tequila Valleys to the Southeast, but still within the same chain of lake basins at the junction of th e transcontinental volcanic range and the Sierra Madre Occidental. The basin settlement area contains no guachimontones or iconographic similarities to the Teuchitln culture (Valdez 1998:230 231) and is therefore considered unaffiliated with the Teuchitl n. The area is given a special mention here as a neighboring area within the same geological and environmental region that shared other West Mexican cultural aspects, such as the shaft tomb tradition, and likely had a trade relationship with Teuchitln. Th e Sayula Basin and surrounding area appears to have been a hub for trade and interaction between several groups in Nayarit, the Jalisco

PAGE 108

93 highlands, Colima and Michoacn (Valdez 1998:224) T he Sayula Basin has no obsidian of its own, but imported large amoun ts of it: 582 lithic items from excavated contexts were dated to the Middle to Late Formative periods, the majority of which were obsidian (Reveles 2006:387). A marked increase in foreign items at the Sayula sites of Cerro del Agua Escondida and Case ta, including "all types related to jewelry and fine objects of obsidian and calcite (necklaces, anthropomorphic earrings, nose rings, tubular beads, circular finger scrapers)" for the Late Usmajac phase (200 B.C. to 200 A.D.) corresponds to the appearance of shaft tombs in the later sites and reflects an increase in the size and complexity of the exchange network (Liot, et al 2006:414 416) T he Sayula Basin experienced a parallel decline of shaft tombs with those of the Teuchitln culture during Sayula's Verdia Phase, ca. 180 A.D. 320 A.D. (Valdez 1998:224). The decline of the shaft tombs in both Teuchitln and Sayula basin contexts may have resulted from a decline in the related wealth economy, and its associated networking/exclusion strategy for the acquisition of wealth items. However, the decline may not have occurred equally through all semi peripheral and peripheral areas. Also, the shaft tombs' roles in wealth trade, as determined thus far, is one of a bonding of members to the lineage networ k, reaffirmation of its legitimacy and a final ritual display and cache of wealth. None of these functions were necessarily exclusive to the tombs. Rituals and symbols which served to reinforce the lineage may have transferred to other ritual spaces, or ot her, non lineage related networks may have formed for wealth acquisition. The increased presence of foreign goods in other contexts was likely at least partially due to the fact that wealth items which were previously destined for elite graves were still

PAGE 109

94 w idely circulated, but deposited elsewhere. Ultimately, the decline of shaft tombs is connected to changes in the trade economy, but not necessarily the trade economy's decline. Conclusion The variability evident in the organization of the Teuchitln culture holds a potential to teach us a great deal about the dynamics of changing economies and power financing strategies in very different ecological situations within a variable natural landscape. Excavation and detailed site studies are still few in n umber, but exhaustive survey of the Tequila Valleys continues S urveys and excavations published thus far suggest that material culture differences between core and semi peripheral sites demonstrate deviation from the core corporate ideal which increases w ith distance from the core, and therefore distant sites show a weakened association with the core. Large walls at sites within passes between the Tequila Valleys and the outside world suggest an attempt at creating a territorial identity which encompasses the semi periphery sites, but semi periphery sites still varied substantially from the uniform, corporate symbolism demonstrated by the core this indicates that core affiliation likely remained at arm's length. The walls at Llano Grande and other sites wi thin passes may only demonstrate the risk involved in attempted polity expansion with inadequate resources to maintain the necessary relationships for full assimilation of a remote site into the culture (Kowalewski, et al. 1983) O hnersorgen and Varien's ( 1996) measure of social interaction via the gravity model also supports the

PAGE 110

95 notion of a very week relationship between the core and at least some of the more distant and geographically isolated sites such as Llano Grande. At the site level, this chap ter has described the emphasis on wealth economy primarily associated with the shaft tombs, concurrent with an emphasis on staple economy mainly associated with the g uachimontones. The data further demonstrates how the Teuchitln wealth and staple economie s, which are viewed elsewhere as conflicting, articulate within the same socioeconomic system to support the social structure. Points of articulation between financial bases are reflected in the utilization of ritual artifacts in some corporate guachimont n rituals, and the reflection of staple finance in the presence of many shaft tomb food offerings along with prestige gifts. However, this relationship apparently did not exist without tension, since the corporate institutions still served to restrain and limit the power of individual lineages. Since more distant sites were also more culturally distant from the corporate ideal, the corporate system appears to have been manipulated by the stronger lineages, reflected in asymmetrical emphasis on specific buil dings placed around the guachimontn circles. The weakening of the corporate structure in at least some parts of the semi peripheral zone would have opened the door for elites to expand their power laterally by capitalizing on additional abundant res ources in each area, a pattern outlined in Hirth's matrix control strategy (1996). Accumulating controlled resources would have served to lower the risk of failure for any one resource by creating an option to lean on the successes of others (much like sto ck portfolio diversification in modern times), which may have been especially important as in periods of drought or other unforeseen environmental change that may cause general crop failure. Elites would also have viewed each resource as an

PAGE 111

96 opportunity for increased political power, as additional elements of the society come under their control. A staple oriented economy would have remained where agriculture was viable, while those in power likely sought to extend their reach by acquiring control of additio nal economies and their associated networks A t Llano Grande, the primary "alternate" economy would have been obsidian. Llano Grande was certainly close enough to some of the richest land in the valleys to capitalize on staple resources, yet the ceremonial center and a substantial number of buildings were instead placed beside and partially on top of the nearby obsidian mine situated at the top of the pass. Conversely, Navajas Circle 5 and the Los Guachimontones ritual center were both built adjacent to far mland, where access to obsidian sources appeared to be a secondary consideration A large obsidian mine existed near Navajas, about 2 3 km away from the ceremonial center (Hoedl 2013:6,110,figure 4.14), and the Los Guachimontones center was also only 3 km from its nearest source, La Mora Teuchitln (Spence et al. 1980:358 359 2002 67 68 ; Weigand 1985:88 89) Navajas' remote, yet exposed high plain environment is also in a good position for exchange and the group almost certainly did engage in such activity Unlike Llano Grande, however, travelers are not restricted to passing through Navajas on their way in and out of the Teuchitln semi periphery L lano Grande's chosen position where an obsidian source and restricted access happen to coincide is ideally si tuated for regular exchange with outside groups on a moderate scale, as a workshop based exchange center (Hirth 1998:453; figure 2.1). For these reasons, it is hypothesized that Llano Grande embraced both staple finance and wealth finance, but emphasized o bsidian over agriculture more than Navajas. Navajas is expected to have followed the tradition of

PAGE 112

97 nearby core more closely, and continued the core's emphasis on agriculture, while also trading with nearby areas for luxury goods

PAGE 113

98 CHAPTER IV LITHICS BACKG ROUND This chapter discusses lithic forms and relevant lithic studies within West Mexico, and primarily within the Tequila Valleys. Lithics in other selected parts of Mesoamerica are also discussed and are utilized along with the West Mexico data to form the basis for expectations regarding wealth and staple industry related products further described in Chapter V As Jay Johnson asserted in his study of the Mayan obsidian economy, "a clear understanding of what actually is being made is necessary be fore hypotheses about state level economics can be tested" (1996:171). In order to determine whether, and to what extent, lithic assemblages within semi peripheral sites were likely a result of wealth trade with external groups rather than agricultural, fo od preparation and/or other common functional activities, it is necessary to determine which forms tend to be utilized for each of these purposes S ome quantities of prestige forms produced locally were also likely used for ritual purposes, so context beco mes especially critical to interpretation. Distribution studies can then describe not only what was made, but what portion of the process may have been included in source, production and use contexts. This chapter is divided into two main sections: De scriptions of l ithic forms related to both elite prestige and general purpose utilization; and distributions of various forms related to local, regional and long distance trade. T he primary site under test is in a semi peripheral zone which may have mainta ined economic relations with core populations, external groups or both Therefore, these two sections are further divided into sub sections on internal Teuchitln data, and external group data. Some external groups might

PAGE 114

99 not have traded directly with the T euchitln groups, but probably heavily influenced other groups with which the Teuchitln people did interact. Overview of Mesoamerican Forms Chipped stone items throughout Mesoamerica were made from both chert and obsidian. Chert is a cryptocrystall ine quartz formed as a sedimentary deposit in lakes and streams. It is very durable, forms sharp edges and often very workable for most knapped stonework, although heat and pressure from volcanic activity may reform it into a more macrocrystalline structur e to a point where it may become unusable for tool manufacture (Phillips 2011:46) O bsidian, however, is a direct product of volcanic activity. It is a natural glass produced when rhyolitic or s i licic magma is extruded to the surface. Material just under t he surface of the magma cools too quickly to crystallize, and forms a glass layer. The magma surface forms into a layer of porous glass from the release of gases into the atmosphere, which results in a very rough and visually distinctive cortex layer. For the most part, obsidian remains just below the surface except for a few places where it breaks through and forms an obsidian spine (figure 4. 1 ). Obsidian spines then become access points for mining operations. As a natural form of glass, obsidian contains no crystal structure, which makes it extremely workable for knapping (Phillips 2011:115 116). Knapped obsidian edges are also unparalleled in their degree of sharpness (Clark 1988:11), with an edge width of only 3x10 6 mm (Phillips 2011:116), which is shar per than surgical steel, and obsidian blades have been tested as scalpels on both animal and

PAGE 115

100 human tissue with very positive results (Buck 1982). These properties have made obsidian a most desired material for knapped stone objects (Phillips 2011:116). Each group utilized both chert and obsidian somewhat differently, at least partially due to the population's degree of access to these materials Prismatic blades were generally only made from obsidian since it is the only material with fracturing proper ties that work sufficiently well for the required manufacturing technique Teotihuacn used obsidian for both prismatic blades and more common items such as bifaces (Spence 1981:774,776) whereas the lowland Maya, who had much less access to Figure 4. 1 : Profile of a rhyolite dome from a lava flow, showing the obsidian layer and spines. (From Hugh e s and Smith 1993, as found in Phillips 2011:116). obsidian and for whom the material was a rare commodity, typically reserved their obsidian exclusively for p rismatic blades (Johnson 1996:168). Mesoamerican lithic studies tend to concentrate on formal objects, and often on prismatic blades in particular. Yet much of the Mesoamerican flake industry is actually very expedient and typically contains many mul tidirectional (often termed casual ) cores and only a one to two step tool production technique D ue to this level of expedience,

PAGE 116

101 informal tools may be indistinguishable from what may be categorized as debitage, which has prompted Clark (1988:15) to suggest that the very distinction between lithic products and waste may be considered meaningless; any struck edge, even as a byproduct, is a candidate for use. Flake tools in this region involved perhaps one or two stages of operation to obtain a quick edge and a roughly workable shape (Clark 1988:27) T he same is true for Mayan contexts, where such "ad hoc" flake tools are common throughout the region although generally treated as "background noise" in the study of formal tools, although they do often show signs of use. One of the few analyzed examples is the dataset from the Mayan site of Cerros, where expedient flakes were shown to make up 65 percent of the tools bearing use wear (Johnson 1996:161,162). Spence et al. (2002) conducted a surface survey throughout the Tequila Valleys to determine the distribution of lithic materia l including production debitage and product forms. The team also noted that expedient flake retouch was common in the area, and combined the informal t ools with their waste category (2002 :63). Others (e.g. Beekman 1996a; Esparza 2003; Soto 1 9 82, 1990 ) have separate categories for informal flakes and blades, where their status as debitage, secondary use or expedient tools is unspecified. Aside from some eccentric forms, Mesoamerican li thics include typical forms found throughout the world, including knapped items such as bifaces, points, unifaces expedient flakes and blades; and ground stone tools such as manos metate s, mortars and pestles. The Mesoamerican polyhedral core blade indu stry became common at some point between the Early and Middle Formative periods ( Darras 2012:418; Hirth and Andrews 2002:123), and eccentric production and trade became popular among elites

PAGE 117

102 during the Early Classic period (Moholy Nagy 1999:306; Moholy Nagy and Nelson 1990:77 78 ; Spence 1981:780, 1996 ). Polyhedral Core B lades John Clark describes polyhedral core blades more by the technique of manufacture than by morphology, even though morphological characteristics must be used to identify the use of the technique in the field. They are created by orienting a pre shaped, roughly cylindrical core vertically along the major axis so that one of the two flat ends is at the top. Blades are then removed from the side of the core by repeatedly striking the top near the edge and working in succession around the perimeter of the platform W orking around the core multiple times produces successively thinner, more narrow and more uniform shapes with little waste of the original material and a high number of pro ducts for a given volume of source material T he initial macroblades and the next few successive stages are easily accomplished with hard hammer percussion (figure 4. 2 ), but the smallest and finest blades that can be produced with the technique require the precision of a pressure flaking technique I t is usually assumed that the final desired products are very thin and narrow prismatic blades (Figure 4. 3 ). Prismatics have very straight and parallel lateral edges, and one or two very straight and vertical do rsal scars which also parallel the edges (Clark 1988, 1997; Clark and Bryant 1997; Hirth and Andrews 2002:2 4). However, previous stages including macroblades have also been found with utilized edges, and as the basis for more formal tools ( Dockall and Sha fer 1993; McAnany1989:335,342; Parry 1987:37). In areas of West Mexico, l arger blades

PAGE 118

103 Figure 4. 2 : Polyhedral blade production sequence defined by John Clark, illustrated by Bradford Andrews. From Hirth and Andrews (2002:3).

PAGE 119

104 Figure 4. 3 : Prismatic bla de production sequence defined by John Clark, illustrated by Bradford Andrews F rom Hirth and Andrews (2002:4).

PAGE 120

105 are produced with no evidence of continuing the process as far as prismatic blade production (Darras 2012:420 421). Even though the process creates very little unusable material following core preparation polyhedral blade production does create distinctive d ebitage and unused byproducts that can confirm the use of the technique (Clark 1997 ; Clark and Bryant 1997) These forms are further deta iled in the Chapter V Eccentrics Eccentrics are lithic artifacts which include features that have no known practical function other than ritual or a mark of elite status (Whittaker 1994:177). Some eccentric shapes suggest some implied function, su ch as knives with intricately carved ornate handles found in ritual contexts (cf. Moholy Nagy 1999:figure 3a,h ; Whittaker 1994:figure 8.2), and some items defy any functional category ( Moholy Nagy and Nelson 1990:77 78; cf. Moholy Nagy 1999:figure 3b g,i k ; Whittaker 1994:figure 3.20;). Overview of West Mexican F orms M ost Mesoamerican chipped stone items were made from both chert and obsidian, but knapped Teuchitln items were almost exclusively obsidian. Brief informal surveys of Navajas and Llano G rande have not revealed any chert or other alternate raw material sources suitable for knapping Llano Grande is located on an obsidian source and Navajas ha s several abundant sources nearby. Some basalt appears to have been worked, but in a

PAGE 121

106 very expedient fashion and surveyors have discounted all but a few potential artifacts from the material (Christopher Beekman, personal communication, 2014) T he area's abundance of obsidian is due to the volcanic nature of its location, and the superior workability of obsidian has allowed the Teuchitln culture to manufacture any knapped items common to Mesoamerica, though prismatic blades are absent Macroblades are quite common, and some fine blades with irregular dorsal scars, called "first series blades" in Clark's sequence (figure 4. 3 ) also are found. Although large numbers of prismatic blades are found in West Mexican surface assemblages, there has been almost no sign of the prismatic form near activity floors in excavated West Mexican contexts. The prismatics on t he surface therefore appear to be Postclassic (Darras 2012:420 421). The Teuchitln culture produce d four forms of eccentrics, most of the which are have not been found in other Mesoamerican cultures These includ e pointed dowel shaped objects known as pikes ( cf. Beekman 1996a:figure 6.18) small rod ornaments thin and flat jewelry items of various shapes referred to here as laminar flake jewelry and small cruciform shapes Aside from some additional use contexts for pikes elsewhere in West Mexico ( R eveles 2005 ) t he only eccentric items the Teuchitln people are known to have had in common with other group s is the cruciform S ome have suggested that pikes are equivalent to Mayan bloodletters found in Belize (Beekman, personal communication 2014). How ever, no publications found to date suggest any form similar to pikes outside of West Mexico Additionally, the literature generally suggests that the Maya used prismatic blade s for bloodletters, which have no resemblance to pikes ( cf Clark and Bryant 199 7:131 132; Haines et al. 2008; Stemp 2014) The lack of production or use

PAGE 122

107 context in other areas for these items creates extra interpretational challenges to lithicists, regarding both manufacturing techniques and functional purposes. Pikes and Rod Ornam ents Pikes may be the least understood lithic items in West Mexico. Production techniques and use contexts for the pikes are both very much unknown, and there has been very little study regarding them. Pikes are long and roughly dowel shaped objects about a centimeter in diameter, with two bluntly pointed ends or a "knob" shaped protrusion that extends from one end (figure 4.4) There may also be a short protrusion that extends Figure 4.4: Pike collection, photo courtesy of Jose Guadalupe Romero, director, Casa de C ultura Museum, Tala Jalisco M X

PAGE 123

108 perpendicularly from the center of the pike, which may have been used for hafting (Beekman (1996:802) B eekman has observed that pikes are "more or less triangular in cross section" (1996:802), alth ough at least one segment with four sides has been located at Navajas. The sides are also somewhat convex on cross sectional edges, giving an overall impression of a roughly rounded shape. There also appear to be at least two varieties of objects usua lly labeled as pikes. The variety which is termed a rough pike in this thesis has large, deep and irregular flake scars of varying lengths. Some scars are quite large, occasionally exceeding 2 cm in length, and are mainly oriented with scars' major axes al igned vertically, along the major axis of the piece. (figure 4. 5 ). Rough pikes may simply reflect early shaping activity for the production of the more formal refined pikes (figure 4. 6 ), which exhibit much more consistent, uniform dimensions and small fin ishing flake scars. Closer observation of two pike fragments at Navajas shows that these scars are often oriented along the minor axis, around the circumference of the piece. Finishing scars are often aligned in parallel rows that run the length of the pi ke, and the pattern creates a more clear definition of the object's side edges than rough pikes. Although the flake pattern differences are clear in the artifacts recovered from Navajas and the one rough pike from Llano Grande, only a few items have been r ecovered from both sites so it is not known how consistent these patterns may have been, or how widespread they may be among pikes throughout the region. Two personally examined rough pikes from the sites under study closely resemble modified crested blade s from the polyhedral core blade industry. In polyhedral blade manufacture, either at the start of macroblade production, or if the core face has

PAGE 124

109 Figure 4. 5 : Rough pike from Navajas Circle 5, lateral and cross sectional views. Photos by Kathy Beekman. Figure 4. 6 : Refined pike fragment from Navajas Circle 5, lateral and cross sectional views. Photos by Kathy Beekman.

PAGE 125

110 undergone reshaping activity after a failure, the face is chipped along two parallel lines from the platform to the distal en d to create the dorsal side and lateral edges of a first blade. The initial blade is struck from the core before continuing on with the usual macroblade knapping procedure T his distinctive first blade is called "crested" because of its protruding dorsal e dge with wide and deeply set flake scars (Clark and Bryant 1997:125,figure 5; Soto 1990:240). The resulting rough but distinctive scar pattern of a crested blade very closely resembles that of a rough pike. Elsewhere, Reveles describes rough pikes in the Sayula Basin as having "a series of bi marginal adjustments ranging from abrupt to flake like" (2005:352), which closely describes the scar pattern of a crested blade. The transformation from blade to pike would then only require a rounding off of the lateral edges to form a nearly even width to depth ratio. Since crested blades are generally thicker than macroblades, they are likely the most ideal starting point for a rounded object. However, crested blades do not appear in very large numbers since onl y one would be produced to create a starting point for each core. A tentative production sequence for pikes may begin with roughly shaping a crested blade, if available, or macroblade with strikes along the edges of the blade, and then using perpendicular rows of finishing flakes to complete the product. However, the fact that rough pikes have appeared in the nearby Sayula Basin without any complimentary presence of refined varieties (Reveles 2005:352, Fig. 1c), suggests that the rough forms may also have b een utilized directly.

PAGE 126

111 Regardless of the exact method of production, unless the Belize "bloodletter" claim is verified, the process appears to have been unique within Mesoamerica. The distinctive finishing scar patterns on pikes reveal a fine, curved flaking from an unknown pressure technique. Even if pikes internally served some very practical purpose, their unusual shape would make them very collectible prestige items to external groups whether or not their intended function was understood or practic ed by outsiders. Rod ornaments have similar dimensions to pikes but only two convex sides, and no obvious flake scars. These items appear to be produced by polishing a piece of obsidian down to the form of a narrow rod shaped object (Beekman 1996a:802 Figure 6.18d). No rod ornaments were located at Navajas or Llano Grande, and they might be l imited to the La Venta corridor. Based on the contexts of two discovered rod items, they may date to the Postclassic period (Beekman personal communication 2014). Laminar Flake Jewelry Laminar flake jewelry is made from thin obsidian flake blanks, referred to in this thesis as l aminar flakes (termed "cylindrical core flakes" by Beekman [1996:797 799, figure 6.16] after one potential method of manufacture, an d also called "obsidian mirrors" by Long [1966:228 230]) They are extremely thin (less than 2.5 mm maximum thickness) and flat obsidian pieces with little to no detectable bulb of percussion and little to no rippling on either side (Figure 4. 7 ). They are very smooth and reflective, and are usually found retouched into various shapes with small drill holes in ideal locations for suspension as pendants. Shaped laminar flakes are therefore invariably interpreted as

PAGE 127

112 jewelry or adornments for ritual purposes (B eekman 1996a:790 ; Clark and Weigand 2009 ; Reveles 2005 ). Figure 4. 7 : Anthropomorphic laminar jewelry from Navajas. Three proposals for manufacturing methods of laminar flakes have been offered. Stanley Long's initial proposal is based on analysis of two round items interpreted as mirrors at San Sebastian, another site on the west side of the Magdalena Basin near the base of the surrounding hills (southeast of Llano Grande). Long suggested the use of a polyhedral blade core worked to a point where a near cylindrical shape is achieved. The core is laid on its side and thin, disc shaped sections of the platform are removed, in much the same way as one would slice a sausage. The technique has thus been dubbed the "sausage technique" by Schndube (Beekm an 1996a:795) T he core is rotated a

PAGE 128

113 quarter turn between each strike to create a more even surface with a different orientation of the bulb of percussion on each side, which is removed along with the rippling through polishing to form a smooth, lustrous s urface (Long 1966:228 230, Fig. 129) T he proposed method could produce very uniform, repeatable results, and has thus been suggested as a means of production for an extensive exchange network (Long 1966 :40). However, prepared fine blade cores, much less t hose worked down to a uniformly cylindrical shape, have thus far proved to be rare in Teuchitln contexts ( Esparza 2003 ; Soto 1982, 1990). Also, Clark and Weigand (2009: 79 ) have argued that Long's suggestion of polishing to achieve the finished surface is unfeasible. Since obsidian is essentially glass, its highest possible luster is achieved when it is fractured and the interior is initially exposed; the finest polishing material available in the area could only serve to dull the surface Long's basic tech nique remains a viable potential method with perhaps some modification; the bulb of percussion and rippling would need to be controlled through the knapping technique, rather than subsequent polishing. Clark and Weigand succeeded in replicating lamina r flakes by making successive strikes on a core, such that each flake centers the previous flake scar. (Usually, a knapper will strike over a dorsal ridge from the edge of a scar for easier flake removal.) Each removal then creates a larger flake as reduct ion progresses towards the center of the core. The process results in laminar flakes which are quite variable in size. Rippling and the prominence of the bulb of percussion are controlled through slowly delivered blows via a soft hammer, such as an antler or a wooden billet (Clark and Weigand 2009 : 5 Fig. 3) Yet a third potential method was proposed by Clark after close examination of one museum item of laminar jewelry, which revealed a plano convex shape (slightly rounded

PAGE 129

114 on one side, and essentiall y flat on the other). The shape appears to be the result of removing a bulb of percussion from a flake with a pressure technique, and then utilizing only the detached bulb for production (Clark and Weigand 2009:7,8). Some of the flakes from Llano Grande an d Navajas do appear to have had their bulbs of percussion removed, as noted in the field before any awareness of the proposed technique. Several flakes exhibit s ome very small scars and some crushing left behind in position where the bulb is expected. Late r observations by Camilo Mireles and Rodrigo Esparza have confirmed a convex convex shape on some flakes from within and near the core zone, which resembles two ventral surfaces, and is suspected to be the result of Clark's bulb removal method (Camilo Mire les, personal communication, 2014). However, although laminar forms are very much restricted to the region, they can be fairly numerous in West Mexican collections. A usefully reliably rate of production for these pieces based solely on this extremely frag ile method appears unlikely, especially considering that very large bulbs of percussion of up to at least 15 cm in width with no eraillure scars or noticeable rippling or fissuring would be required to match the sizes and features of some laminar artifacts Mireles also has confirmed that the bulb of percussion method could not have produced the largest laminar items. Clark had nearly dismissed his previous experiment based on the one plano convex sample. However, although it is unconfirmed by detailed comp arisons to field data, Clark's original method demonstrates that an alternative which has a potential for high rates of production can successfully produce laminar flakes. Technology often allows for multiple, diverse variations on basic production te chniques that can result in essentially the same final form (cf. Hiscock 2004), and Clark's experimental method certainly allows for a great deal of flexibility in technique as

PAGE 130

115 well. For example, we do not know the degree of core preparation before the fir st laminar flake is created. It could be a matter of simply lopping several of the more protruding angles off the core and then using an opportune flake scar as a starting point. Or, towards another extreme, the core could be worked down to an ideal size a nd shape with arrises aligned for easy successive laminar flake removals. Clark's seminal publications on prismatic blade reproduction (Clark 1982, 1997; Clark and Bryant 1997) deal with a technology designed for maximum material use efficiency (Sheets and Muto 1972), and Clark's proposed laminar flake production method provides the same advantage I n an area such as the Tequila Valleys where obsidian is readily available throughout the territory, material use efficiency is not a major concern, and thus les s efficient and usually more expedient production methods tend to be employed (Beekman 1996a:750, 817; Darras 2012; Hirth and Andrews 2002:9; Magne 1989:22). Potential technique variations can therefore allow for some less efficient means to create a lamin ar flake of the desired size I t is certainly possible and perhaps likely that techniques varied between sites, and to some degree even between artisans at the same site. However, allowing for multiple technique variations does not negate the speciali zed nature of the products themselves. The jewelry disappeared from Tabachines shaft tombs in the periphery at the end of the Late Formative period, but continued to be made in sites closer to the core including Navajas, Llano Grande, and Los Guachimonton es. This pattern paralleled other items interpreted as elite markers, such as hollow figurines, which suggests that elites had brought the technique under a more centralized control ( Beekman and Weigand 2008 : 309 ) C oncentrated p roduction debitage for likel y flat obsidian jewelry was found in only one isolated area in the La Venta corridor which is considered

PAGE 131

116 a probable workshop It also appears to be the result of one or more specialized techniques, and is therefore considered a prestige item (Beekman 1996a :805,815). Distribution In order to understand how lithic technology fits into the overall economy of a culture, it is necessary to understand as much as possible about the full spectrum of what several French lithicists have termed the chane opr atoire or sequence of operation. Although the chane opratoire does include the production sequence of a particular technology, as illustrated above in figures 4.2 and 4.3 for Clark's prismatic blade sequence, it encompasses human activities beyond the p roduction sequence to include the entire industrial cycle, beginning with procurement and continuing through production, use, storage, maintenance and discard. These actions can then provide information about the concepts technology strategies and roles o f the people who carried them out ( Bleed 2001:105; Bo da 1995:43 ; Julien and Julien 1994:15 ; Lemonnier 1986 ) Thus, the chane opratoire aims to describe and understand all cultural transformations that a specific raw material had to go through (Sellet 1993:106) The concept has been criticized as inflexible in its attempt to identify specific technology strategies with specific cultures (Shott 2003). However, t he Mesoamerican literature reviewed below suggests that exchange relationships cause some tran sformations to cross cultural boundaries O perational sequences ar e often expressed as workflows and can form branching paths often from items considered production byproducts to create additional

PAGE 132

117 products from the same material ( e.g. Bleed 2001 :figure 4 ; Sheets 1975 ) as described above for Mesoamerican macrobades and expedient flakes. Therefore gaining understanding of an economic strategy must go b eyond familiarity with lithic for m s. W here and how both prestige and utilitarian products and related materials typically travel in various stages of their lifecycles must be determined as well as the types of locations at which they are found. Such locations may include ceremonial centers, administrative centers, workshops, and neighborhoods within iden tifiable social strata. Since partially worked materials can be transported and further refined at either intermediate or destination locations, part of this process is to determine what stages of the production process were likely transported over local, interregional and interpolity scales of trade. Transported forms may include raw material, finished products, or some intermediate stage such as prepared cores or bifacial blanks. With this background, differences in physical characteristics of artifacts a t the examined sites (such as specific tool forms, debitage forms, and expedient tools ) resulting from production and distribution activity for various artifacts can then be used to determine relative quantities of lithic items which were likely the result of production for trade Forms and production stages of traded prestige items as well as those of staple related items for local and area wide use, can be used to interpret the economic nature of the lithic collections and give us insight into staple and wealth finance mechanisms of the regional political system I tems considered "elite" or "prestige" goods and related distribution strategies may differ between polities Therefore, a survey of product distribution pattern studies in surrounding areas will be utilized in this section to determine what categories of lithic

PAGE 133

118 products were consistently valued as prestige items across different groups, and how these products and related byproducts/debitage were distributed. Workshops and Points of Exchange Most studies of trade and other distribution patterns focus on workshops as indicators of nearby distribution points of exported items. Such locations are often identified and referenced with assumed definitions of exactly what is meant by the term "work shop" However, discussions regarding definition of a workshop, and relatedly, how one may be recognized in the field, show some diverse opinions and explanations. According to Michael Spence, a w orkshop can be identified as a site with unusual lithic depo sition, which may include an especially dense area of obsidian, a high debitage to product ratio, or the presence of unfinished products (1981:771) M oholy Nagy has used a more functional definition to differentiate workshops from other deposit contexts, s uch as middens and collections of offerings. She describes workshops as places "where the principal activity is the manufacture of stone artifacts" (1990:269). Clark's definition is more specific to trade contexts: "A designated space where artisans or spe cialists create products for sale or exchange," and Clark further specifies that production must exceed the consumption needs of the specialist (1981, as found in Soto 1990:216). Johnson (1996) has noted that Clark's requirement for evidence of overproduct ion ignores remote deposition, especially in relatively confined areas such as small buildings where large accumulations must be periodically removed. Enclosed spaces used as workshops will still contain an unusually large concentration of debitage, due to its usefulness as a cache

PAGE 134

119 of potential expedient tools, which creates some degree of reluctance to move debitage to its final contexts (Whittaker and Kaldahl 2001) W hen the span of time for workshop activity at a site is unknown, however, Clark's overpro duction stipulation becomes impossible to measure because the rate of production cannot be derived. This thesis will follow Clark's definition as the most useful to its purpose, but with special attention to the nature of objects produced rather than relyi ng on quantity to determine whether production may have been done for the purpose of exchange. Challenges to Interpreting Inter polity Exchange Differences in products and manufacturing techniques between polities may complicate any interpretation based on data from multiple groups, but different forms can be generalized into broader categories F or example the most obvious difference between West Mexico and other Mesoamerican regions is the apparent lack of any prismatic blade industry. P olyhedral cores worked down to a point where prismatic blades can be produced have been found in a Los Guachimontones area workshop ( Soto 1990) and the few published excavation contexts from the area show heavy emphasis on polyhedral core macroblades (Darras 2012:4 20 421) However, e xcavated contexts in the area have not revealed a substantial number of prismatic blades or exhausted prismatic cores (Darras 2012:417 418, 420 ). Yet elsewhere in Mesoamerica, prismatic blades were ubiquitous since at least 700 B.C. (Hir th and Flenniken 2002:123) D espite these differences, distribution patterns can still be validly generalized and compared along lines other than specific forms, such as production stages (finished formal products, raw

PAGE 135

120 material, and items from intermediate production stages such as "blank" forms and prepared cores), items deemed "elite" or "wealth" items, and common or pragmatic staple goods. Exchange in Mesoamerican contexts The following discussion provide s some background in lithic production and distribution patterns among Mesoamerican groups, and note common approaches to trade which can be considered normative for inter polity interaction. Some t rade norms of external groups w ere also likely followed by Teuchitln semi peripheral groups involved in economic interaction which may be reflected in the material culture. The topic of Mesoamerican lithic distribution patterns, even when limited to specific periods and to a few major groups, is far too broad to tackle comprehensibly for the purpose of this thesis, so just a few summary reports and case studies considered representative of general patterns for Middle Formative Central and South Mexico (De Len et al. 2009), Classic Period Maya (Johnson 1996), and Teotihuacn (Spence 1981) A consens us regarding lithic distribution during the Late Formative and Early Classic periods throughout Mesoamerica held that, in nearly all cases, prepared cores were exchanged and transported rather than finished products (Clark 1987 ; De Len and Carballo 2003 ; De Len et al. 2009 ; Jackson and Love 1991 ; Parry 1987). Core preparation removes most of the external cortex material and shapes the core for easier manufacture. The strategy would expose the severity of inclusions in the underlying stone and also reduce unnecessary mass for ease of transport (Spence 1981:776, 777).

PAGE 136

121 Consumer sites then had the responsibility to manufacture their own finished goods. The pattern is seen as a stark and abrupt change from the Middle Formative practice of trading mainly finishe d products. However, this consensus was based on studies which considered only the presence or absence of cores and products near source and consumer areas. Later, more detailed analysis determined that although the pattern appears true in a general sense and is useful as a framework for comparison of more detailed studies, specific sub regional contexts appear to have varied in the rate and timing of the transition to core trade and local production (De Len et al. 2009) Finished Product Distribution: E arly to Middle Formative Early to Middle Formative data predate the time frame for any known presence of Teuchitln culture. The early data are presented here to provide a background for trends leading up to the Late Formative to Early Classic distrib ution pattern, and to illustrate an alternative to the expected pattern De Len et al. (2009) examined changes to lithic distribution strategies throughout South and Central Mexico over time using lithic data sets previously collected in Formative T laxcala, the Basin of Mexico, and Oaxaca. The De Len and company approach used comparisons of formal attributes associated with both ceramic and lithic artifacts between periods at multiple sites to identify distribution differences, following Hirth (1998 ). De Len et al. interpret the presence of primary debitage ( cores and debitage distinctive to polyhedral core blade production such as facial correction flakes, platform correction flakes, crested blades and cortical flakes) within a consumer context as evidence of imported raw material and local production, whether or not cores have

PAGE 137

122 actually been found on site A presence of actual cores of course serve s to strengthen the argument but since they are few in number and may have been deposited in differ ent contexts, their presence in a manufacturing debitage context is not necessary (De Len et al. 2009) A n absence of both cores and core formation debitage along with a large quantity of flakes that appear to be from middle to later stages of the product ion process (such as various forms of failed blades or small biface flakes) suggest the use of itinerant craftsmen who travel with the core material and knap products to order at various consumer locations. However, one potential problem with the approach noted by De Len et al. is that an absence of core material may just as well result from a partial excavation of the site that missed the evidence because it was spatially displaced, possibly because of culture norms that dictate different handling of exha usted cores (De Len et al. 2009:114). The presence of larger quantities of cross sectionally broken blades in use contexts suggest that fine blades within Middle Formative Mesoamerica were often intentionally broken into multiple segments, and the st raight, uniform medial sections were favored for use. An unequal distribution of proximal, medial and distal blade segments at a site was also used by De Len et al. as a clue to determine whether blades were segmented, and therefore also knapped, at a dif ferent location. Conversely, a preponderance of medial sections suggests a consumer site where blades were externally knapped and segmented. A nearly even count of all segments suggests local segmentation of whole blades, although the determination of loca l or remote blade production then requires additional evidence (De Len et al. 2009:115 118). The medial blade distinction can therefore be used with the identification of primary core debitage to identify mixed contexts which

PAGE 138

123 reflect a combination of both final product imports and local production (i.e., both a predominance of medial segments and the presence of cores, rejuvenation flakes, and other identifiable blade industry debitage.) De Len examined blade and debitage data retrieved from the sit e of San Jose Mogote and the E arly to M iddle F ormative Loma de Atoto site in the Basin of Mexico, and concluded that most of the blades at both sites were imported. The sites of Amomoloc, Tetel and Las Mesitas of Formative Tlaxcala show a more complex scen ario. Amomoloc is the earliest site, and contained a high degree of prismatic medial segment import along with some secondary (non core) debitage distribution. Tetel, which spans most of the M iddle F ormative period and ends just before the start of the Lat e Formative (700 400 B.C.), shows an apparent gradual transition from imported pre segmented blades in the earlier contexts, to local household production in more recent layers. Las Mesitas, which was occupied only during the later part of the middle for mative (500 400 B.C.) shows evidence of both pre segmented blades and core debitage, indicating a mixed model of product and raw material import (De Len et al. 2009:124). For the most part, De Len et al.'s analysis confirmed the accepted view of c hanges in Formative Period trade and production patterns H owever, it did identify some details about a more gradual and earlier than expected transition from the import of blades to the import of cores and local production in Tlaxcalla T he later Las Mesi tas data is also from a time period too early to suggest that it negates the expected change in traded forms for the Late Formative period, but illustrates a need to consider the possibility of multiple distribution strategies for any specific region or ti me period

PAGE 139

124 Raw Material Trade: The Late Formative and Classic Maya De Len et al. also noted a lack of status differentiation based on elite and commoner access to obsidian during the Middle Formative, which only began to emerge in South and Central Mexico during the Late Formative period (2009:119 121). The Maya began to more clearly display elite/commoner differentiation via imported goods throughout the Mayan Lowlands during the Late Formative period as well, and the trend appears to have peaked d uring the Early Classic (Barrett 2004:43, 281) T he Lowland Classic Maya mainly imported obsidian as a relatively scarce, exotic material from distant lands (Johnson 1996), which included areas in the Guatemalan Highlands and Central Mexico (Moholy Nagy an d Nelson 1990 ; Moholy Nagy et al. 2013 ) A lthough obsidian blades do appear in earlier contexts, evidence of local manufacture does not appear until the start of the Late Classic period (Moholy Nagy et al. 2013:78). Debitage distribution analyses in the a rea show that obsidian was brought into major settlements mainly in the form of cores, and nearly all major Mayan centers followed a pattern of spatially and politically centralized obsidian import and production, but likely only partial control of the dis tribution for ceremonial purposes O bsidian cores, rejuvenation flakes and other core related flake types have been found mainly at city center workshops, including the public center of Nohmul in Northern Belize (Johnson 1996). Nohmul is viewed by Jo hnson as a good representative case study of Mayan obsidian distribution strategies, since the center's lithic distribution data have been thoroughly studied throughout the area, and they appear to exemplify the most common distribution strategies found th roughout the Maya region during the Classic Period (Johnson 1996:170). Obsidian tools used in common household contexts at Nohmul show a similar

PAGE 140

125 pattern to that of the Late Formative households, in that obsidian appears to have been utilized to a greater e xtent by elites than by commoners. Material utilization intensity studies using multiple, disparate methods including measures of density in counts per unit, ratios of obsidian to chert, and cutting edge to mass ratios, have all demonstrated that aside fro m the relative rarity of obsidian in all contexts, elite household contexts consistently show greater utilization of obsidian over that of commoner households (Johnson 1996:166). Johnson also utilized pottery sherd density as an indicator of the intensity of non lithic related activity in different locations, and compared sherd to flake ratios to determine where work with obsidian was emphasized over other activities J ohnson concluded that within the No hmul city center and periphery, the emphasis on obsidian working appears to have been about 30 percent greater in the city center area (excluding the apparent workshop area) than in the commoner households, relative to ceramic usage T he proposed workshop location shows a much higher concentration of fl akes relative to sherds than other city center areas, with an increase in relative flake density ranging from 85 98 percent which further confirms its use as a specialized lithic related workshop The study's use of ceramic quantities as a normalizi ng factor does not appear to differentiate between expected amounts of ceramics in ceremonial contexts relative to that of households (and indeed, Johnson does warn that sampling within Mayan city centers is problematic, due to their "large, complex and he terogeneous" nature [1996:168]), but Johnson's data show a consistent reduction in obsidian density as one moves farther away from the center B ased on the workshop's location quite close to public spaces (and actually directly under a later Classic Period ballcourt), obsidian

PAGE 141

126 appears to have been brought directly into the city center and worked by attached artisans, primarily for public ritual purposes. Surplus goods (primarily prismatic blades) may have been released to the surrounding elite residences fo r more common uses, and eventually dissipated outward to the peripheral commoner communities through a relatively uncontrolled form of trade (Johnson 1996:170 172). Once released to the general public, in both elite and commoner residential contexts, blade use appears to be quite utilitarian and diverse in form W ithin the private sector, therefore, obsidian was likely primarily used for a wide array of common household tasks (1996:166). More recently, residue analysis from the lowland Maya site of Lo s Naranjos and nearby communities in Honduras confirmed this notion. Prismatic blades have produced residues of soft botanicals such as grasses, palms, maize and other starches likely utilized for culinary purposes, as well as woody plants more likely expl oited for industrial uses, such as building construction and hafted tool components T he residue data reveals a more strikingly varied use of even the relatively small and fragile prismatic blades than previously surmised through contextual data (Morell Ha rt, et al. 2014) N on utilitarian uses are also considered quite likely M ayanists and other pre Columbian Mesoamerica scholars have noted that especially small, narrow blades within both ritual center and elite household contexts appear ideally suited for blood rituals, and commonly refer to these items as "bloodletters" or "lancets" (e.g. Carballo 2009:494 ; Johnson 1996:172 ; Joyce 2004:195). Another apparent consequence of the short supply of obsidian at Nohmul is a more extreme conservation of core material, and a desire to maximize material use efficiency by utilizing nearly every possible cutting edge D orsal flake scar patterns show reuse of

PAGE 142

127 shattered polyhedral cores as flake cores, with use wear patterns on flakes as short as 10 15 mm. There als o is evidence of attempts to create new platforms at the distal ends of prismatic cores so they can be worked from the other side to extend the life of the core. Yet, three relatively under utilized polyhedral cores and one macrocore were discovered in a Late Classic Nohmul tomb, along with several high status items which include 14 whole obsidian blades, jade, and chert eccentrics (Hammond et al. 1987:265, fig's 8 and 9). The inclusion of largely unused obsidian cores in a Mayan tomb would have been considered a major sacrifice of a highly sought resource considering the apparent level of demand for useable obsidian edges. Fine imported obsidian has also been located in Mayan tombs at several other sites, including Kaminaljuy, Tikal, Altun Ha and Cop an in several forms, including blades, bifaces, points and eccentrics (Becker 1992 ; Hendon 1991 ; Sheehy 1991 ; Spence 1996 ) H endon (1991:910) has stated that blades were actually quite common in both elite and commoner graves at Copan, but Copan is conside red an atypical site regarding overall obsidian distribution in that it is near an obsidian source, and production appears decentralized in household contexts (Johnson 1996:170 ). A few exhausted cores also have been located in tombs at Tikal (Moholy Nagy e t al. 2013), but the use of cores as grave goods does not appear to be a common pattern among the Maya. Although the cores in the grave context do provide further evidence of emphasis on local manufacture from imported cores, perhaps more importantly they also show the importance of obsidian to elites even as a lightly worked raw material prior to product manufacture. Obsidian was highly sought by elites even though high quality chert which could be made into very refined objects was more readily ava ilable from sources within

PAGE 143

128 Northern Belize (Shafer and Hester 1986). Elite affinity to obsidian, therefore, does not appear to be entirely related to its superior workability and performance as a tool, but to a large extent, also to status and symbolic pow er. Despite the evidence for local production in general, the chert eccentrics within the Nohmul tomb may well have been produced elsewhere T he small Northern Belize site of Colha appears to have specialized in the production of eccentrics and highly specialized tool forms distributed within the Northern Belize area. The sheer volume of debitage at the site (over 100 piles averaging 30 meters in diameter and 1 m deep) suggests tool production rates well beyond local needs The site is estimated to hav e produced about 4.5 million pieces through the Late Formative period, and given site population estimates, this translates to 150 tools per worker per year (Shafer and Hester 1986:162). Colha has been an area of contention with some archaeologists who hav e viewed the site as more of a common debitage dumping ground for multiple sites, rather than a major workshop area ( Mallory 1986 ; Moholy Nagy 1990). Moholy Nagy (1990) has pointed out that ethno archaeological evidence consistently demonstrates cores prep ared at the source, worked into final form at workshops removed completely from the source area very near artisan residences, then dumped at a remote location to free up space for other domestic purposes. However, consumer site debitage studies within the area surrounding Colha do suggest the presence of a nearby centralized production site from which finished tools were exported. Colha chert tools were found at Pulltro u ser Swamp approximately 35 km from Colha, and no production debitage other than small re sharpening/recycling flakes have been located at the site Colha chert" tools have been discovered in similar contexts at several other sites as well, within a radius of approximately 40 km from Colha.

PAGE 144

129 Distinctive eccentrics from Colha also appear at seve ral of the largest Mayan sites up to 100 km away (McAnany 1989) C olha appears to be one of several specialty workshop areas discovered within the Southern Lowlands at smaller sites near major polities (Shafer and Hester 1986:163). Apparently, one exceptio n to the Mayan pattern of local production was the import of uncommon items that require rare skill sets or lesser known techniques. Some of the data presented by Johnson appears to parallel De Len's proposed evidence for mixed model distribution (i mported products combined with local production) for core blades, in that the number of medial blade segments (N=1002) in an apparent workshop context far exceeds the number of proximal and distal segments (N=367 and 191, respectively) (Johnson 1996:Table 1). Additional pre segmented blades may have been brought in from elsewhere within or near the city center, especially if the site served a dual production purpose where blades were both manufactured and utilized for some other purpose at the workshop. The possibility of a dual production workshop in Mayan territory is not without precedent; such utilization has been determined at Kaminaljuy via apparent use wear on expedient "waste" flakes among other evidence for core blade manufacture (Anderson and Hirt h 2009), and some workshops in Teotihuacn previously thought to be producing lithic items for exchange were also found to contain a high percentage of use wear, and re interpreted as craft production areas (Spence 1987:430 432) Of course, any set o f generalizations that attempts to cover an expansive segmentary group such as the Maya may be especially subject to De Len's caveat about the likely presence of local variability in distribution strategies, and such exceptions may especially

PAGE 145

130 apply to eco nomies within areas that feature unusual ecological profiles. Both the highland sites of Copan and Kaminaljuyu show some variation from the pattern of centralized production and initial control of distribution. Production at these sites appears to have bee n done primarily in household contexts, rather than elite workshops within city centers. A difference in material access may be a primary cause for the difference in control in both cases, since both sites are also near obsidian sources (Johnson 1996:170 ). The Blue Creek area, within the alternating hills and fertile valleys just west of the alluvial plain of coastal Belize, provides an example on the other end of the resource availability spectrum. The Blue Creek territory contains very few sources of work able stone of any sort S patial correlations between concentrations of known elite goods (e.g. jade, cacao and spondylus shells) and concentrations of stone tools and debitage from local outcroppings show that elites likely controlled areas of common lithi c sources (primarily chert) as well as obsidian (Barrett 2004). This variability, however, offers an opportunity to observe at least one factor affecting the dynamic aspects of distribution, rather than making static predictions which assume uniform circumstances between groups In all of the above cases within Mayan territory, the degree of centralized, elite control of obsidian appears to inversely relate to the degree of availability of the material. Since rarity is one important aspect of a design ated prestige item, the control pattern matches the expectations of Earle's (1994) concept of a wealth financed economy, controlled through Blanton et al. 's (1996) concept of networking/exclusion power strategy

PAGE 146

131 Teotihuacn Teotihuacan was one of the primary Mesoamerican centers in central Mexico from the latter part of the Late Formative period through the Middle Classic ( ca 0 750 A.D.). The city is the closest major contemporaneous Mesoamerican center to the Tequila Valleys and even though no trade relationship has been found between the two areas (Beekman 1996 a :997 ; Weigand 1993:231), some aspects of Teotihuacan economy likely influenced patterns of production and trade in all surrounding Mesoamerican groups in some way, either directly or in directly, including the Teuchitln. Common obsidian tools created in and around Teotihuacn were distributed primarily locally and regionally, with a small amount of long distance trade that reached as far as Mayan territory (Spence 1981). Also, the city w as largely formed by emigrants from various surrounding communities, one of which was near the Tequila Valleys, within Michoacn C eramics and obsidian from grave goods found in the Michoacn barrio were sourced via INAA and XRF analysis, and found to be o f Michoacn origin. Yet, items in living floor contexts were local, and completely in the Teotihuacn style. If ties to Michoacn were maintained, some ideas related to Teotihuacn culture, including lithic production and distribution strategies, probably also made their way back to Michoacn (Begun 2013). Some pressure blades of a high quality, distinctive green obsidian from the Cerro de las Navajas source have also been found in the Sayula Basin in what appears to be a Late to Terminal Classic Period con text (Reveles 2005:368) The source (also commonly called the Pachuca source [Stemp et al. 2012]), is within the Central Mexican Highlands about 50 km from Teotihuacn and more than 500 km from the Tequila Valleys Phil Weigand has determined a single gree n blade within the Teuchitln core area to be from the

PAGE 147

132 Pachuca source as well which indicates a possibility of at least some rare interaction between Teotihuacn and Teuchitln (Spence et al. 2002:71). The Sayula and Teuchitln items, however, were recove red from surface surveys, and therefore the chronological data are unconfirmed (Darras 2012). Green obsidian at various Teotihuacn workshops has been identified as material from the Cerro de las Navajas source via neutron activation analysis. Obsidi an from this source (often referred to as simply "green obsidian" by Spence since the color was commonly associated with Cerro de las Navajas) was the primary material for core blade production as well as all known exported eccentrics G reen obsidian was b rought into the site in core form and utilized in workshops within or near the city center (Spence 1981). Cores were located in all workshops engaged in core blade production D uring the Tlamimilolpa phase (A.D. 200 450) 90 percent of the core and blade m aterial in workshops within the city have been found to originate from Cerro de las Navajas. Another source of lower quality grey obsidian from the Otumba region 16 km from the city was utilized for bifaces and scrapers. Otumba obsidian was also apparently imported in core form, evident from a very common presence of cortex at each workshop, despite the fact that most of the cortex was most likely removed from these cores by field preparation at the source location (Spence 1981:776, 777) Some well sh aped but unfinished bifaces have also appeared at some workshops, and are interpreted as probable blanks which were left unfinished to retain a shape more resistant to breakage during transport, and finished by the recipient upon arrival at their destinati ons. Blanks were rarely found locally outside of the workshop contexts, and according to Spence's analysis, local finished points appear to not have been

PAGE 148

133 manufactured from blanks, which indicates that the blanks were likely made strictly for export purpose s (Spence 1981:771). Bifacial blanks are emphasized in the first known workshop set up for regional distribution during the Tzacualli phase, ca. A.D. 1 150, and their production grew with the expansion of the regional distribution system, ca. A.D. 200 450 (Spence 1981:781). Spence suggests that Teotihuacn may have held direct state control over workshops closest to the city center, while those in more semi peripheral areas were encouraged and fostered by the state but essentially self directed. Spenc e's conclusion is based on communication and transportation issues related to the physical distance of the workshops from state resources and the workshops' strategic positions, which were more directly exposed to both local and remote consumers (1981:770) The distributions of product forms in regional and local consumer sites were compared to those within semi peripheral and centralized area workshops, and Spence concluded from the distribution data that internal city workshops produced mainly locally uti lized goods, and semi peripheral area workshops produced items for external trade (mainly, the aforementioned biface blanks) as well as local and regional distribution H owever, later use wear analysis has concluded that some of the workshops actually manu factured obsidian tools primarily for internal work on other crafts, such as woodworking or textiles, with little or no redistribution of the obsidian itself. But internal use of obsidian tools was only indicated at a minority of workshops; the amount of e stimated production for export remains substantial, involving at least 44 regional workshops (Spence 1987:430 432, tables 2,3), and Spence's reduced estimation of the magnitude of Teotihuacan's obsidian export

PAGE 149

134 industry did not change the overall view of th e role of Teotihuacan obsidian within Mesoamerica as the central region's primary exported commodity (cf. Algaze 1993:312). Teotihuacn also shows some evidence of a difference in control over each of the two sources. Spence suggests that the acquisit ion and dispersal of the green core blade obsidian to final destinations was controlled by the state, whereas the collection and transport of grey obsidian was an independent effort from each respective workshop T he green obsidian is not found in workshop s elsewhere in central Mesoamerica, which suggests Teotihuacn control of the source, and it shows up in core blade contexts in nearly the same source material proportions at each workshop, which indicates a central, coordinated effort for dispersal. The g rey obsidian source includes one color variant which shows up in different proportions at each site. The variant differences likely resulted from different workshops working independently to access different areas of the same source. One apparent state ass ociated workshop at the Ciudadela, a large public open space near the city center which produced large bifaces and could take most advantage of the larger block cores from the grey source, but does not have a proportionally higher supply of grey cores with in external biface production areas than what has been located in semi peripheral workshops (Spence 1981:777 779). Spence has suggested that Teotihuacan fully control led the green source during its Tlamimilolpa phase, during the Early Classic period (A.D. 200 450) (Spence 1981:782,783). As with the Maya, Teotihuacn elite graves include obsidian in forms of blades, points, tools and eccentrics (DeLucia 2008; Sempowski 1992; Sempowski and Spence 1994). Obsidian is found much more frequently in Teotihu acn interments than in the Mayan tombs, at an approximate proportion of 25 percent over all excavated graves. It is

PAGE 150

135 still found disproportionately among graves of relatively high status, identified by the abundance of other elite offerings (Sempowski 1992 :33,36). In Teotihuacn, obsidian was not an exotic item, and therefore not high on the list of internally wealth related goods unless elaborately worked (1992:34). Although its internal use is relatively common, it is also commonly used for wealth export, so its use as a grave good underscores some level of identity of Teotihuacn elites with a wealth industry. Inter polity Distribution Some Teotihuacn manufactured ritual ceramics and obsidian eccentrics have been found dispersed through Mayan terri tory. These elite trade items, and the nature of the inter polity relationship that their presence represents have been a topic of study spanning several decades (e.g. Kidder et al. 1945 ; Moholy Nagy 1999; Moholy Nagy and Nelson 1990:72 73; Pendergast 1971 ; Spence 1981:780, 1996; Stemp et al. 2012). Sourcing via x ray fluorescence and neutron activation analysis has confirmed that Central Mexican obsidian from six sources of varying quality were exported as far as the Mayan city of Tikal in Guatemala, at a distance of over 1,200 km, from the Middle Formative to the Early Postclassic periods (C.A. 800 B.C. 900 A.D.). Some of the material is from the aforementioned Cerro de las Navajas, which is found in ritual contexts at multiple Mayan polities (Moholy Nag y 1999:300 303, Table 1; Spence 1981) T hese items have occasionally been found in elite tombs at several sites, including La Lagunita, Yaxha, Kaminaljuy, Tikal, Rio Azul, and Altun Ha in the forms of blades, bifaces, points and eccentrics (Spence 1996), although a larger portion from the green source has been found in utilitarian forms within household contexts, and use wear on some of these items also

PAGE 151

136 indicates common utilitarian use (Stemp 2014 ). Relatively little debitage from the Mexican obsidian, how ever, appears in Mayan contexts. Product frequencies even outnumber frequencies of debitage pieces, so debitage appears limited to edge retouch and occasional tool modification/restoration. Obsidian was therefore likely exported over the longest distances of inter polity trade primarily as finished products for bifaces and eccentrics, as well as prepared prismatic blade cores for the widest uses (Moholy Nagy 1999:310, Table 2) Spence's view of the role of luxury items in Mesoamerica echoes that of Wa llerstein for the European world system (1974), in that such trades are infrequent and very small in scale, and thus considered relatively inconsequential to the overall economy (Spence 1981:781). However, Schneider's response to Wallerstein (1979:52) also applies to Classic Period Mesoamerica: These occasional items likely represent diplomatic ties that are key factors in maintaining relationships between polities, and thus also forms of inter polity exchange. The regional expansion and partial contro l of Teotihuacn's obsidian trade during the Tlamimilolpa phase corresponds with the likely beginning of the expansion of the polity into the nearby highland valleys surrounding the Valley of Mexico between 200 and 300 A.D. (Algaze 1993:310 ; Cowgill 2000:2 77 280 ) T he expansion also coincides with the expansion or dispersal of the Teuchitlan culture including the occupation of Llano Grande over the same time frame H owever, given the lack of any evidence of material exchange between the two polities, any r elationship between Teotihuacn and the Teuchitln culture would have been an indirect one through common trading partners or perhaps a more general influence on trade interactions throughout the area.

PAGE 152

137 Common Mesoamerican Patterns Several common lith ic distribution patterns emerge from the literature among the two most prominent and influential groups for the Late Formative and Classic periods, Teotihuacn and the Lowland Maya. Remote transportation of obsidian after the Middle Formative is generally in the form of prepared cores, and most products, especially those of lower quality material most suited to common tools, are distributed locally L ower quality material, whether low grade obsidian or locally sourced chert, does not appear to be subject to elite control, but is handled by a relatively uncontrolled direct market system. Despite elite control of production, social stratification was only moderately reflected in the Late Formative lithic data in both areas. Although more obsidian consistently appears in elite rather than commoner households, there is no readily apparent difference in the forms of obsidian used in locations that reflect social position. The same pattern of mixed elite and commoner household use was also noted by De Len et al. f or Late Formative contexts in Central Mexico (2009:119 121), and therefore appears quite consistent throughout Mesoamerica. Both Johnson and Spence interpret the distribution pattern as an initial elite control of the supply with the surplus of higher qual ity obsidian redistributed for common use contexts among elites households, whether to non centralized workshops as raw material in the case of Teotihuacn, or along family lines to households as finished products, as is suspected with the Maya F rom there according to Johnson and Spence, a relatively small amount of obsidian flows downward through the social hierarchy to commoner households, although the mechanism for continued redistribution is not apparent in either case

PAGE 153

138 Both Teotihuacn and the Maya also utilized specialized workshop areas in outlying areas for specific products destined for wider distribution, but still within each city's sphere of control. Area exports from these two polities differ in that Teotihuacan's exports from regional w orkshops are primarily intermediate forms (bifacial blanks), whereas Mayan area exports are often complete, specialized tools A lso, some of Teotihuacan's regional trade workshops appear within the core city area, as well as the outer edge of the city Finally, both areas produce eccentrics and apparent ritual items from high quality material (although among the Maya, some of these items are from a high quality local chert rather than obsidian) which have been located in centralized ceremonial context s within a much larger, region wide area of distribution. These items appear targeted for ceremonial purposes among rulers of remote polities. Teuchitln: Internal Distribution "...this [excavated] mound is near the great obsidian workings of Teuchitlan where miles of hillside are strewn with refuse cuttings." [Breton 1903:133] It has been clear even from the first casual observations made by Adela Breton that obsidian production was a major industry for the Teuchitln culture The initial impress ion of lithics within the Te quila Valley s from early surface surveys was that a high degree of site specialization occurred throughout the region Breton (1905) also

PAGE 154

139 noted differences between waste areas, where polyhedral cores appear quite frequently in s ome areas, but not at all in others. Different sites vary greatly in product types, debitage to product ratios and manufacturing technique (Spence, Weigand and Soto 1980, 2002). A high degree of variability also appears at quarries. Core blade technology, flake cores, and laminar jewelry all appear to be emphasized differently at different sites B ifaces appear in a number of places, but never as the sole type of product (Clark and Weigand 2009:79 80) Debitage distribution observations from the Tequi la Valleys thus far suggest the same pattern of core transport from the source and production at the local destination seen throughout the rest of Mesoamerica. The pattern has been observed through all published study areas within the Tequila Valleys (figu re 4.8) including the core center of Los Guachimontones (Esparza 2003:77 ; Soto 1990 ; Weigand 1985), the La Venta Corridor to the east (Beekman 1996a:753), and around the Magdalena Basin to the northwest (Spence 1980 ; Spence et al. 2002). One excepti onal case is the material from the La Joya source on the eastern side of the Magdalena Basin (figure 4.9) O bsidian from the La Joya source appears to have been distributed through the Magdalena Basin area since the Middle Formative period Most of the La Joya material was found in the form of unmodified material, although it appears also to have circulated in the form of complete blades among several sites (Spence et al. 2002:65,66,70) T he La Joya source is best known for its association with the nearby L as Cuevas workshop, which was very heavily utilized during the Postclassic period (Spence et al. 2002:71 72). However, utilization of La Joya during the Teuchitln occupation span appears very light.

PAGE 155

140 Figure 4.8: Major areas and sites studied within the Tequila Valleys. Adapted from Beekman 2007 :figure 1.01. (Los Guachimontones) Lake Magdalena Basin

PAGE 156

141 Figure 4.9: Tequila Valley obsidian source locations mentioned by Spence, et al. Adapted from Spence, et al. 2002:figure 6.2. Late Formative and Early Classic obsidian distributio n appears to have been limited to local areas surrounding each source location S urveys and a few excavations show that guachimontones are quite often associated with obsidian workshops (Beekman 1996a ; Esparza 2003 ; Soto 1990 ; Spence, et al. 1980), and th ey appear to be involved in some way with obsidian distribution (Beekman 1996a:976). Based on regional differences in platform preparation and the sizes of most workshops throughout the area, lithic production and distribution appears to have involved inde pendent networks surrounding secondary guachimontn centers, rather than central control from the Los Guachimontones site, despite the presence of a very large obsidian production area near Los Guachimontones (Beekman 1996a: 748 749; Soto 1982, 1990 ; Spenc e et al. 1980) which is by far the largest and most artifact dense workshop in the Tequila Valleys (Spence et al. 1980:358), and spans about 1.5 acres ( Soto 1990:217). L. Magdalena

PAGE 157

142 Judging from the lack of overall frequency or grouping of flakes with cortex, the workshop appears to have taken in obsidian in the form of prepared cores. Most, and possibly all, of these cores are from the nearby La Mora Teuchitlan source, which contains an estimated 1000 1200 tons of debris and "millions" of decortication flakes (W eigand 1985:88 89). Spence (1980:357 358) has suggested that much of the lithic mass in the workshop can be attributed to items used for woodworking, rather than lithic products destined for direct distribution. However Soto 's (1990) interpretation has bee n directed mainly toward lithic export, and her dataset does not show a presence of more than a few actual tools and tool fragments A craft workshop context should show a high number of worn tools as they are used up and discarded. Source quality di fferences could have potentially drawn people towards more distant sources, based on Clark and Weigand's ( 2009:79,80) remark that d ifferences in product types appear to be based on the quality and limitations of available materials. However, the evidence i n at least one area suggests that raw material differences may not have been a major consideration for the culture. The more moderate quality local sources in the La Venta corridor appear adequate for most Teuchitln lithic production, and even laminar jew elry, which appears to be the primary elite item of the Teuchitln occupation period, could still be made from the local obsidian L ithic sources used in the eastern part of the region were found to be predominantly from the most convenient e astern volcani c formations of the Sierra de la Primavera, despite the relatively mediocre quality (Beekman 1996a:833, 846). No patterns that suggest centralized obsidian control were seen in the La Venta corridor, except for the laminar jewelry products which were distr ibuted only among elite sites (Beekman 1996a:986).

PAGE 158

143 With adjustments for the later corrections to the architecture timeline (Beekman 2010:64; Beekman and Weigand 2008:315), the La Joya quarry appears to have dominated the Magdalena area during the pea k public architecture period of the Late Formative (Spence et al. 1980:358, 359; 2002:65 66), although some obsidian circulated from an area that includes several sources in the surrounding hills, including Llano Grande, about 25 km northwest of the L a Joy a source. Approximately 38 percent of the obsidian found in the core occupation area surrounding Los Guachimontones is visually similar to that of the hill sources near Lake Magdalena, but most or all of it is likely from the much closer source near San Ju an de los Arcos, approximately 18 km to the southeast (in the opposite direction from the Magdalena hills). The San Juan de los Arcos obsidian falls under the same visual type d escription in Spence's typology as the Magdalena Hills sources (Spence et al. 2 002:68) The primary site of Los Guachimontones appears to have mainly utilize the nearby La Mora Teuchitlan source, just three km northwest of the ceremonial complex (Spence et al. 1980:358 359 2002 67 68 ; Weigand 1985:88,89). Some Llano Grande obs idian as well as a minority of remote materials from Nayarit and Michoacn appear to have made their way to Los Guachimontones ceremonial center (Esparza 2003:75,88; Spence et al. 2002:68). But for the most part, material from sources in the west and north west, including Llano Grande, were limited in internal distribution to the Magdalena basin. Also, no obsidian identified as La Mora Teuchitln material has been located in the Magdalena area (Spence et al. 1980:359). Obsidian export outside of the Teq uila Valleys has been mentioned on a few occasions, but thus far the evidence is scant. Some obsidian artifacts from the Navajas

PAGE 159

144 source as well as San Juan de los Arcos have been located in the Sayula Basin at the site of La Pea. Recovery contexts of the artifacts were determined to be either Middle or Late Formative layers. Most were in the form of debitage, and one included a knife which appears to have been manufactured locally at La Pea, based on dimensional characteristics (Reveles 2006:388). O f 27 successfully sourced n orthern Mexico INAA tested samples, four prismatic blades from sites in s outhern Zacatecas and n orthern Jalisco were traced to obsidian sources within the Tequila Valley s but dates for those deposits are uncertain (Darling 1995, as found in Beekman 1996a:823 824), and as discussed below, were likely from the Postclassic period. Esparza (2003:90) has also determined that obsidian from the La Mora Teuchitln source was traded to areas well beyond the Tierra Caliente region of Micho acn, approximately 320 km from the Tequila Valleys and elemental analysis of a single blade segment found in an Epiclassic context at a Mayan site on Ambergris Cay off the coast of Belize most closely matches sources from the Tequila Valley s but an exac t match could not be confirmed. Although the context of its deposition is very late, blade s likely began to circulate outside of the Tequila Valley s some time before the end of the tradition, around 500 A.D. (Braswell et al. 2011:143). Finally, although no t a definitive match, burials within the Feathered Serpent pyramid in Teotihuacn contain items very similar in form to the laminar jewelry of the Tequila Valleys (although not in technique the Teotihuacn items are bifacially knapped) which are dated t o 150 250 B.C. One issue with Spence's collections is that they are all from surface surveys. Spence himself has stated that although surface densities of artifacts are important evidence, by

PAGE 160

145 themselves they are "not always reliable" (1987:30). Both Spence and Esparza mention the appearance of "fine" pressure blades or prismatic blades, and these items were almost ubiquitous around West Mexican surface ruins, yet too few are associated with excavated contexts to suggest any prismatic blade industry f or the area during the span of occupation for the Teuchitln culture (Darras 2012). Large quantities of prismatic blades within the area rather suggest Postclassic activity in the region. Also, due to funding and time limitations, obsidian sourcing in the area has historically relied on visual properties such as colors, patterns, translucence and tactile characteristics rather than geochemical analysis. Beekman initially utilized visual identification for his La Venta corridor survey, but later submitted sa mples for INAA analysis and found that most samples were from the Primavera source, including three samples originally suspected to be from Llano Grande (1996a:833) The practice of visual sourcing has been common for some time, but it has been teste d against laboratory analysis using XRF with materials from known sources, and was deemed unreliable by Moholy Nagy and Nelson (1990). Following the publication of these tests, analysts have shown an ability to learn from previous test results, and have im proved their success rates from roughly 50 percent to 95 percent (Braswell, et al. 1994; Clark personal communication to Beekman, as found in Beekman 1996a:819 821) B ut in practice, visual sourcing attempts are invariably done by analysts who are not inti mately familiar with the sources associated with the collections under study. However, visual sourcing may still have valid applications in the right circumstances. Moholy Nagy and Nelson's test was primarily run on the least distinguishable gray varieties of obsidian in all but three cases, and more recent pXRF sourcing has confirmed that previous visual

PAGE 161

146 identifications of Teotihuacn's Pachuca green obsidian were correct in 16 of 19 tested samples (Moholy Nagy 2013:75). It must be noted that pXRF is relat ively new and its reliability is less certain than conventional XRF or INAA, but the base XRF technology on which it relies does have a proven track record. T hese later results suggests that more visually distinctive sources might be reliably identified th rough visual observation B eekman and Spence have determined that the lithic sources in the Tequila Valleys are quite visually distinctive, but some visual types, particularly Spence's "Group G," are observed in several sources across most of the area Nev ertheless, results from additional elemental analysis comparisons to visual sourcing, such as one currently pending for the Tequila Valleys (Hoedl 2013), should be used to further test the validity of the approach in the area under study. One item no t included in Spence's survey analysis is the most abundant material at the Llano Grande source location, a fine dark green black obsidian B ased on the visual description, it may be the missing source of the fine "group B" obsidian mentioned by Spence, wh ich he speculates to be from a source external to the Tequila Valley s and only occurs in very small quantities within the V alleys. Obsidian originating from the hills surrounding the Tequila valley basin s is described as visually atypical in color and ton e, relatively poor quality and likely not practically useful for pressure blades (Spence et al. 2002) M y limited casual test knapping experimentation with a variety of visually distinctive sources at Llano Grande suggests that the dark green black variety at that location is quite workable, especially relative to other visually distinct types found near the Llano Grande quarry site which do fit Spence's characterization of the area's obsidian. The source database recently produced by Glascock, et al. (2010 :204, Table 12.1) does

PAGE 162

147 recognize several additional visual variations including the dark green material, and rates the overall obsidian quality at both Llano Grande and Navajas as "excellent". The finer quality material at Llano Grande may have been more r arely distributed within the basin because it was more useful as an external trade item, given the site's difficult access to the interior, and relatively easy access to the periphery discussed in Chapter III Area and Site Studies The few published lithic studies from the Tequila Valleys and surrounding areas to date are primarily based on surface surveys limited to specific areas, and most West Mexican studies are still almost completely descriptive in the nature of the culture history approach (Be ekman 2010:45). Although they do offer valuable data from excavation results for future analyses, publications resulting from these studies are very general; no truly detailed studies of ceremonial circles have been published (Beekman 2007:20). As a result with the exception of Soto 's work (1982, 1990) detailed technological lithic production studies in the area have also been lacking despite the abundance and perceived importance of obsidian to the area. Most publications on lithic studies in West Me xico offer only descriptions and interpretations of processed data, and little access to the data itself, other than glimpses of certain attributes regarding specific points. Also, as purposes for these datasets differ, collected attributes and database de signs also vary. As a result, none of the literature on areas compared in this chapter show perfectly congruent datasets for a comparison of all attributes and calculated results. However, much overlap in methodologies and data

PAGE 163

148 definitions does occur, so t he data which are publicly available and useful for comparison are utilized where possible. Some types within the more detailed typologies are collapsed here into more general "least common denominator" categories. Collapsing accomplishes two goals: It cre ates an additional degree of compatibility between data sets for comparison, and focuses the data on details which are most pertinent to the economic questions and related attribute categories addressed by this thesis. Typologies in West Mexico are us ually a mix of Clark's technological types for prismatic blades, various flake technologies (1997; Hirth and Andrews 2002) with some generalizations and modifications, and functionally oriented descriptions of various retouched tools. Most of the literatur e reviewed here relies heavily on functional interpretation, as does this thesis. The use of subjective morphological characteristics to interpret tool function has been criticized by some lithicists because perceived functions are often assumed a priori a nd unverified. (Hirth and Andrews 2002:1; Whittaker 1994:27). Many functional tool types, such as projectile points, manos and metates, have been verified many times ethnographically and ethnohistorically, and also via contextual and residue analyses. Cert ain functional types have been long accepted by nearly all researchers (but see Aoyama [2005] for a study on alternate use of projectile weapons among the Maya), and some degree of functional interpretation is necessary to arrive at any meaningful conclusi ons. However, unverified interpretations of less understood forms should be viewed with caution, but still considered along with contextual data, as future opportunity to test these interpretations may become available. Four areas and sites are examin ed in this section: First is t he core site of the ceremonial center and second the Feature 83 workshop at Los Guachimontones. Third,

PAGE 164

149 the semi peripheral area of the La Venta corridor, and fourth the peripheral La Pea site within the neighboring unaffiliat ed area of the Sayula Basin. Los Guachimont o nes The primary site of Los Guachimont o nes is chosen as a centralized representation of Teuchitln core culture. In C hapter III environmental and architectural aspects of the Navajas site were suggeste d to be more representative of core culture than Llano Grande. To the extent that the datasets are comparable, a review of lithics in Los Guachimontones should be useful for comparison to lithic data in Llano Grande and Navajas as further described in C hap ter V to determine whether the proposed relative degrees of adherence to the core between the sites is also borne out in the lithic data. Specific aspects of the lithic collections provide additional details about the nature of similarities in what was pr oduced and/or used at each site. Lithics at both the ceremonial complex and the obsidian workshop are described here. Ceremonial Complex The ritual complex of Los Guachimontones is composed of nine guachimontn circles and two ballcourts. 16,000 lith ic artifacts have been extracted from the complex. A lithic spatial distribution study was conducted by Rodrigo Esparza (2003) for the largest features of the ceremonial center. Units of analysis previously defined and mapped by Weigand were used to record the spatial distribution, but for the purpose of the current analysis, only site totals for various artifact types are considered. Artifacts were mainly

PAGE 165

150 recovered from the surface, along with some from several sample pits in building platforms around the two largest circles (I and II), and in the largest ballcourt situated between the two circles. Where possible, distinctive tools were functionally defined, but described by technological form where functionality could not be discerned. Esparza followed Cla rk's technical typology for both blade and flake industries ( 1997; Clark and Bryant 1997 ), although some of Clark's types were generalized into "macro" and "smaller" or un qualified flakes and blades. Some attention also was given to retouch subtypes that h ave been tied to functional purposes (e.g. denticulation, concave and convex edging, and notching). Esparza's typology was designed to determine the degree of spatially differentiated labor between circle buildings, and thus group specialization. Esp arza's Circle II contained the highest number of obsidian pieces, at 5,020. Most pieces in Circle II were flakes, and more than half showed use wear ( Esparza 2003:81). Esparza found that decortication flakes in the site were minimal, and therefore core pre paration likely took place prior to the material's arrival at the complex. Most cores within the complex were casual/multidirectional; very few polyhedral cores were located. Many expedient flakes or flake byproducts of other production activity, including macro flakes, were retouched and utilized as tools ( Esparza 2003:76 78). Esparza also suggests a similarity of guachimontn production strategies to that of the major nearby workshop excavated by Soto (1982,1990), and cites the proportions of blade s triking platform types as evidence. The platform preparation techniques utilized at both locations are indeed similar (Table 4.1), although Soto found a more even distribution with the greatest proportion (47%) pecked, whereas the ceremonial center contain ed a majority ( 70 %) of multifaceted platforms. The platform preparation

PAGE 166

151 technique was therefore more standardized at the ceremonial center This may only be due to a much lower number of people involved in the production process at the ceremonial center s ince more individuals introduce more of their own variations Another possibility is less variability in the types of items being made at the ceremonial center. The most expedient type, the single facet or smooth platform, is a minority in Table 4.1: Pl atform preparation percentages, center vs. workshop. From Esparza (2003:78). Platform type Ceremonial Center Esparza/Ponce (1999 2000) Workshop Soto (1982) Single facet 11% 17% Multi faceted 70% 36% Pecked 18% 47% both contexts, indicating both site s may have emphasized formal production techniques. Some macroblades and over 20 first and second series core blades were found in Circles I and II, and the ballcourts S ome of these were refitted to cores, demonstrating that blade knapping took plac e on location ( Esparza 2003:79) Very few bifacial products and biface blanks were found M ost retouched tools were flake based, and functional interpretations include potential staple processing items such as choppers, knives, "mashers" (perhaps pes tles), as well as awls, polishers, scrapers and scratchers (Table 4.2). There were not enough bifacial products to define a detailed formal typology ( Esparza 2003:80). Items translated as "razorblades" are not described, so are assumed to be a relatively f ine, 1st or 2nd series blade s (figure 4.3), and blade

PAGE 167

152 segments are likely horizontally broken pieces. The only item that can be considered a luxury or ritual artifact is an impact fractured Clovis point tip (verified as a true Clovis Table 4.2: Lithics fr om the Los Guachimontones public center, adapted from Esparza 2003:Table 3 Type N % Blades 1,644 9.9% Fine blades ("razorblades") 1,116 6.7% Flakes 12,072 72.8% Blade s egment s 332 2.0% Points 44 0.3% Preforms 58 0.3% Cores 167 1.0% Specialized tools (awl, chopper, knife, masher, polisher, scraper, scratcher, unkno wn use) 1,142 6.9% Staple tools (Masher pestle?) 5 0.0% Total 16,580 100% from independent sources) in a ball court offering context, which may have been an ancestral offering ( Esparza 2003:80,81). Esparza found that different tool types were emphasized within different buildings, suggesting specialization on certain tasks between groups. But despite the workshop like organization, he concluded th at the center did not include a workshop, primarily based on the predominance of internally worn tools which indicates manufacture for local use (2003:90). Esparza does not mention any prestige or ritual items other than the Clovis point and other archaic period points previously extracted from a ritual deposit in one of the smaller circles within the complex. None of the distinctively Teuchitln prestige forms (pikes, rods or laminar flake items) are specifically mentioned, although a category for items of "unknown use" included 21 artifacts (0.1 percent of the total count).

PAGE 168

153 The Los Guachimontones study does suffer from the same problem encountered by Spence et al. (2002) in his regional distribution study, in that most of his data are from surface co llections H owever, Esparza's data appear contemporaneous with excavated contexts produced thus far for the period Only a very small percentage of first and second series blades were located compared to the number of blades from earlier stages (1.2 percen t of the total number of blades), and 3rd series (prismatic) blades were not present. Feature 83 Workshop The primary Los Guachimontones workshop area designated Feature 83 was thoroughly analyzed by Maria Soto (1982, 1990). It is in an entirely dif ferent context from the ceremonial centers under study, but with the contextual difference in mind, provides additional clues about the nature of lithic production and use in the central core area in general, and also the nature of lithic workshops in the region. The workshop area is quite expansive compared to any other lithic concentration in the region; Soto defined a 75m x 78m grid for the area which covers approximately 0.6 hectares, or 1.4 acres (1990:217). Like Esparza's ceremonial center study Soto examined the differential distribution of lithics within the site, but concentrated more on the distribution of technological forms in the far more abundant debitage to determine labor divisions and degrees of specialization for different production stages as well as tool forms. Soto very closely followed all of Clark's polyhedral blade forms (Clark 1997; Clark and Bryant 1997; Hirth and Andrews 2002) and other commonly recognized attributes such as core platform preparation types

PAGE 169

154 and platform angles to determine the clustering of attributes within distinctive concentrations of debitage. Again, at the scale of the current study, only the emphases on specific technologies and forms from the site as a whole are reviewed for comparison. As with mo st other Teuchitln contexts (excluding only Spence, et al's [2002] identified use of the La Joya source within the Magdalena basin), the lack of any appreciable quantity, size or clustering of decortication flakes indicates that obsidian arrived at the wo rkshop already prepared as macrocores (Soto 1990:225) S oto located an abundance of virtually every type of polyhedral product, byproduct and debitage specified by Clark, down to but not including final stage p rismatic blade manufacture (Table 4.3). She al so noted use wear on some items which indicated some utilization for other activities within the workshop area, but only occurring on 1 to 4 percent of items for each typological category, with an overall percentage of 2.16. Nearly all of the use wear (98. 4%) was within a single cluster which was also the only concentration out of 11 clusters that contained no microdebitage indicative of a lithic production area Table 4.3: Lithics from the Feature 83 workshop, with used item percentages. From Soto ( 1990 :238, 240 ). Type N % of Total Used % of type used Crested macroblades 269 0.3 6 2.23 Macroblades 20,179 20.2 226 1.12 Macroflakes 10,000 10.0 95 0.95 Blades 28,578 28.6 1246 4.36 Flakes 23,364 23.4 250 1.07 Cores 45 0.0 1 2.22 Ground stone (manos, metates) 11 0.0 n/a n/a Debitage (platform preparation, error correction flakes, flake and blade fragments) 17,390 17.4 91 0.5 Total: 99,836 100 1915 n/a

PAGE 170

155 (Soto 1990:238). The same cluster was also the only one which contained evidence of ground stone products, specifically four metate fragments and seven manos, as well as the majority of ceramics in the area T he cluster was therefore interpreted by Soto as a living quarters and the rest o f the workshop was determined to be designated solely for the manufacture and redistribution of lithic products within the area. The majority of forms were related to polyhedral blades, with a parallel industry for the generation of large flakes for flake based blanks. Soto suggests that most blanks were likely designated for exchange as preforms (1990:241), as also suggested by Spence for similar items found in Teotihuacn (1981:771). Soto does not detail exact quantities of retouched blanks or tools remai ning in the shop in her synopsis of the workshop data. She does, however, report that all of the retouched pieces are either broken or show signs of production failure, which indicates that all viable finished products had been exported. Also, none of the retouched pieces outside of the living quarters assemblage show signs of use (1990:241). No pikes, laminar flakes, flake jewelry, or debitage related to either eccentric industry are mentioned by Soto in relation to the workshop. The La Venta Corridor Beekman's (1996a:123 124) study of the semi peripheral La Venta Corridor area also primarily relied on a surface survey, with sampled areas which show unusually high artifact density more systematically and comprehensively covered. Sample excavations a t targeted sites followed the survey (1996a:144). Beekman also notes that the majority of

PAGE 171

156 fine blades recovered from the area are first or second series items, rather than third series (prismatic) blades (1996a:756), and fine blades are also very low in nu mber (Table 4.4). Therefore, like the Esparza collection, the blade types present suggest that the majority of items in the area are contemporaneous with the Late Formative and Early Classic sites. Various functionally specialized tools are defined for the area, as well as rough bifaces and point preforms (Beekman 1996a:763,768,figures 6.3 6.5). The latter two items may have been utilized as bifacial blanks for trade purposes, as described by Spence (1981:771,781). Beekman's ( 1996a:756) lithic analysi s for the area also used much of Clark's typology, but eliminated Clark's distinction between macroblades and blades, and also between macroflakes and flakes because the "macro" distinction is typically defined by size rather than morphology, which is also dependant on the initial core size The study also included some common morphological distinctions of debitage not present in other studies of the area, including some types specific to core blade debitage. More specific types are collapsed into more basi c categories here, as noted at the beginning of this section. Laminar flake based items appeared in the area, and include various geometric shapes such as anthropomorphic figures, lunates, discs, crosses and axe shaped items. Whole flakes are noted t o consistently possess a bulb of percussion and some rippling on both sides, and the markings are oriented in different directions on either side. The Table 4.4: Lithics from the La Venta Corridor (nodules excluded). Adapted from Beekman 1996a:Table 6.1. Type n % of total Debitage (shatter, blade core fragments and 2 196 45.2

PAGE 172

157 f lakes, cylindrical core) Flake 2 061 42.4 Finishing Flake 53 1.1 Macrocore 22 0.5 Casual core 1 0.0 Percussion blade 312 6.4 Fine blade 6 0.1 Scraper 25 0.5 Stripper 12 0.2 Multi functional (awl/knife, awl/scraper) 15 0.3 Knife 15 0.3 Rough biface (blank/preform) 7 0.1 Chopper (Reworked core fragment) 1 0.0 Chisel 1 0.0 Ground stone, staple processing (manos, metates, molcajetes) 10 0.2 Unknown (too fragme ntary or no known use) 10 0.2 Points 74 1.5 Cylindrical core flake 17 0.4 Cylindrical core jewelry 15 0.3 Pikes and rods 10 0.2 Total 4 863 100 evidence supports Long's ( 1 966:228 230) proposed "sausage" manufa cturing technique for these items, since turning the core between blows as Long described would produce this pattern. Many laminar pieces which may have been broken during manufacture were found within the site, and Beekman suggests that the site was a wor kshop dedicated to laminar jewelry (1996a:790, 795 799, figure 6.16). Several pikes and rod ornament fragments were also located throughout the area (Beekman 1996a:802). Artifact counts from the s urface collection were compared to those recovered from excavated contexts. Laminar jewelry was found to be very much underrepresented by survey data, and all other types decreased in number below the surface. There is little evidence of differences in both production and distribution of various product types

PAGE 173

158 between sites, except for the sites at Cerro Tepopote, which shows a much lower than average count of core/blade debitage (Beekman 1996a:805). Beekman ( 1996a:742) notes that the political boundaries of the La Venta corridor defined by the walled semi peripheral sites did not affect the distribution of artifacts, which showed no change in density on either side of divisional walls; the local economy therefore appears to have followed an independent course from any attempted political control. Artifact t ypes collected from the area include very small quantities of prismatic blades, some discoidal scrapers, jewelry, long narrow points and long handled scrapers. Beekman also notes some difficulty among less experienced staff in determining natural obsidian cobbles from artifacts, due to the abundance of naturally occurring obsidian throughout the area. The problem is compounded by the prevalence of more expedient and therefore less obviously human altered tools (Beekman 1996a:750). Very little cortex in the La Venta corridor sites once again indicates that cores were prepared before distribution to consumer sites (Beekman 1996a:753). The Sayula Basin The Sayula Basin is the only site in the periphery which contains verified obsidian from the Tequila Va lleys Excavation in the Sayula site of La Pea in 2000 and 2002 yielded 4,301 lithic artifacts, of which 3,203 (74.5%) were interpreted as imports, and 3,180 of the imports (99.3%) are obsidian. Most of the remaining imports were of elite materials which included greenstones of unspecified mineral type, pyrite, chalcedony and specular hematite (Reveles 2006:373).

PAGE 174

159 Some obsidian from the site has been sourced to the Tequila Valley s mining sites of Navajas and La Joya via neutron activation analysis (NA A) O ne obsidian knife is also sourced to Navajas. Based on dimensional characteristics, the knife is of local manufacture (Reveles 2006:388) Other items were visually sourced to the Teuchitln sites of San Juan de los Arcos and La Mora Teuchitln; as wel l as the Teotihuacn controlled Pachuca (Cerro de las Navajas) source in Hidalgo. Of the visually identified sources, all except one (Ixtan del Rio) have previously been confirmed within other parts of the Sayula B asin via NAA T he use of these distant sou rces was necessary for the area, which has no obsidian of its own (Reveles 2006:373), and obsidian was utilized in the Sayula Basin more than any other lithic material (Reveles2005:350). The proximity and abundance of obsidian in the Tequila Valleys made t he Teuchitln culture a highly desirable primary trade partner. However, only 582 lithic pieces were attributed to contexts which were contemporaneous with the Teuchitln culture. These were discovered in Atotonilco and Usmajac phase contexts (Middle Formative into the Early Classic.). The two phases needed to be combined because they could not be accurately differentiated at the site for knapped tools (Reveles 2006:374), although Reveles also notes that this is partly because there appears to be cons iderable technological and morphological continuity between these phases (2006:387). 574 of these items were knapped, and they include 556 debitage pieces. The majority of the obsidian debitage is from polyhedral cores. Artifacts include nine obsidian proj ectile points, one obsidian biface (an elongated item suggested to be a knife), one andesite scraper likely made from local material, four polished items (two pebbles, a piece of partially polished slate and a mano) and four jewelry pieces T wo of

PAGE 175

160 the jewe lry pieces are specular hematite flakes, one of which was perforated; the remaining two are earrings, one of chalcedony and the other of a "fine volcanic rock" (2006:374). Some of the more distinctive points are very wide triangular items with nearly equil ateral dimensions and wide, short rectangular stems (Reveles 2006:387), and resemble some of those found in Teuchitln contexts (e.g. Be e kman 1996a:figure 6.14f). Reveles' description of the Sayula Basin's material culture through most of the Formati ve and Early Classic periods reflects a group primarily engaged in a wealth trade economy. Types and styles of artifacts in the area were diverse, and most often included similarities in styles to several surrounding groups. Foreign artifacts increased in number during the basin's Usmajac phase (200 B.C. 300 A.D.)(Liot et al. 2006:415 416). Most of the lithics were also retrieved from contexts dated to the Usmajac period, described by Reveles as "pre agricultural" (Reveles2005:351) which from Earle's per spective would likely indicate investment in a wealth economy S ome ground and pecked tools also were located. Artifacts are described as diverse, and include more categories of items and diversity of forms than reported in any Teuchitln context. Artifac ts from the area include anthropomorphic items, necklaces, lip plugs, eccentrics, drills, end scrapers, racloirs, points and knives with only a small presence of blades and prismatic blades. Direct percussion appears to have been used until the very last f inishing step in the production process for most tools, including blades. No exhausted cores were found, and Reveles suggests that finished blades were imported. Oval, semi circular and circular scrapers are common S ome necklace pendants and beads made of obsidian were located in a Sayula shaft tomb (Valdez 1998), and some rough pikes and laminar anthropomorphic figures have also been recovered from the area (Reveles2005:351 352).

PAGE 176

161 If blades were imported, the Sayula Basin is the one area that essenti ally breaks the post Middle Formative pattern of local blades manufacture. As discussed earlier, however, additional evidence of core blade industry byproducts may be necessary to determine the presence of a local core blade industry. The focus on lithic imports and prestige items, coupled with a paucity of items suitable for food processing are clues to a heavily wealth oriented economy a short distance from the Teuchitln semi peripheral zone. Inter Area Comparisons One additional caveat must be n oted for any comparison of literature on previous lithic studies in the area: The few datasets available with comparable type and frequency data are from different scales of research. The La Venta Corridor dataset is an area wide collection, which may incl ude a wide range in types of assemblage contexts, whereas the two Los Guachimontones studies are at the site level. Fortunately, source sites at La Venta were avoided (Beekman 1996a:751), as were shaft tombs during the excavation phase (Beekman 1996a:144) but assemblages frequently included midden deposits (1996a:156), and likely household deposits along with the production and public centers which better correspond to the contexts of the locations at Los Guachimontones The Feature 83 workshop area did inc lude a likely dwelling unit, and the public center only shows clear evidence of the use of utilitarian objects, but such use is likely to have differed to some degree from that of households. Beekman's observation of the nearly exclusive presence of lamina r jewelry in the few excavations done in the La Venta

PAGE 177

162 Corridor (1996a:805) require some consideration that such items were very much underreported in the surface survey data. A concentration on only excavated contexts would drastically reduce the sample si ze, Nevertheless, these datasets are what we have available, so comparisons must be made with these noted differences in mind. As shown in Table 4.5, the La Venta Corridor dataset is dominated by debitage and flakes. Flakes are also prevalent in the Los Guachimontones sites, but La Venta's proportion of debitage differs from both Los Guachimontones contexts by the widest margin of any category, at a d ifference of 27.8 percent from the Feature 83 workshop. Approximately 98 percent of the debitage categ ory from La Venta is classified as shatter, or broken and unidentifiable worked pieces. A consumer context such as the Los Table 4.5: Percentages of each lithic type per location Location La Venta Corridor Los Guachimontones Ceremonial Center Los Guachi montones Feature 83 Workshop Blades 6.5 16.6 49.1 Flakes 43.5 72.8 33.4 Points 1.5 0.3 n/a Blanks/Preforms 0.1 0.3 n/a Cores 0.5 1.0 0.0 Staple tools 0.2 0.0 0.0 Other tools 1.3 6.9 n/a Eccentrics 0.9 0.0 0.0 Debitage 45.2 2.0 17.4 Unk nown 0.2 0.0 0.0 Total 100 100 100 Unspecified by Soto (1990); the only remaining retouched artifacts are production failures. Guachimontones ceremonial center would not be expected to contain a large quantity of shatter, and a polyhedral core worksho p starting from prepared cores may contain mostly

PAGE 178

163 readily identifiable pieces since the distinctive products and byproducts of the industry are easily recognized, even in a broken state. The prevalence of shatter at La Venta is likely due to an emphasis on other industries, such as expedient knapped items, bifaces or eccentrics. Midden contexts might also show a much higher concentration of shatter, since shattered cores and macroblades/flakes often have no reuse potential. La Venta 's blade industry is quit e low, and contains a 10.1 percent lower proportion of blades than even the flake domin ated ceremonial center. The ceremonial center shows a preference for flake tools over blades by a wide margin, even though the workshop emphasized more blades than flake s, which suggests a wider area of distribution than the ceremonial complex for the workshop. It also suggests that the balance of the distribution outside of the ceremonial complex, which would have been primarily household contexts, mainly included blades and blade products. As discussed above, other Mesoamerican contexts which include Teotihuacn, the Maya and Oaxaca, primarily used polyhedral blades in household contexts for a wide variety of utilitarian purposes The comparison also suggests that the lithic industry within the La Venta Corridor is only partially similar to Teotihuacn's regional workshop industry. Unlike the more distant Teotihuacn workshops, bifacial blanks do not appear in greater numbers in La Venta. R ather b lanks appear to be more popular for the creation of tools within the central ceremonial center, most of which were likely provided by the parallel flake industry of the workshop. However, like Teotihuacn's workshops, La Venta contained eccentrics (laminar jewelry, pikes and rod ornaments) in small quantities which were likely used, at least in part, for elite trade with nearby external groups. These items are not seen in the public areas within the core This does not suggest that such items were

PAGE 179

164 rare or absent in the core. Rather, the lack of evidence for production or use of these items in the primary workshop or the larger components of the ceremonial center at Los Guachimontones indicates that eccentrics were likely not particularly emphasized for either trade or corporat e ritual. As network acquired prestige items, they may only appear in smaller kin oriented production contexts and shaft tombs within the core, where the networked wealth trade institutions could (quite literally) maintain a low profile in an area which em phasizes corporate ties over networked ones. Again, the data likely reflect a difference in relative degree of utilization, rather than a categorical difference. Stone tools specific to the processing of staples are quite few in number for a comparis on of relative percentages of entire datasets, and these are invariably ground stone items, which may arguably fall into a different category from knapped tools. The La Venta corridor does marginally contain the highest relative proportion of these items, which is likely due to a frequent occurrence in household assemblages. Three of the ten items (a mano and two metates), however, were found in a La Venta burial chamber (Beekman 1996a:317). The ceremonial center contained five pieces that Esparza termed "m ashers" (2003:Table 3), which may have been pestles or items of a similar function. The items was therefore tentatively placed in the staple category. Oddly, the Feature 83 workshop contained the highest number of staple specific items (N=11), namely the s even manos and four metate fragments in the living area noted above. However, o ther retouched tools were given somewhat more generic functional categories such as choppers and knives, which have a number of applications but are also common for staple proce ssing. These utilities appear in a higher concentration in the ceremonial center than in La Venta. Soto mentions the presence of retouched tools in one group at

PAGE 180

165 the workshop (1990:241), but her synopsis of the investigation does not include specific number s for them. N one of the retouched products were very frequent in any context compared to flakes, blades and debitage However, the percentage of projectile points in the La Venta corridor, at only 1.5 percent was still five times that of the ceremon ial center at 0.3 percent Some increase in points may be expected for defense in a semi peripheral zone which pursues an exclusionary economic strategy, or which functions as a border monitoring area. Points may also find increased utilization for hunting game (Beekman 1996a:52) to supplement staple crops in agriculturally poor semi peripheral zones. The difference between the point percentages at La Venta and the ceremonial center, however, is minimal at 0.2 percent. Teuchitln material and eccentrics whi ch include some similar forms appear in at least one known peripheral area known for its trade economy, along with eccentric items of other styles which may be locally produced, or from other trade partners. Conclusion Several patterns are shown in the external and internal lithic data comparisons which can be used to inform expectations specified in Chapter V Two of these are counter intuitive F irst, unlike Teotihuacn, t he percentage of bifacial blanks produced at the semi peripheral zone of La Venta for possible trade opportunities is quite low, and does not surpass those utilized in core A lso, staple processing tools are more frequent in the semi peripheral zone, which is suspected to be less agrarian than the core As stated

PAGE 181

16 6 above, the latter result is likely due at least in part to a mismatch in types of sites. The La Venta survey includes a combination of public and private contexts that includes households, and the Los Guachimontones sites are public contexts, each with a very specific purp ose. The lack of bifacial blanks appears to simply reflect a lack of preform exchange in the corridor. Since groups in different semi peripheral areas within the same culture can engage external groups in different ways, forms of exchange not seen in the L a Venta corridor may still occur elsewhere, but the likelihood of the practice is diminished because the corridor does not provide a precedent for it. The rest of the comparisons support the notion of a semi peripheral area which is more involved in wealth trade and less involved in staple finance than the core area. These include a decreased proportion of blades and other utilitarian tools commonly used for staple processing, and increased proportions of points and eccentrics. Even though the pe ripheral Sayula Basin data do not include comparable statistics, they do provide ample evidence for an active exchange relationship with other groups in the area, as well as evidence of a strong wealth trade ec onomy. Te quila Valley s obsidian has also b een traced to the Sayula Basin and a diverse array of eccentrics including some forms strongly associated with the Teuchitln culture, have also been found in the area This review has shown some patterns in the lithic data between the central core area, one semi peripheral area and one peripheral area, but has also revealed a need for isolated ceremonial center and workshop lithic studies from both semi peripheral and peripheral areas W e can now compare and address expectations for the ceremonial center within the semi peripheral site of Llano Grande, and the distant but strongly core affiliated site of Navajas T he case of Navajas becomes especially interesting in light of a known possible

PAGE 182

167 trade partner in Sayula, and a more balanced combination of trade and staple emphases may be demonstrated at the site, which would show an exceptional case to Blanton et al.'s (1996) and Earle's (1991) expected spatial separation between staple and wealth economies.

PAGE 183

168 CHAPTER V METHODS This chapter introduces t he lithic assemblages recovered from the guachimontn public architecture at Llano Grande and Navajas. Methods of recovery and data collection are then discussed, followed by methods of comparison. The Llano Grande and Navajas Datasets All lithic a rtifacts under study were previously retrieved from Llano Grande Group 14 and Navajas Circle 5 guachimontones I n the case of Llano Grande, one building very near the guachimontn circle and apparently associated with it was also excavated, but contained o nly a few pieces At both sites, artifact locations were tracked with 2x2 meter excavation units used in conjunction with a numbered lot system. Lots define site features delimited by vertical soil and architectural layers as well as horizontal archi tectural areas, such as buildings, patios and altars. Some lots also define features which c ut through layers and disturbances such as postholes, tree roots and looters' pits F ollowing the collection of artifacts in each unit and lot, all remaining soil w as run through 1/4" mesh screens to capture remaining fragments (Beekman 2001:6, 2007 :20). One major difference between locations which affects the analysis is that only a small proportion of the Llano Gra nde guachimontn was excavated, namely three buildings adjacent to the circle, the one off circle building noted above, and a trench across the

PAGE 184

169 diameter of the circle (figure 5.1). At Navajas, six of the eight buildings were excavated, along with the majority of the patio and the central altar (figur e 5.2). Therefore, between site comparisons are restricted to relative proportions and artifact densities Figure 5.1: Llano Grande plan, with excavation units. From Beekman 2008:figure 4

PAGE 185

170 Figure 5.2: Navajas plan, with excavation units. From Beekma n 2007:figure 2.91. One environmental difference between sites which affected the collection of artifacts was the ubiquitous presence of naturally occurring obsidian cobbles gravel and fragmented spalls at Llano Grande, which presented the same prob lem as La Venta in the identification of human altered items among the natural ones (Beekman 2001:4). As a result, many natural pieces were retrieved, stored and reported among the artifacts. Since Navajas was 2 3 km removed from its obsidian source, such natural cobble "noise" was not an issue, and recorded artifacts were primarily isolated to one or two layers immediately above the activity floors

PAGE 186

171 Preliminary analyses were performed by Dr. Christopher Beekman for the lithic artifacts at Llano Grand e, and by Bruno Calgaro for the lithics at Navajas. For Llano Grande, context information was recorded by unit and lot, along with typological classifications weights, visual material properties and cortex presence These data were recorded and entered in to electronic spreadsheets. Many items were grouped by type and entered as individual records, with a count entered for the number of pieces in each record. T he general data format was not usable for the current project but most of the typolog y was used as a starting point for the data definitions. The Navajas data included more detail on individual pieces and the data definition was quite similar to the one I had extended from the Beekman data but was handwritten in paper tables and only photocopies wer e available to the lab, which were largely unreadable Therefore, artifacts needed to be re examined and recorded for the current project, from both sites Artifacts were stored in a back room at the Tala Museum according to their site contexts by lot and excavation unit identifications T he time window for access to these artifacts was limited and the required time to record the necessary data was uncertain, so debitage artifacts were sampled and transported according to prioritized lots P riority and thus lot order for analysis was based on likelihood of contemporaneity with the architecture and the degree of protection from post occupational disturbance L ots directly above and below the latest activity floors in each excavated building received the highest priority and were therefore the first prioritized for examination, with lots immediately below each activity floor receiving secondary priority I n most cases, activity floors were described in the excavation project documentation as well defined c lay layers atop gravel subfloor, although in some areas the clay was not preserved,

PAGE 187

172 which at times posed challenges to floor and activity lot identification at Llano Grande (Beekman 2001:appendix). Activity areas above previous floor layers were targeted n ext, and finally central patio artifacts were examined. All debitage pieces within each targeted lot and unit were examined. Due to laboratory space and transportation limitations between storage and our hotel room lab areas, artifacts were transported and examined for one site at a time. Navajas was examined first, and all of the priority areas were covered for that site. However, the higher volume within the Llano Grande guachimontn did not allow time for examination of debitage within its central patio Due to the very small sample of obsidian products (including decorative and ritual items as well as tools) at both centers relative to the amount of debitage, product analysis received secondary priority. All products were examined, but only for a v ery limited number of attributes. Observations and measurements for each item were recorded in Microsoft Excel spreadsheets according to lot and unit location data, and all items within each unit and lot combination were given unique sequential IDs T he s preadsheet data was also imported into IBM's SPSS statistics program for statistical comparisons between sites Attributes Debitage For debitage, recorded physical attributes pertinent to the current project are morphological type, cortex co verage, length, width, and flake scar count. Physical attributes were recorded as follows:

PAGE 188

173 Morphological type General shape characteristics were used to determine a basic category for each piece. The type characterization was then used to de termine which of the following attributes applied to the piece D efined types relied on previous work in the area and leaned towards an expectation of a dominant percussion blade industry for the region ( Calgaro 2003 ; Darras 2012 ; Soto 1990 ; Spence et al. 2002). Types are grouped into two broad categories: General types, which may be either part of a casual flake or biface industry, or byproducts of any more formal industry; and polyhedral industry types that are specific to percussion blades and polyhedral macro cores, following Clark's observed reduction sequence (1988, 1997). Basic categories such as "debitage" and "tools" were pre defined for the purpose of organiz ation and stor age of artifacts by these categories, so they were kept as is D ebitage categories generally follow Spence's scheme, where flakes, blades and any expediently retouched items which do not form recognizable formal tools were included as debitage (2002:63) R etouch is recorded as a separate attribute for these items. Debitage a nd e xpedient t ool t ypes are defined as follows : casual core: A core with unpatterned, multidirectional flake scars. Chunk: Follows Beekman's (1996:754) definition of the term (not to be confused with Clark's definition [1988:16, 1997:125], which define s a "chunk" as synonymous with this project's definition of shatter). A chunk is an angular piece of workable

PAGE 189

174 material without any specific indications of human alteration, even though it may be within a debitage context. It may well have been a byproduct of stone manufacture, but no reliable evidence is present. Shatter: A piece defined as shatter may resemble a chunk, but contains at least one distinctive indicator of knapping, such as one or more positive and/or negative bulbs of percussion, or concent ric rippling. However, the piece also bears an ambiguous form that cannot be oriented in such a way that it defines a core, flake or blade. Macroflake: Macroflakes are generally created early in the reduction process, primarily to reduce the size of the core while working towards a general core shape. They usually hold at least some cortex, and are relatively thick, irregular, and often much larger than later stage flakes. Macroflakes are invariably hard hammer pieces with prominent bulbs of percussion. Shaping/finishing flake: Shaping and finishing flakes are relatively flat, uniform and thin ( Spence et al. 2002:63,64 ) with little to no cortex, and are usually moderate to small in size ( Clark 1988:17 ) with a major axis typically under 4 cm Types specific m acro/polyhedral blade production are defined as follows : Macrocore: An early stage polyhedral blade core, as defined by Clark and Bryant (1997:fig. 3), that contains primarily vertical but irregular dorsal scars. blade core fragment: A f ractured macrocore or blade core. Macroblade: A large, irregular blade generally wider than 2.5 cm (Clark 1997:113). 1st series blade: A uniformly shaped blade generally smaller than 2.5 cm, but with irregular flake scars. 2nd series blade: A unifo rmly shaped blade much like a 1st series blade, but with only one or two vertical dorsal arrises extending from the proximal end towards the center, and irregular flake scars from the same point near the center to the distal end. Platform correction fla ke: A round, polyhedral or semi round/polyhedral flake with an

PAGE 190

175 inclined edge thickness that creates a wedge shaped profile. Created from a horizontal removal of the striking platform for the purpose of leveling the platform. Core correction flake: A fla ke with a series of straight and parallel dorsal scars, either horizontal or vertical, resulting from removal of an error on a polyhedral core Natural object : A stone object of either rare or visually striking material, or with a simple geometry that suggests a human crafted product, but shows no reliable signs of manufacture such as a bulb of percussion or impact rippling. The object may even be nearly or completely covered with a finely textured cortex. Its presence within a ritual space is suspected to be intentional. As stated in the type descrip tions, the distinction between reduction and shaping/finishing flakes i s according to size, morphology and/or the presence and amount of cortex (figures 5.3 through 5.5) T he size distinctions par allel Clark's typological distinctions between macro flake s and flakes (Clark 1988 ; see also figures 4.1 and 4.2 of this thesis ) In the current typology, t he "shaping/finishing" prefix is added to the smaller flake category in this project, in order to avo id confusion with the more common use of Figure 5.3: Reduction flake Figure 5.4: Small Figure 5.5: Large shaping/finishing flake shaping/finishing flake

PAGE 191

176 from a lithic core. The use of morphological characteristics as added criteria for identification borrows from Spence et al. 's blade typology which distinguishes between rough and irregularly shaped flake blades and uniform fine blades with parallel edges (Spence et al. 2002:63 6 4) S pence et al. note that their fine blades closely match Clark's 2nd and 3rd series blades However, shape unifo rmity is used in a somewhat different context within this thesis, to distinguish the two defined flake types as well as earl ier blade sequence distinctions between macroblades and 1st series blades. Clark's typological distinctions are based on exper imental observations. However, a s Clark himself has stated, his method of identifying flakes of various types is essentially subjective (Clark 1988, Clark and Bryant 1997). A width division of 2.5 cm has been suggested as a more concrete means of distingui shing between blade types (Clark and Bryant 1997:113), but the figure is based on past experience in certain areas, and subject to localized variations. Beekman based much of his own West Mexican lithic typology on Clark's work, but chose not to utilize th e macroflake category in his typology because flake size is also largely dependent on the size of the original core as well as the stage or degree of reduction (1996a:755 756). Spence et al. 's definitions also move away from Clark's size dependence to som e degree by including morphological characteristics which can be associated with each type (specifically, relatively parallel lateral edges and a more regular overall shape without angular protrusions associated with shaping flakes ). However, Spence does n ot indicate any measure of the degree of uniformity required to distinguish between types. The use of subjective morphological characteristics has been criticized in the past by lithicists, as they have lead to assumed designated tool functions based solel y on perceived morphology that may be incorrect (Hirth and Andrews 2002:1;

PAGE 192

177 Whittaker 1994:27), and are prone to inconsistent interpretations and labeling between analysts (Whittaker et al. 1998). The macroflake/flake distinction does, however, repres ent a distinctive step in a known formalized production process based on ethnohistoric data (Clark 1982) The problem is in verifying the correct identification of each step's products in the field. Therefore, a measurable and objective statistical test f or distinctiveness between items labeled as reduction and shaping flakes is provided for the Teuchitln data from both Llano Grande and Navajas, in the form of Kolmogorov Smirnov tests. In the current context, t he Kolmogorov Smirnov tests indicate whether the two groups represent genuine differences in frequency distributions by size. The results of the tests consistently show that the size distribution curves are significantly d ifferent at the p<.05 level Since any one or more size measures may show a significant distribution difference between the two types, three size attributes were evaluated: l ength, width and the product of length and width (labeled "lxw"), where lxw is considered a rough approximation of the ventral surface area. As shown in Tab les 5. 1 and 5. 2, s ignificantly different size distributions were found for all tested size variables (p<.001 for all dimension variables at both sites, except for length in the Llano Grande data, where p=.001). Note that there are no positive extreme diffe rences (a Kolmogorov Smirnov term for maximum points of difference between data relationship curves) for any size category, although each size parameter shows a negative extreme difference. Clark's reduction flake type was used as the initial or "control" group for all cases, so the lack of positive differences simply shows that those items labeled as reduction flakes are consistently larger in each

PAGE 193

178 dimension. Since Clark's primary criteri on for separating these types is size, these results support his obse rvations. Sample sizes in these tables are different for length width and lxw because dimensions truncated by breaks were not recorded. This was done to insure that recorded samples reflect original knapped dimensions, rather than post production f ragmentation from trampling and post occupational formation processes. Table 5. 1: Llano Grande size frequency distribution comparison for different flake types a l ength (n=2 80 ) width (n=297) l xw (n=231) Extreme Differences Positive .000 .000 .000 Ne gative .243 .358 .336 Asymp. Sig. (2 tailed) .001 .000 .000 only) Table 5. 2: Navajas size frequency distribution comparison for different flake types a length (n= 146 ) width (n= 132 ) l xw (n= 123 ) Extreme Differences Positive .000 .000 .000 Negative .4 85 .5 04 .582 Asymp. Sig. (2 tailed) .000 .000 .000 only) Cortex coverage Rather than utilize a typical b inary "presence/absence" criteria for cortex, following Mauldin and Amick (1989:68,70,Table 1), cortex coverage quartiles were recorded to

PAGE 194

179 expediently obtain a finer distinction between cobbles and cortical flakes that had very few flakes struck from them, and those which may have been worked to a point just short of a finishing phase with a very small amount of cortex T he cortex coding scheme is defined as follows: 0 = no cortex present, 1=1 25 percent cortex coverage, 2=26 50 percent coverage, 3=51 75 pe rcent 4=76 99 percent, and 5=full coverage (a natural cobble). For the needs of the current project, t his scale is co a rse enough to allow an accurate visual estimate, and fine enough to avoid the aforementioned overgeneralizations A lso in keeping with th e Mauldin and Amick method, o n flakes and blades, the ventral surface is not counted as part of the total area for the estimate, since by definition it cannot hold cortex (1989:68) Metrics Due to a high degree of fragmentation at both sites, onl y complete edges were measured for length and width L ength and width measurements were done with standard lab calipers to a 1 mm resolution Following Odell (1989:168), length was measured as the maximum distance from the platform to the distal edge, perpendicular to the striking platform A s Odell also notes, and especially with a preponderance of irregular flakes that are often struck at a lateral angle that is not perpendicular to the platform, this is not always the major axis, or the maximum leng th when following the direction of the strike. This method was chosen for its repeatability in the field with irregular flake morphologies, and to allow for comparability with other datasets.

PAGE 195

180 Width is defined as the maximum distance between lateral e dges, as measured perpendicular to the length measurement (and thus parallel with the striking platform). Due to the same flake irregularity problem described for length, the width measurement defined here is not necessarily the geometric minor axis of the piece. Products The term "product" is preferred here over "tool" in contexts which may include items that are ornamental or ritual in nature P roducts are divided by type as points, blades, bifaces, scrapers, and miscellaneous tools and eccentrics T he latter category includes apparent ritual and jewelry items, as well as apparent specialized tools that do not fit any common lithic categorization. Three basic attributes are used in the same manner with all product categories as they are with deb itage T hese attributes are length, width and cortex. Descriptive names are used in place of type codes because product forms are very diverse, yet sample sizes are quite small and therefore groupings of common formal attributes beyond the most basic type s could not be determined. Descriptions included form attributes commonly referenced in lithic studies, such as stem, notch and blade forms, so that examples from similar sites in the region can later be incorporated to determine groupings of common morpho logical attributes that can define a useful typology O ther tool/product attributes were recorded beyond basic categorization and description, however that level of detail is not required for the current project.

PAGE 196

181 General Approach for Analysis Analy sis proceeded in three phases: First, p ercentages of generalize d formal types from both sites were compared to those from the previously discussed datasets between the site of Los G uachimontones and the La Venta corridor to determine whether similarities a nd differences in the lithic industries follow the general material culture patterns seen in core and peripheral areas, as described in C hapter III The analysis then moves to comparisons between the sites of Llano Grande and Navajas. Since Navajas contain s the only excavated guachimontn circle of a comparable size to Llano Grande and both sites have comparable recorded l ithic attributes, additional c omparisons are made between the two sites Second, e ach site's debitage collection was then tested f or primary or secondary archaeological context and the collection's degree of completeness. The criteria for completeness is based on the shapes of size distribution curves as outlined below in the Nature of the Deposits section. Third, o nce the coll ections were deemed adequately complete for the subsequent tests direct comparisons of percentages of assemblage properties that suggest either a trade industry, or internal use for staple processing, were made Most of the focus of these comparisons is p laced on the difference in emphasis on trade rather than attempting to find direct evidence of agriculture practices via lithics. Typical markers of agriculture found in the old world, such as sickle sheen on stone blades from wheat harvest, do not appear or suggest any analogs in Mesoamerican contexts. Also, most Mesoamerican cultivation tools, if similar to manual tools used in the area historically, were likely made

PAGE 197

182 primarily from wood and were therefore perishable (Stuart 2003:1). Differences in tools u tilized for domestic processing of food items may well increase in number and/or in size for feast related activity over regular household production (Adams 1999), although there are still only a few stone tools, such as manos and metates, that tend to be exclusive to food processing tasks in the area. A distinction must also be made between prestige items useful for trade, and common tools that may be intended for either trade or local utilization A s described in C hapter IV obsidian within Mesoameri can sites is invariably found to be more concentrated in elite household and ritual contexts, but is also found among commoner households, and is often in the form of common tools in residences of all classes. Johnson therefore suggests that unfinished obs idian and utility items held a status midway between staple and prestige goods (1996:171). However, since the importation and initial distribution of obsidian was under direct elite control and prioritized for the local production of prestige goods (Johnso n 1996 ; Spence 1981), evidence of obsidian export from Teuchitln sites in the form of prepared cores, utility products bifacial blanks and debitage resulting from these forms also are considered evidence of a wealth trade economy. Criteria for Comparing Staple and Wealth Lithic Economies The Nature of the Deposits As a first step towards the economic analysis and interpretation of the site, the nature of how debitage came to be deposited within the ritual centers must be addressed. Understanding th e nature of the debitage can resolve how completely the debitage/flakes

PAGE 198

183 represent a complete manufacturing assemblage, and whether it is reasonable to pursue certain forms of analysis that require the full complement of debitage from the production process for accurate results. The usefulness of "waste" flakes as expedient tools tends to work strongly against any tendency towards frequent offsite dumping, even within enclosed areas (Whittaker and Kaldahl 2001), and the common use of expedient and/or b yproduct flakes within Mesoamerica is well documented As discussed in Chapter IV, expedient flakes and blades often considered as only cast offs from formal tool production can also be viewed as products which follow a different branch in the chane opra toire ( Julien and Julien 1994:15; Lemonnier 1986; Sellet 1993 ) which stems from more formal tool production ( Bleed 2001:figure 4; Sheets 1975) T herefore, it is not surprising that flakes are found stashed within public architecture, although the consiste nt presence of common flakes within guachimontn ritual architecture is somewhat curious. Logistically, at least three possibilities exist: 1.) Lithic artifact production took place directly within the guachimontn circles, 2.) production took place somewhere nearby, but the remaining flakes were gathered and moved into the ritual centers, or 3.) specific flakes were selected from the original location and brought into the site as tools T hese possibilities are, of course, not mutually exclusive. If t he first or second possibility is true, more detailed statistical analysis can be done to determine more about what was produced, although the second scenario invalidates any analysis that specifically targets the smallest flakes for comparison I f only th e third scenario is true, all further analysis must proceed by treating most flakes as individual expedient tools rather than debitage.

PAGE 199

184 T he three scenarios show differences in frequency distributions of various flake size range categories. Complete pr imary deposits contain exponentially greater counts as size range categories decrease in value, and form a reverse J curve trajectory when size and count are graphed. When lithics are swept and moved together, perhaps from a workshop to a midden or a secon dary cache for expedient flakes, the J curve should still hold true for all but the smallest items, which are usually left behind from the cleanup operation and thus the resulting curve peak is truncated for those smaller pieces (Root 2004). One recommenda tion states that the ratio of flake lengths ranging from 1/8" to 1/4" to those ranging from 1/4" to 1/2" sh ould be greater than 1.48:1 in order to consider the assemblage a primary deposit (Behm 1983:14, as found in Baumler and Davis 2004). A more recent r ecommendation is that flakes smaller than 5.66 mm in length should outnumber those greater in size by a ratio of 4:1 (Root 2004:86,87). The third possibility of "picking and choosing" only those pieces desired as tools and transferring them to the gu achimontn will likely show an entirely different distribution trajectory. Such an operation would involve transporting flakes a few at a time, at most, to the guachimontn and selected flakes would comprise a narrow range of sizes and shapes that fit most usefully and comfortably in the hand (or perhaps with a makeshift haft such as wrapped maguey fibers) as tools, and also provide a useful blade length, position and shape for the task at hand. Such an assemblage would likely not show the expected J curve pattern, and might better resemble a bell curve pattern which peaks towards the larger end of the size scale, around the most frequently useful tool sizes. The most likely of the three possibilities w as thus determined through size category distribut ion plots. One centimeter incremental ranges were defined and unfragmented

PAGE 200

185 piece lengths were tallied within each range. Size ranges were then plotted against counts for each range to determine a distribution pattern, and also determine a "best fit" from e ach of the three scenarios, while watching for potential crossing of patterns or "hybrid" plots to reveal any multi modal transport considerations. The implications of the overall volume of lithics at each site are then discussed, given unequal amoun ts of excavation and estimated projections based on artifact densities of data excavated thus far, before moving on to more specific tests based mainly on percentages and ratios of artifact types discussed below Expectations and Tests for Economic Distin ctions A greater economic reliance on obsidian trade at Llano Grande than at Navajas would result in one or more of the following eight differences in the lithic data. Each expectation corresponds to a specific test, with d ifferent implications in su pport or rejection of its null hypothesis. The total combination of all outcomes, however, involves an exponential increase in possible scenarios regarding the nature of lithic trade at both sites. Therefore, implications of each test are limited in scope to the test in isolation here, and only the actual result s are synthesized. Tests 1 and 2 determine the relative degree of specialized obsidian producer or consumer roles for each site. Tests 3 and 4 pertain to relative emphases on prepared obsidian core s, tests 5 and 6 pertain to utilitarian items, and tests 7 and 8 pertain to relative emph ases on trade oriented products. Navajas is considered the core control site, therefore each of the following tests describe outcomes specifically for Llano Grande and relative to Navajas

PAGE 201

186 1. A lower number of products relative to the amount of debitage: If Llano Grande is more involved in production for export than Navajas, then a much lower proportion of artifacts at Llano Grande should be in the form of formal products relative to the amount of debitage (Magne 1989:20 ; Morrow 1996:357 358; Spence 1987:431 ). Heavy specialization in export should result in a large amount of "ghost" debitage that has no corresponding tools present at the site. Conversely, a lithic import site should have more "orphan" tools that have no corresponding debitage present (Morrow 1996:357 358). Test Procedure: Product counts were tallied and compared to the total number of examined artifacts (products and debitage) at each site. Retouched bypr oduct flakes and blades, in this case, were also counted as debitage. Since items categorized as chunks are indistinguishable from natural stone fragments that are abundant at Llano Grande (Beekman 2001:4,12), they were not included in the total artifact c ounts for either site R esults are presented as product to total artifact count ratios, and as product percentages of the total number of artifacts. The null hypothesis is supported by an equivalent or higher percentage of products to debitage at Llano Gra nde. Support for the null hypothesis of this test could result from manufacture of lithic items for both internal use and trade at a location away from the circles, underutilization of Llano Grande's lithic source, and/or thriving lithic production within or near Navajas' circles

PAGE 202

187 2. A higher ratio of obsidian flakes and debitage to ceramic sherds: If Llano Grande site did specialize in lithic production, evidence of other activity is expected to diminish, which tends to be best represented by ceramic waste (J ohnson 1996:170 172; Spence 1987:430). In the context of a guachimontn, ceramic vessels and bowls are especially indicative of staple ritual activity (feasting) (Johns 2014; Tyndall and Beekman 2007:174). Test Procedure: Ceramic sherd totals for Llano Grande were taken from Tyndall and Beekman (2007:154,155) and totals for Navajas were taken from Johns (2014: 83 ) The ratio of lithic debitage pieces to ceramic sherd totals was then calculated for each site, compared, and presented as both ratios and eq uivalent indices found by dividing lithic totals by ceramic totals. Support of the null hypothesis is indicated by an equivalent or lower debitage to sherd index at Llano Grande, in which case Llano Grande may have continued the core based staple rituals, possibly in conjunction with increased trade. 3. A lower percentage of prepared cores: If Llano Grande functioned as a quarry site for trade, it would most likely have continued the Mesoamerican Late Formative and Early Classic pattern of emphasis on export of prepared polyhedral macrocores. M ost distributed cores, whether internal (Beekman 1996a; Esparza 2003; Soto 1982, 1990; Spence et al. 2002) or external (De Len 2003; Johnson 1996; Parry 1987; Spence 1981 ; among others) to the area inhabited by the Teuc hitln culture were decorticated to a large extent T herefore, exported Teuchitln cores from sites near

PAGE 203

188 sources were uniformly prepared near the source in some way. Llano Grande is an obvious candidate for showing evidence of source preparation, but Nava jas may also have served in the same capacity, to a large extent, with easy access to a nearby mine. Therefore, both sites may have engaged in wealth finance to different degrees. Morrow's concept of "ghost" debitage may also be applied to cores. The presence of a relatively small number of polyhedral macrocores or cores lacking cortex relative to the amount of debitage indicates a preparation site for a core export economy. A large percentage of prepared cores would be more indicative of local utiliza tion of lithic resources T here is little precedent for the distribution of u nprepared cores within Mesoamerica in the periods under study, and are more likely used for expedient internal tools. The majority of recognizable cores will be at least partially worked (as indicated by negative bulbs of percussion and impact rippling), although completely unworked obsidian pieces of adequate size to form products may also be interpreted as unprepared cores, if found in isolated contexts. Test Procedure: Relative proportions of cores were compared and presented as a percentage of total artifact counts within each site. Both the total amount of cores and categories of prepared and partially decorticated casual cores were compared U nworked obsidian was not consider ed, since it is ubiquitous at Llano Grande and therefore any intent for use could not be confidently determined. Support of the null hypothesis in this case is indicated by an equal or higher proportion of prepared cores at Llano Grande, which would indicate preparation primarily for internal use. A presence of casual cores with partial cortex does not

PAGE 204

189 affect indications of trade, but does affect the interpretation of cortical flakes, as described below. 4. An increase in the amount of cortex on flakes : This expectation is the complementary evidence for prepared core export L ithic manufacture removes the vast majority of cortex very early in the reduction process (Clarkson 2008:288; Magne 1989:17; Mauldin and Amick 1989:70 71). Debitage from core prepar ation would primarily involve cortex removal, and should result in higher proportions of flakes with high percentages of cortex coverage, than debitage from later stages of production. Test Procedure: Since cortex percentages were recorded as quartile cod es a measure of association for ordinal data was chosen for this test. Cortex proportion frequency distributions of flakes a nd blades were compared through a Pearson Chi Square test. Upon discovery of a significant association at the p < .05 level Kendal l's t a u c was used to determine the strength and direction of the association Additionally, the mean difference in cortex coverage between sites was determined for all items with partial cortex (i.e., excluding those coded as "no cortex" and "full cortex ") to show the overall difference between sites. The null hypothesis is supported if either a lack of significant association or a significant positive association, is found between sites and amounts of cortex The null hypothesis is also supported if an equal or lesser mean percentage of cortex coverage per piece is found for Llano Grande. Conversely, the null hypothesis is

PAGE 205

190 rejected if a significant negative association is discovered between sites and amounts of cortex, and a greater mean percentage of cortex per piece is discovered at Llano Grande Failure to reject the null hypothesis here would indicate that unaltered cores were prepared within or near the Navajas guachimontn at least as frequently as Llano Grande. Bar charts of the cortex d istributions are then compared to help determine any differences in the nature of production between the two sites. The differences in "full cortex" and "no cortex" codes are also compared to more clearly show the magnitude of differences in decortication between sites. 5. A lower percentage of specialized tools for staple processing (e.g. manos, metates molcajetes 2 mortars and pestles) F ollowing the logic behind previous ceramic studies in Navajas (Johns 2014; Tyndall and Beekman 2007), the consistent inc lusion of staple tools in ritual contexts such as guachimontones links staple processing to public ritual activity, and therefore provides an additional form of evidence for feasting rituals related to a staple economy A lower percentage of staple tools m ay therefore indicate reduced emphasis on staple related ritual in a consumer context. Test Procedure: Percentages of staple processing forms relative to the total number of recovered products at each site were compared and presented. T he null hypothesis is supported by an equal or greater percentage of specialized staple tools at Llano 2 a mo l cajete is a tripod supported grater bowl used for gri nding spices.

PAGE 206

191 Grande, which would indicate that staple related rituals were likely continued and emphasized to an equal or greater degree than at Navajas. 6. A lower percentage of debitage associated with to polyhedral blades and blade based products : Even early stage polyhedral macroblades are produced by a very distinctive, methodical and formal process that creates very identifiable morphologies in several flake forms (Clark and Bryant 1 987; Hirth and Andrews 2002:2 3) Y et, as Johnson observed, blade production is "an apparent enigma, the specialized production of a generalized tool" (1996:167). Fine polyhedral blades and macroblades are forms of very flexible utility by nature. Site con texts and use wear evidence have shown that blades were used for a wide variety of common tasks, such as shaving and food preparation, and mainly within domestic areas of Mesoamerican core population centers (Johnson 1996:166, Spence 1981:780). Blade s and blade byproducts are therefore likely common at both Llano Grande and Navajas for internal utilization. For trade purposes, much more refined core blade manufacture was commonly known and practiced through all other areas of Mesoamerica, so there wou ld not have been any demand for West Mexican macroblade production. Also, if the above expectations regarding the trade of cores for later production by the consumer holds true, material for blades was probably not traded in its final form, but rather as p repared macrocores. Test Procedure: Debitage pieces which are primarily associated with the polyhedral blade industry (blade core fragments, core flakes, macroblades, 1st and 2nd series

PAGE 207

192 blades, platform correction flakes and core correction flakes) were t allied, and type counts were aggregated to determine percentages of blade specific items relative to the total count of debitage pieces at each site. T he null hypothesis is supported by an equal or greater percentage of blade based types at Llano Gra nde, indicating greater emphasis on a polyhedral blade industry than Navajas. An increased polyhedral industry at Llano Grande may mean production of blades for internal use as utilitarian or staple processing items, and also possibly a return to (or conti nuation of) the Middle Formative practice of direct trade of blades and blade products 7. A higher percentage of bifaces : Despite a lack of evidence for the production of bifacial blanks in the La Venta Corridor described in Chapter IV remote areas may di ffer in their exact function, depending on internal cultural and environmental differences between them, and the political and economic differences between the peripheral groups with which each semi peripheral group interacts (Stein 1996:36, 2014:57 58). T herefore, given the popularity of bifaces as a regional export item s for Teotihuacn, Llano Grande and/or Navajas may still have produced them for trade. Blanks would be particularly advantageous to traders in the hills around the Tequila Valley s since th ey are more robust for transport compared to finished products (Spence 1981:771), much lighter than prepared cores, and require less production labor than cores at the destination. Theoretically, the relative percentage of bifaces and other specific types could be derived from debitage attribute statistics in much the same manner as the procedure

PAGE 208

193 followed here for the blade industry, but debitage from bifacial blanks is much more difficult to distinguish from expedient flake production or early stage production from other industries. This is especially true when multiple technologies are known or suspected to be present within a single assemblage, which is most often the case in the field (Ahler 1989:206; Carr and Bradbury 2001; Tomka 1989:137 ). Carr and Bradbury (2001) and Ahler (1989) have developed debitage attribute based formulas from their own experimental data to address the mixed technology type identification issue. However, like most experimental debitage studies, reliable use of these formul as across different datasets or in the field is still untested. Therefore, only counts of identifiable products which remain from the production process can be used as indicators of the emphasis on each type produced at both sites. Using the Feature 83 workshop as a general pattern for production deposit expectations ( Soto 1990), it also appears that even the largest Teuchitln workshop did not stockpile products for later distribution on site, which left only broken and unfinished products. Therefore incomplete and flawed items may be the only clues to indicate each site's relative degree of biface production Test Procedure: Unrefined and failed bifaces were tallied for each site, and percentages relative to the total number of tools at each site were compared. Support of the null hypothesis is determined by a lower relative percentage of bifaces at Llano Grande than at Navajas, which would indicate that bifacial blanks were less utilized as a form for trade than at Navajas.

PAGE 209

194 8. An increase in emphas is on production related to esoteric items (eccentrics) unique to the Teuchitln culture (i.e. laminar jewelry, pikes and rods) : Trade in items considered unique and rare to remote locations is an exception to the usual core trade pattern. Such items are especially useful to outsiders as status markers and also easier for external elites to control when the method of manufacture is not widely known or easily duplicated (Beekman 2000:386 ; Earle 1991:7). E xternal refined pike distribution is unknown, bu t rough pikes have been located in the Sayula Basin, and laminar items are found throughout West Mexico (Reveles 2005:352). Production debitage for laminar jewelry has previously only been found at one site within the La Venta corridor (Beekman 1996a:816,8 46 ) and possibly at the San Juan de los Arcos quarry so uth of the core population zone ( Clark and Weigand 2009) It is considered a probable prestige item (Beekman 1996a:816,846 ) and also appears to be the result of one or more unusual and spatially limit ed techniques ( Clark and Weigand 2009 ; Long 1966 ) Since the discovery of laminar items has been limited to West Mexico and thus far the evidence for the production of these items has been further limited to the Tequila Valleys l aminar production a nd use is less understood than o ther items previously discussed. Because of this lack of understanding, and because these items are also especially pertinent to questions of wealth and staple finance, the bulk of effort in the analysis is spent on identify ing markers of manufacture for laminar technology

PAGE 210

195 Refined pikes fall into the same category However, much of refined pike production would be revealed in the retouch pattern, and flakes of the required size are missing from the field data. On e basic method of determining relative degrees of laminar production is the presence of final stage laminar debitage itself, which may be identified as flakes and blades with only one dorsal scar and with very low (0 25%) cortex. (Thickness would be a bet ter identifier, but unfortunately thickness data were not recorded.) The se items can be differentiated from late and final stage biface debitage since even small biface flakes usually have multiple dorsal flake scars. M ost or all of the larger laminar piec es were likely utilized, but fragments and pieces of unwanted size or shape w ould have been left behind, which may be numerous in the case of a specialty workshop for such items as discovered by Beekman in the La Venta corridor (1996a:790, 795 799). Howeve r, such items may be sparse in secondary contexts where laminar production was done in much lower volumes. Due to their fragile nature, many remaining pieces may be quite small due to fragmentation, and may not have been transferred to their final location s. Although multiple methods have been proposed for the production of laminar blanks and more than one method may well have been practiced throughout the valleys, the main distinction between laminar jewelry and most other product types is that the je welry has no dorsal flake scars, so reduction must work towards that end. The usual expectation is that the number of dorsal scars per flake, and therefore also the density of scars, increases as reduction progresses (Andrefsky 2005:106; Magne 2001; Philli ps 2011:71 72; Shott 1994:80). Logically, flakes must also progressively

PAGE 211

196 decrease in size through the reduction process. Magne has concluded that dorsal scar counts, along with counts of platform scars, statistically appear to be the best predictors of the degree of reduction (2001:23) H owever, the scar eliminating goal of laminar jewelry is at odds with this process. Even in mixed typological and technological contexts the presence of laminar obsidian manufacture should at least weaken the negative corres pondence betwee n size and flake or blade scar density. In some c ases with relatively high percentage of laminar jewelry, a scar density/size correlation might become insignificant or even change to a positive correlation Just one other, far better k nown technology in the area may create a positive scar density to size correlation, namely the polyhedral core blade technique. However, core blade debitage also leaves more recognizable morphological patterns (e.g. failed blades, core correction and facia l correction flakes) in the debitage to reveal its presence. Using either Long's or Clark's proposed methods (aside from Clark's observed bulb of percussion removal method), isolated laminar production may only leave behind some core preparation, fail ed laminar flakes, and partial "scrap" flakes from forming specific shapes. Failures from both techniques may be indistinguishable from thin platform preparation flakes, in that they can be uneven in thickness, even but thicker than desired, a wedge shape formed by failure to pass completely through the core, or fractured upon removal L ong's "sausage" technique also would show forms of debitage diagnostic of core blade production activity and possibly one or more cylindrical cores of variable lengths. In e ither case, if laminar production was an exclusive or overwhelmingly dominant form of production, the presence of small

PAGE 212

197 fragmented and whole thin flake s along with broken retouched pieces should be readily apparent, as discovered by Beekman in the La Venta corridor (1996a:795) B ut for the large majority of lithic collections, the likelihood of mixed production technologies within the same assemblage is quite high (Carr and Bradbury 2004 ; Larson 2004 ; Root 2004 ; Tomka 1989), and despite some earlier experim ental progress with multivariable regression and discriminant analysis techniques using a limited number of product forms ( Ahler 1989 ; Bradbury and Carr 2001) there is currently no verified method to determine the proportions of each industry (blade, lamin ar jewelry, bifacial core, etc.) within a debitage collection H owever, the strengths and directions of any correlations between scar counts and flake sizes should directly relate to relative proportions of laminar vs. conventional flake production within each site The strength and slope of negative correlations is also affected by other factors, such as the combination of types of other products being made, the material being worked, and the combined techniques of individual knappers (Andrefsky 2005 :107). The correlation between size and scar density should not, however, become insignificant or turn to a positive sloping trend with any commonly known product or method. Test Procedure: The term "flake" is used here in a generic sense to mean any pie ces with flake or blade dimensions and characteristics. Both flakes and blades are tested in this procedure since laminar flake dimensions can't be assumed.

PAGE 213

198 Scatter plots were produced of dorsal scar density to size relations as a preliminary visual check, in order to determine sloping trends and possible grouping patterns in the relationship between the two variables. One chart per size variable (length, width and the product of length and width as a rough approximation of area) was created for each site. Since grouping patterns may either be isolated to one flake type or occur across the entire dataset, shaping/finishing flakes and reduction flakes are included as separate groups within each chart. The same scar density and size parameter relationshi ps were then checked for statistically significant correlations by running Navajas and Llano Grande P earson's r figures were also calculated for the experimental data to verify a significantly negative correlation in conventional products. Evidence for laminar production at either site is manifested in a lack of significant correlation between size and scar density, or a reversal to a positive correlation. If both sites we re strongly engaged in laminar production then a relatively greater emphasis on laminar production at one site will produce a statistically stronger positive correlation at that site. Conversely, the null hypothesis is supported by a statistically signific ant negative size to scar density correlation for all size parameters at both sites, more strongly positive correlations at Navajas, or a lack of a statistically significant correlation at Navajas if Llano Grande shows any result other than a significant p ositive correlation. Support of the null hypothesis in this case would indicate ritual items were not shown to be a greater part of the trade system at Llano Grande than at Navajas, and therefore Navajas shows a likely greater emphasis on wealth economy re presented by these items and their exchange with the Sayula Basin

PAGE 214

199 sites, and possibly others. The null hypothesis is also supported if results are mixed between different size variables. Counts of laminar waste flakes are also used as supplementary ev idence to assist in determining the relative proportions of laminar debitage. Laminar waste flakes at each site were counted and compared as relative percentages. The null hypothesis is rejected if a higher percentage of laminar items is observed for Llano Grande relative to Navajas. Import Considerations One obvious additional criterion was considered, but had to be eliminated due to a lack of data from the excavations at either guachimontn site: t he presence of incoming "exotic" elite items from o utside areas would show evidence of reciprocity in a trade relation. However, obvious imports were not present in either site, and may never be found because they are often very perishable. Perishable trade goods are often staple items, but many are usuall y considered wealth items. Examples of perishable wealth goods may include salt, fe athers or certain animal hides (Nassaney 1996). Absence of evidence of foreign goods, therefore, is not considered a strong argument against the existence of a wealth trade economy Some less perishable items, such as foreign minerals or seashells, may later be discovered within other guachimontn satellite buildings at Llano Grande or in other contexts, such as nearby shaft tombs or elite occupational units with additional e xcavation

PAGE 215

200 Outcome Analysis Navajas has thus far shown a greater degree of affiliation to the core through its architectural configuration and staple related artifacts Therefore it is expected that Navajas may show a greater relative degree of sim ilarity to the more utilitarian and staple oriented lithic industries discovered in Los Guachimontones Circle II ( Esparza 2003) and the Feature 83 Workshop (Soto de Arechavaleta 1982, 2000) whereas Llano Grande may show relatively greater emphasis on one or more trade economies identified in Chapter IV namely, eccentric bifacial blank and/or prepared raw material tr ade T he null hypothesis therefore, is supported if no differences in trade or staple economies are seen between the sites, if only Navajas shows greater emphasis on one or more of the three trade economies, or if the results are mixed (i.e., Llano Grande and Navajas both show greater emphasis on different trade economies) O therwise, if the analysis indicates a higher relative emphasis on ob sidian trade only at Llano Grande or of a wealth economy at Navajas, the null hypothesis will be rejected It is tempting to speculate on a particular role or status of each site within the social structure, based on the outcome of these tests H owe ver, given that these tests involv e only relative differences between sites, we can currently only determine the state of the economy of each site relative to the other. Llano Grande and Navajas may well both have "primarily" operated under the same econom ic base; Llano Grande still had fertile land available nearby, and Navajas had a large, high quality source of obsidian in close proximity, as well as possible trading partners in the Sayula Basin. However, there is no criteria established for the amount o f staple or wealth production that warrants either

PAGE 216

201 categorical distinction. These data are also currently limited to two specific sites, which happen to be in certain positions within the social structure. Additional sites will need to be included for test ing in the future in order to generalize whether the same relative differences may apply more generally to populations within the core and semi peripheral zones. Conclusion Since this is an initial probing study for the data at both sites, aside fr om a couple of more involved statistical comparisons, most of the above tests involve only descriptive statistics. The descriptive information may, however, actually be the most useful choice for an initial interpretation of the sites under test. Multiple lines of evidence are, of course, always recommended for verification of overall results, which is certainly no less true in lithic studies (Magne 2001:23 24). But when also considered as separate clues about different aspects of the theoretically informed questions, unanticipated details about the nature of the sites beyond the basic statistics can emerge, which provide the best chance for generating additional questions as a basis for moving forward. Therefore, although the obvious goal of Chapter VI is t o seek answers to the overarching theoretical questions through these tests, the more important goal in this first detailed look at the data may be to discover better questions.

PAGE 217

202 CHAPTER VI RESULTS This chapter provides results of the tests defined in C ha pter V along with discussion of the significance of each result. Implications of these results are summarized and synthesized in Chapter VII Organization and Review of the Test Criteria The results discussion begins with an overview and comparison of morphological type proportions in the lithic collections from Navajas and Llano Grande, with those published from other sites reviewed in Chapter IV The background comparison is intended to provide an indication of relative adherence of each site under test to economic patterns of other core and semi peripheral sites. The nature of origin (primary or secondary context) and the degree to which each debitage collection from the sites of Llano Grande and Navajas may represent complete production activity, as determined through piece length sample distributions, is then discussed. Following confirmation of adequately complete assemblages for an analysis of production related aspects of the data, all of the following tests are run to determine the relative em phases on the economic bases in question at each site:

PAGE 218

203 1. The number of formal products relative to the number of informal flakes, blades and debitage to determine whether either site represents a consumer or producer context. 2. The ratio of obsidian flakes, blades and debitage to ceramic sherds to indicate the degree of specialization in lithics relative to other activities at each site. 3. The percentage of prepared cores relative to debitage to determine the relative likely degree of prepared core export if accompanied by a larger quantity of debitage, or internal utilization if accompanied by a lesser degree of debitage. 4. The amount of dorsal and platform cortex on flakes as an additional indicator of internal core preparation. 5. The percentage of specialized lithic tools for staple processing relative to the total number of samples at each site to indicate the degree to which each site utilized lithics for feasting rituals. 6. The percentage of debitage associated with polyhedral blade manufacture and blade bas ed products relative to the total number of samples at each site to indicate the degree of emphasis on utility items most often produced for internal use. 7. The percentage of bifaces relative to the total number of samples at each site, to indicate the degr ee of emphasis on items commonly prepared for trade. 8. The relative emphasis on production related to eccentrics unique to the Teuchitln culture. The only testable item in this case is laminar flake production, which is examined through relative degrees of disruption of the normal flake scar density size distribution pattern.

PAGE 219

204 Lithic Type Proportion Comparison to Background Contexts The lithic background contexts discussed in Chapter IV are reviewed in Table 6.1 below, and Table 6. 2 shows relative per centages of lithic types for sites compared in C hapter IV along with the same categories from the sites under study. The table includes the data from table 4.5, with the addition of data from the sites of Navajas and Llano Grande. Table 6.1: Summary of compared background locations Site Descriptions La Venta Corridor A semi peripheral zone east of the core The collection is from combined area wide survey and multiple site excavations Lithics from the entire zone are compared from all periods. Surface lithics appear contemporaneous with Teuchitln occupation, but surface eccentrics are underrepresented compared to excavated contexts. Staple tools and debitage may be elevated by the excavation of household and midden contexts, respectively. Los Guachim ontones Ceremonial Center A surface survey of t he entire center, with sample pit data from the two largest circles and the ballcourt between them. Los Guachimontones Fe a ture 83 Workshop A comprehensive excavation of t he primary core workshop, which s pans 0.6 hectare. Table 6.2: Inter site comparison of lithic proportions by percentage relative to total item counts, including sites under study. Location Llano Grande (n=962) Navajas (n= 356 ) La Venta Corridor (n=4,863) Los Guach. Public C n tr (n=16,5 80) Los Guach. Workshop (n=99,863) Blades 3.8 7.6 6.5 16.6 49.1 Flakes 65.3 61.1 43.5 72.8 33.4 Points 0.3 2.6 1.5 0.3 n/a Blanks/Preforms 0.3 0.7 0.1 0.3 n/a Cores 3.9 0.7 0.5 1.0 0.0 Staple tools 0.0 0.2 0.2 0.0 0.0

PAGE 220

205 Other tools 1 .4 11.2 1.3 6.9 n/a Eccentrics 0.1 1.9 0.9 0.0 0.0 Debitage 24.8 13.8 45.2 2.0 17.4 Unknown 0.0 0.0 0.2 0.0 0.0 Total 100 100 100 100 100 The samples used in this comparison for Llano Grande and Navajas include only the verified artifacts ( all types excluding chunk s ) analyzed for this thesis, not excavated sample totals. Again, the aim of this section is to determine similarities between areas, with Navajas proposed to align with Los Guachimontones as a core affiliated site, and Llan o Grande proposed to align with the La Venta Corridor as a semi peripheral site However, several of the differences in percentages do not align as expected. Navajas is closer to the semi peripheral La Venta Corridor in the percentage of polyhedral blades. Los Guachimontones does, expectedly, show a relatively high percentage of blades, and Llano Grande shows even fewer blades than La Venta. Points are much lower in proportion at Los Guachimont o nes relative to the La Venta Corridor. An increase in poi nts for military purposes may be expected within a semi peripheral zone with exposure to external groups. However, the heavily fortified site of Llano Grande contains a site relative proportion of points as low as Los Guachimont o nes. If the walls did invol ve a centrally directed construction effort, then perhaps any anticipated threat from external groups at the time of construction either did not materialize or greatly diminished before the site was abandoned. Points may also be concentrated elsewhere on s ite at Llano Grande. In any case, the reason for Llano Grande's low point percentage also requires further investigation. Points do not appear to have been primarily made from source or workshop blanks in any area, since relatively few blanks are reported for any site. However, blanks are most concentrated at Navajas

PAGE 221

206 where the highest percentage of points is also located. Those blanks may, however, only be early stages of production or failures since they are still only a fraction of the number of points fo und on site The percentage of eccentrics is also higher at Navajas than at any other site, even though they are apparently non existent at the Los Guachimont o nes ritual center S mall percentages exist at Llano Grande and the La Venta Corridor. Much like other product types discussed so far, however, a more important factor in determining the relative amount of export of such items is the evidence for manufacture relative to the number of items which are present. Flakes are the most frequent item s and also show differences in percentages betwee n study areas, but knapping produces very large quantities of flakes and the percentage of secondary flakes which were actually utilized at each site is unknown without use wear and/or residue studies, and t herefore cannot yet be compared. Also, relevant data regarding cores cannot be discussed without knowing specific types (casual, decorticated casual, polyhedral, etc.) which were often not provided in the literature. The majority of types, including blades, points, cores, staples and eccentrics all show closer percentages between Navajas and the La Venta Corridor than between any other two locations. some of these similarities may be due to greater regional affinity between the two eastern areas, than between core and semi peripheral zones. For example, the increased presence of points in both areas may be due to increased hostilities from external groups to the east O ther similarities, however, appear coincidental. The presence of staple processing i tems, for example, would be common in household areas within the La Venta corridor, and may have no association with ritual areas.

PAGE 222

207 The results of this comparison suggest that some social factors other than core and semi peripheral area relations may, in certain circumstances, have greater impact on economic similarities and differences. However, although it is an important consideration, this is a different question which runs tangential to the present purpose of this thesis. In the current comparison such factors may only serve to mask the impact which core to semi periphery area relations may actually have on the economy. To observe differences specific to the relations under test, it may prove more beneficial to look at specific sites from different areas in relative isolation. General observations on the Collections Under Test Overall, each collection at Llano Grande and Navajas gives the initial impression of an expedient flake industry. Flakes at both sites are very wide, and in some areas they average wider than they are long U nusually wide dimensions may be due to expedient flaking, or possibly a technique which intentionally favors more rounded dimensions. But the majority of obsidian pieces lack any defining features or patterns, even after setting aside the over 40 percent of the data in Llano Grande which are classified as chunks R elatively low proportions of polyhedral blade related debitage exists at either site; blade industries were present, but do not appear to have been emphasi zed. The few cores remaining at each site are nearly all casual; no polyhedral cores and only one very small macrocore were located. Edge retouch also appears on 2.8 percent of flakes and blades at Llano Grande, and 6.2 percent of the same categories at Na vajas. If left to casual observation, one may conclude that nearly all lithic activity associated with the

PAGE 223

208 guachimont o nes is highly informal, and therefore used primarily for internal purposes. Internal use could include craft work for ritual items such as clothing, decoration, or food processing. However, what appears to be strictly expedient in nature can easily represent secondary byproduct utilization from a more formal industry, such as bifacial or unifacial tool production A closer look is therefore required to determine the nature of each site's lithic industry. Much like the neighboring Sayula Basin (Reveles 2005:351), bulbs of percussion are obvious through all recovered size grades at both sites, indicating that percussion was used until the final stages of production S ome blades and flakes do appear to have been produced by pressure flaking, with very shallow and sometimes nearly undetectable bulbs of percussion The most obvious of these are the laminar jewelry flakes. However, even when in cluding the jewelry flakes, pressure flakes are relatively uncommon. This created an additional challenge in the determination of Clark's division for "series" blades as well as the corresponding flake categories. The switch to pressure flaking techniques is a usual hallmark of Clark's series blade production (Hirth and Andrews 2002:2). It also illustrates the earlier point that local variations often occur within what otherwise appear to be equivalent lithic industries. Overall Quantities and Site Covera ge Differences Of the 1,636 examined pieces from Llano Grande, 674 items (41%) were determined to be of questionable artifact status and are thus categorized as chunks T he Navajas guachimontn contained substantially less natural stone; 128 out of 484 analyzed

PAGE 224

209 obsidian items (26%) from Navajas are chunks Y et even with chunks excluded, the Llano Grande debitage sample is approximately 2.7 times larger than that of Navajas But the overall count difference s discussed thus far only involve the s amples examined for this study A s discussed in C hapter V although excavation at Navajas Circle 5 was nearly complete, the Llano Grande guachimontn was only partly excavated. Site totals can be estimated, however, based on average artifact densities to a chieve some idea of the magnitude of artifact counts within each guachimontn In most excavated applications of artifact density volum etric densities are used in contexts which span time periods long enough to use layer depth as a proxy for time. I n the current context, however, we have single occupations over 100 to 150 year periods The guachimontn contexts also contain artificial "strata" composed of compacted earth floors over artificial subfloor material (Beekman 200 1 2007). Further, collapse and post occupation layers have shown no evidence of post period occupation at either site, so items in the more recent layers were either items stored higher up in each building, or the product of post depositional formation processes which vertically di splaced the artifacts E xcavated volumes in this case has little to do with occupation periods, and will therefore produce misleading differences between sites. Therefore, area rather than volume densities are used for the estimated site totals T he total excavation area at Navajas is 706 m 2 and Llano Grande 's excavation covered a total area of just 104 m 2 or approximately 15 percent of Navajas' coverage. Based on preliminary analysis data recorded by Beekman on entire recovered datasets from each s ite, the total number of verified lithic artifacts is 1,747 for Navajas and 2,980 for Llano Grande. The average artifact density at Llano Grande is therefore 28.6

PAGE 225

210 pieces/m 2 compared to just 2.47 pieces/ m 2 at Navajas W hen Llano Grande's density average i s projected over the entire guachimontn it provides a comparable estimate of approximately 34,165 pieces. Since lithic counts in different areas of the Llano Grande guachimontn vary considerably, ranging from just over 100 to over 1200 pieces near the activity floors of each building, continued excavation at Llano Grande will most likely yield counts that vary considerably from this projection. But even in the very unlikely scenario where further excavation fails to yield any additional lithic arti facts at the Llano Grande circle, at minimum Llano Grande's total artifact count remains nearly twice that of Navajas, which still suggests greater emphasis on lithic activity in or near the Llano Grande circle Even if none of the production happened dire ctly on the circle, the volume of production within the vicinity of the guachimontn may well be definable as a workshop, although Clark's workshop criterion of overproduction beyond the needs of the site population cannot be resolved within the scope of t his thesis The Nature of the Deposits The count distribution patterns for piece lengths are shown in figures 6. 1 and 6. 2. Both Llano Grande and Navajas show a size distribution pattern of a negative exponential curve that would be produced by a fu ll complement of lithic production debitage, minus the smallest pieces. The sites show very similar patterns of divergence from a complete debitage distribution in that in both cases, smaller items begin to diverge from what would be expected if all pieces were present, and sample counts of the smallest items

PAGE 226

211 drop off quite drastically. The only differences in patterns between sites are the size thresholds where both of these changes occur A t Navajas, the curve begins to attenuate at around 3 cm, whereas L lano Grande begins to diverge near the 2 cm mark L ikewise, the Navajas c urve drop s steeply at 2 cm and the Llano Grande c urve drop s at around 1 cm A dditionally, there are only 15 flakes reported to be smaller than 1 cm at Llano Grande, but none at Navaja s. As discussed in Chapter V, b oth curves are likely a result of Count Count Size (cm) Si ze (cm) Figure 6.1 Llano Grande Figure 6.2 Navajas flake length d istribution flake length d istribution secondary depositions, since primary depositions of even ear ly percussion stages of a lithic production process consistently include a majority of flakes under 1 cm, and m oving debitage typically results in most or all of the very smallest pieces left behind at the original location This is primarily due to sweeping tool inefficiency and the embedding of smaller items in dirt or grass floors ( Root 2004 ) However, retention of the reverse J curve pattern in the larger pieces is likely the result of transfers of entire debitage piles into the guachimontones, perhaps as an

PAGE 227

212 expedient way to gain convenient access to an assortment of edges rather than picking and choosing large numbers of flakes for secondary use I n both sites, it appear s that the actual production had likely occurred in some nearby location, and the resulting flakes were gathered together and moved into the guachimontn buildings. Although the loss of the smallest flakes and primary context is not the most ideal situatio n for a production study, the fact that flakes appear to have been moved together does allow for closer examination and interpretation of all aspects of the nature of the debitage relevant to the study, therefore all eight tests can validly proceed. Lithi c Industry Tests Test 1: Formal Products vs. Debitage and Expedient Tools Product counts are very low in the sample data from both sites, with the exception of scrapers at Navajas, and polyhedral blades at both sites. Only one item in the ground sto ne category appears in the Llano Grande collection, which is a large symmetrical oval rock approximately 20 cm in length, that shows evidence of grinding activity in the form of a groove carved around its circumference, perpendicular to the major axis near one end. The groove resembles those used on axe heads for hafting, but the stone is very round and blunt on the opposite end and does not readily suggest any known function. Despite their low numbers, there is a disparity in relative percentages of products between sites (Table 6. 3 ), which crosses all product types except laminar flakes.

PAGE 228

213 Table 6. 3 : Percentages of formal product counts relative to total analyzed sample (products and debitage) n Bifaces Points Bifaces Other Scrapers Blades Ground stone Pikes Laminar Jewelry Total Navajas 356 2.6% 1.4% 10.0% 7.6% 0.2% 1.2% 0.7% 23.7% Llano Grande 962 0.3% 0.3% 0.4% 3.8% 0.0% 0.1% 0.0% 4.9% The consistently greater product percentages at Navajas combined with the comparatively sparse quantity of debitage at the site further suggests a greater emphasis on product consumer activity, and conversely, more of a lithic producer role for Llano Grande. The null hypothesis is therefore rejected for this test. Test 2: Debitage to Ceramic Sherd Ratios Debitage (including expedient and secondary flakes and blades) and sherd quantity comparisons also utilize Beekman's verified site totals, as well as Tyndall and Beekman's (2007:164) sherd counts for Llano Grande and Johns' (2014:83) counts f or Navajas. Again, differences in excavation coverage do not permit a direct comparison of counts across sites, but only within each site. Therefore, only the site ratios between artifact types will be directly compared. Table 6. 4 : Debitage and s herd t ota l c omparison Llano Grande Navajas Debitage 2,980 1,747 Sherds 1,525 9,475 Ratio (debitage:sherds) 2.0:1 1:5.4

PAGE 229

214 It appears that Llano Grande emphasized lithics to a much greater extent than ceramic related activity as compared to Navajas, but this does not mean that ceramic related activity by itself was actually lower at Llano Grande. The ratio between sherds and lithics is only a relative comparison between two independent industries. The average artifact density for sherds (1525 sherds over 104 m 2 or 14.66 sherds/m 2 ) can also provide an estimated count for an equivalent excavated area to Navajas, and calculates to an estimated 10,325 pieces, or actually a 9 percent increase over the count at Navajas. As with the debitage, sherd counts vary considerably from building to building ( Tyndall and Beekman 2007:Table 4.2 ; Johns 2014:91). Also, as with the debitage, the degree of artifact fragmentation and the specific forms of ceramic utilization may have varied between sites. Therefore this es timate is quite tentative H owever, given the information we have to date, it appears that although lithic related activity was more prominent in and around the ritual structure at Llano Grande than Navajas, ceramic related activity was at least as intensi vely pursued, and may have involved the same feasting rituals (or some modified versions adapted to the area) as Navajas T herefore, lithic processing could not have been the sole industry at Llano Grande, but it did rise to a relative prominence as other activities continued apparently undaunted within the circle T herefore, the null hypothesis in this case is only partly rejected Lithic activity did increase relative to other activities, but there is no evidence that other activities diminished.

PAGE 230

215 Test 3: Core Totals Only two polyhedral core s w ere found between the two sites, one at each site All other cores at both sites are casual items with substantial amounts of cortex (mostly ranging from 25 to 75 percent ). Navajas contained only three casual cores, or 0.8 percent of the frequency of all debitage pieces, and Llano Grande contained thirty seven (3.8 percent of its debitage count). Chi square and T test statistics could not be run on core quartile values, due to the extremely low core count, but it is nevertheless clear that these cores do not show an obvious indication of trade Table 6. 5 Core quantities as a percentage of debitage at each site Type/Site n polyhedral blade core casual core Navajas 356 0.3% 0.8% Llano Grande 962 0. 1 % 3.8% Llano Grande's core collection demonstrates some lithic production from the initial part of the reduction sequence within or near a guachimontn, which has not been seen elsewhere in the Tequila Valley s However, the presence and expedient nature of the cores within the circle shows a likelihood of production for local use and possibly some form of product trade, rather than trade in the form of prepared cores T he null hypothesis is therefore supported for this test.

PAGE 231

216 Test 4: Amounts of Dorsal and Platform Cortex on Flakes and Blades As mentioned in C hapter V proportions of cortex coverage were recorded by quartile, but two categories define extreme states of no detected dorsal or platform cortex, and full cortex over these areas S ince thes e properties are absolute states and do not scale with the quartile ranges they are included here as separate comparisons relative to their respective site's totals in Table 6. 6 Table 6. 6 : Site percentage s of flakes and blades which possess full or no c ortex Llano Grande (n= 649 ) Navajas (n= 262 ) No cortex 50.2% 68.3% Full Cortex 11.2% 2.3% A decided majority of Navajas' flake and blade debitage bears no cortex, and very little of it (2.3%) bears full cortex T hese figures show further affinity of Navajas with previously discussed areas within the T equila valleys which reflect the prepared core import model, in that relatively little cortex is seen in those sites I n contrast, roughly half of Llano Grande's debitage is cortex free. Although Llano Grande's proportion of full cortex pieces is not exceptionally high, it is nearly five times that of Navajas. Llano Grande appears to reflect a greater degree of decortication and early stage production, although not such a substantial number of cortical pieces to suggest that Llano Grande was solely engaged in core preparation for trade.

PAGE 232

217 Quartile D istribution D ifferences The Pearson C hi S quare result for frequencies of each cortex quartile value, combining both flake and blade types between Navajas and Llano Grande (n=649 for Llano Grande, n=262 for Navajas), show s a significant association between sites related to the cortex distribution (p<.001). The t au c value shows a negative association, although a week one at .191 (p<.001 ) These results indi cate definite, but relatively subtle differences in lithic processing between sites. Overall mean cortex quartile values between sites for flakes and blades with partial cortex are compared in Table 6.7. The mean difference is small, at 0.39, which t ranslates to an actual difference in percentage of 9.75% T his value is very approximate, given the low resolution of quartile values, but nonetheless reveals a small difference in lithic processing behavior Table 6.7: Means and standard deviations for flakes and blades with partial cortex Site Code n Mean St andar d Deviation cortex Llano Grande 250 2.19 1.110 Navajas 75 1.80 .959 Calculated from the ordinal code values from 1 to 4 which represent successive quartile ranges This result aligns well with the complete and absent cortex p ercentages as well as the chi square result. None of these results present such a wide difference between sites as to suggest a difference between a quarry which is solely invested in core prepa ration for trade, and a remote consumer site. Core preparation consists primarily of the removal of

PAGE 233

218 cortex along with some rudimentary shaping, therefore most of the resulting debitage from that process alone should contain high proportions of cortex H owe ver, the results do suggest that a somewhat lesser degree of later stage production was also done at Llano Grande. Since the differences are statistically significant, the null hypothesis is rejected, but with the observation that some combination of core preparation and later stage production likely occurred in or near the Llano Grande guachimontn. Cortex Distribution Details Bar charts more clearly show the potential mixed economy in the data distribution pattern as a bimodal distribution ( Figure 6 .3): At Navajas, flakes and blades with no cortex expectedly dominate, and flake counts are sharply reduced as cortex proportions increase T he same pattern is shown to a lesser extent for Llano Grande, except that there is a small but noticeable increase in proportion for flakes and blades with over 75 percent coverage T his increase may indicate some presence of increased decortication for the purpose of core preparation for export, although in small proportions compared to the amount of further lithic pr ocessing for local use or trade. Cores at Llano Grande show a cortex distribution that is nearly complementary to the flake cortex distribution, with a peak in the 5 1 75 percent range ( Figure 6.4), whereas cortex on the three Navajas cores are all in the 25 50 percent coverage range. These differences may only reflect the relative proximity of sources, as closer sources tend to be utilized for more expedient and less material efficient knapping (Newman 1994:499). With virtually no travel required to o btain material, Llano Grande knappers could pick and choose from multiple cobbles and spalls that suit their needs with no concern about

PAGE 234

219 Figure 6.3: Flake and blade comparisons of cortex coverage distributions per site Figure 6.4: Core comparisons of cortex coverage distributions at Llano Grande.

PAGE 235

220 efficient utilization of material, whereas Navajas knappers would tend to utilize the material more intensively before retrieving more of it. A more relevant observation is that most of these cores in bo th cases are in the middle quartile ranges, where they are obviously not either pre prepared, or new cobbles cached for later processing. Instead, their intermediate state suggests their use on an as need basis. As noted above, the larger proportion of the se casual cores are at Llano Grande, which suggests a greater expedient industry at the site Test 5: The P ercentage of S pecialized T ools for S taple P rocessing Items in this category are especially sparse. A single metate was found at Navajas withi n the excavated contexts, on the activity floor of one of the satellite buildings. Two additional metates and three manos were discovered among post occupation debris at Navajas, as well as several unidentifiable ground stone fragments in unspecified conte xts (Beekman 2007:253 figure 12.1 ), but only the activity floor metate is included in this analysis, as well as Table 6. 3. The lone metate is also a very small, "single handed" item that may have been used for a number of purposes beside staple processing such as pigment mixing, and by itself is scant evidence However, in the context of t he previously discussed ceramic vessel and diorama evidence for feasting rituals associated with the guachimontones its utilization in some capacity for staple processi ng appears likely. No ground stone implements known or suspected to be associated with food processing were discovered at Llano Grande (Beekman 2001:11) H owever, the complete

PAGE 236

221 lack of staple related items at Llano Grande may well be due to the very lo w coverage of the circles, which did not reach a majority of the buildings. The null hypothesis is quite tentatively rejected for this test, and the results should be considered with a high degree of caution due to the inadequate sample for such a low freq uency item. Confirmation is therefore pending further excavation and analysis. Test 6: The P ercentage of D ebitage A ssociated with P olyhedral B lades All forms of debitage related to polyhedral blades (Table 6. 8 ) are also quite sparse, but frequent e nough to suggest with some confidence that some polyhedral blade industry did exist at both sites. Navajas contains the largest internal proportion of debitage pieces related to polyhedral blade production overall, and also consistently for each subcategor y of debitage Polyhedral blade technology is considered primarily a utilitarian form of production and was more prominent at Navajas, where it was most likely employed for local use rather than trade. Therefore, the null hypothesis is rejected for this te st. Table 6. 8 : P ercentage Comparisons of Blade Types by Count Site/ percent n polyhedral blade core M acro blade 1 st series blade 2 nd series blade Platform correction flake Facial correction flake Total Blade Artifacts Navajas 356 0.3% 2.5% 5.9% 0.6% 0.3 % .8% 10.1% Llano Grande 962 0.0% 1.0% 2.7% 0.1% 0.0% 0.3% 4.2%

PAGE 237

222 Test 7: The P ercentage of U nfinished or F ailed B ifaces This test specifically addresses bifaces prepared for export as blanks, either in an early stage of processing or finely sha ped but not completely thinned and without defining features. Some of the bifaces represented above in Table 6.2 appear to be finished products and are therefore not appropriate for inclusion here. At Navajas, two bifaces are finely finished non standard s hapes: an elongated ovoid and a circular item. Also, a third biface contains very steep edges and is therefore interpreted as a bifacial scraper. All three items appear to be finished utilitarian pieces which were likely used locally. From Llano Grande, on e biface had a broken base and corner retouched into edges, and was therefore a secondary use item. The remaining items, however, were rough bifaces, one of which had a prominent plateau on one face, an early production failure caused by too many step frac tures surrounding material intended to be removed Table 6. 9 : Bifacial Blank Percentages by Count n bifacial blanks Navajas 356 0.6% Llano Grande 962 0.2% Although counts at both sites are again very small, even after ruling out apparent fin ished items, it appears that Navajas bifac ial blanks take up a larger percentage of their respective assemblage than do those at Llano Grande, and the null hypothesis is supported for this test A pparently, much like the La Venta Corridor, Llano Grande did not engage in frequent trade of blanks. Navajas' proportion of bifaces was larger, but still

PAGE 238

223 quite infrequent. Therefore, neither site is likely to have developed a noticeable industry for bifacial blank trade. Test 8: Emphasis on P roduction R elated to L aminar F lakes and B lades For the sake of simplicity in the text, t he term "flake" in this test refers to any informal flake or blade shaped item The only blades excluded from this test are 1st and 2nd series items, which show vertically oriented fla ke scar patterns indicative of formal polyhedral core processing Since thickness was not a recorded dimensio n, laminar flakes are identified as single faceted shaping flakes with less than 25 percent cortex. Although the lack of a thickness paramete r is not ideal, the current method of identification may also catch some failed laminar flakes struck at an angle such that thicknesses were uneven. The method could also misidentify some platform correction flakes, but correction flakes were initially rec orded as a separate type from shaping flakes The percentage of debitage identified as laminar flakes is virtually equal between sites: 1.0 percent of flakes and blades were identified as laminar for Navajas (n= 262 ) and 0.9 percent were identified f or Llano Grande (n= 649 ) These data suggest that laminar production was a minor industry pursued equally at both sites, and may have been intended mainly for internal use in both cases. However, as previously discussed, some of the smaller members of this infrequent and especially fragile category may not have survived intact, or may not have been transported to the guachimontones in equal

PAGE 239

224 proportions between sites. Therefore, possible evidence earlier in the laminar production process may be a deciding fac tor. Tables 6. 9 through 6.1 1 density and each size parameter (length, width, and length width product). Correlation results are shown for all flakes, and also separately for each flake type. T he re is no pre supposition of a positive or negative correlation so p values are calculated as two tailed. Sample sizes differ for different test cases, since certain dimensions on some flakes were not included due to fragmentation, and outliers were also re moved. The null hypothesis is supported by a significantly negative r value, which corresponds to the usual reduction progression towards increased dorsal flake scars, rather than a single scar. The null hypothesis is rejected in cases where a positive cor relation or no significant correlation is discovered. An initial evaluation of all flake types as a single category tests for a flake scar density show a more posit ive trend for laminar flake production whether or not the type distinction truly reflects a distinctive stage boundary on the part of the artisan. The 10 indicate that the correlations of al l dimensions at each site are significantly negative T he strength of the correlation is weak for length and width, but moderate for length x width. Since strong negative correlations would not be expected in assemblages with mixed production methods, this is an expected pattern. The Navajas trend is also slightly less negative than that of Llano Grande. However, the negative correlation in each case is still statistically significant at the p<.05 level. With such a subtle difference between sites, it is un clear

PAGE 240

225 whether the difference is more strongly influenced by factors other than the degree of laminar flake production. Table 6. 10 : Correlations of flake scar density with dimension/size variables for all flake types Site l ength w idth l x w Navajas n 124 125 125 p .002 <.001 <.001 r .280 .376 .487 Llano Grande n 136 138 128 p .001 <.001 <.001 r .285 .296 .480 As shown in Table 6.1 1 when isolating reduction flakes, both sites still show a weak to moderate trend of flake scar densit ies increasing relative to a decrease in size in all cases, and all correlations are again significant at the p<.05 level. No differences which can be attributed to laminar jewelry production are evident in reduction flakes. Such differences are not partic ularly expected in this flake type, since the laminar flakes themselves can only meet the definition of shaping or finishing flakes Table 6.1 1 : Reduction flake correlations of flake scar density with dimension / size variables Site l ength w idth l x w N avajas n 71 73 72 p .027 .002 <.001 r .262 .362 .521 Llano Grande n 65 67 63 p .008 .002 <.001 r .326 .372 .504

PAGE 241

226 However, when shaping/finishing flakes are isolated, the pattern changes. As seen in Table 6.1 2 the significantly negat ive correlation between flake scar density and all size parameters still holds for Navajas, but the scar density correlation between length and width dimensions at Llano Grande is not statistically significant, and is only we a kly negative. T he scar density correlation to the length x width area approximation remains statistically significant. Table 6.1 2 : Shaping/finishing flake correlations of flake sca r density with dimension / size v ariables Site l ength w idth l x w Navajas n 51 50 49 p .022 .010 .0 01 r .320 .363 .478 Llano Grande n 68 69 65 p .116 .083 .001 r .192 .210 .417 The correlation is not significant at the p<.05 level. Within our current understanding of West Mexican lithic technology, the only known technique that may cause an abrupt difference in scar density distribution only within the shaping flake category could be due solely to the presence of final stage laminar flakes. However, as shown above, surviving laminar flakes are essentially equal in proportion bet ween the two sites. Polyhedral core blade technology logically must also produce a reduction in scar density with decreased size down to the prismatic blade stage, since the final products only result in two or three vertically oriented dorsal scars but t his is only necessarily true if cores are worked down to the 2 nd or 3 rd series in Clark's sequence 2 nd series blades were excluded from this test, and no 3 rd series blades were found Also,

PAGE 242

227 Navajas contains twice the proportion of blades compared to Llano Grande ( 9.0% n= 356 for Navajas vs. 3.8 % n= 962 for Llano Grande). S ome additional method of setting up the core for faster and/or easier removal of laminar flakes of the desired shape and size may have been responsible for the observed pattern change at Llano Grande In any case, since only Llano Grande shows some disruption in the scar density to size correlation, the site may be a more likely candidate for larger scale laminar flake production H owever, the requirements defined in Chapter V for th e test state s that all size parameters must show a lack of significance at Llano Grande to reject the null hypothesis. Since the length and width product ( lxw ) still show s a significantly negative correlation, the null hypothesis is supported for this test Also, s ome caution should be exercised because this test relies on logic rather than archaeological or experimental debitage data from actual laminar production. Two of the three parameters, length and width did behave as suggested for an increase in la minar production at Llano Grande. Therefore known laminar debitage should be observed to confirm the interpretation of this test. Of the three techniques considered by analysts thus far for laminar flake production (Beekman 1996 a :797 799 ; Clark and W eigand 2009 ; Long 1966:228 230), all suggest that either the core material was prepared by some form of conventional reduction, or no preparation may have been necessary, prior to creating flakes with single dorsal scars. Since any disruption of the flake scar to size trend does not appear in reduction flakes, but only shaping/finishing flakes, t he results of this test l end some further strength to the idea that only conventional early preparation, if any, preceded finer work. However,

PAGE 243

228 removals done later i n the sequence may have involved more than a single step production of laminar flakes. Flake S ize/ C ortex D istribution G rouping P atterns Scatter plots of the relationships between dimensions and scar densities are shown in Appendix B Reduction and sh aping flakes are marked differently within the same plots for easy comparison. All charts show essentially the same general negative sloping (though non linear) trend determined through statistical analysis, although it is less obvious in the shaping flake category of most of the charts, mainly due to their more limited size range For Llano Grande, slopes are especially difficult to detect because of their very weak correlation s

PAGE 244

229 CHAPTER VII DISCUSSION AND CONCLUSION This chapter begins with a sum mary of the results of each test, as each relates to the original research questions, followed by a discussion of the nature of the lithics within the contexts of the guachimontones, and the significance of the results to differences in economies at Llano Grande and Navajas in general P roject limitations and issues are then discussed, along w ith future work that may stem from this project and the questions it has addressed. Question 1: Do the lithic assemblages at Llano Grande show a greater emphasis on production over use contexts than the assemblages at Navajas and other sites more closely affiliated with the core population zone? A greater emphasis on production in general at Llano Grande is well supported from the available assemblages. The av erage lithic density at Llano Grande is more than ten times that of Navajas and projecting the average density over the entire guachimontn suggests that the ritual center may contain over 30,000 pieces T he assemblages at hand are specialized contexts wi thin public centers, and therefore show a specific view of the political economic emphasis of the site on obsidian production T he guachimontones are likely secondary contexts, so it is also likely that neither guachimontn represents the total sum of flak es p roduced at its respective site Therefore, i t is not suggested that one site has a greater throughput of obsidian production over the other, but only that the

PAGE 245

230 specific public context reflects a political economy at Llano Grande which emphasized the pro duction of obsidian as an important commodity to a greater degree than at Navajas. However, based on the current evidence, only the amount of obsidian at Llano Grande might be deemed large enough to suggest the presence of a workshop on site Ad ditional de posits, such as middens and actual workshop locations, may show evidence of production on an even larger scale The Navajas guachimontn which covers nearly the same area as the Llano Grande ritual center and was much more thoroughly excavated, cont ains less than 3 000 pieces. It also contains proportionally more formal products by count at 23.7 percent than Llano Grande at 4.9 percent P roportions of e very product form are greater at Navajas, which includ e non prismatic polyhedral core blades cons idered to be utilitarian items, unifaces and steeply edged bifaces interpreted as scrapers one metate and two apparent ritual/eccentric product types, pikes and laminar jewelry These data suggest a larger role for Navajas as a lithic consumer site than L lano Grande Again, this does not mean that a high degree of production did not occur somewhere on site, but only that lithic production was not well represented within the public ritual architecture at Navajas Question 2: Does Llano Grande show evidence for a greater emphasis on the production of potential trade than Navajas, including commonly exported forms, known elite items, and/or items which require specialized production techniques? Evidence for the production of most items known to have be en created for trade, however, is not as clear. Items commonly traded within Mesoamerica are prepared (at

PAGE 246

231 minimum, decorticated) cores; bifacial blanks, which were very common trade items at Teotihuacn; and eccentrics. Only prepared core production presen ted clear evidence of greater emphasis at Llano Grande than at Navajas. Distributions of different percentages of cortex suggest that Llano Grande likely did function as a quarry for core preparation, but not exclusively. Core preparation is clearly indica ted by a significantly larger component of flakes with very high amounts of cortex at Llano Grande. However, the lack of cortex on the majority of flakes suggests a larger component of additional processing for specific products, although it is not possibl e at this point to confidently suggest specific forms which may have been created, solely based on the recorded attributes of the debitage. The number of bifacial blanks is very low at both Llano Grande and Navajas, and actually accounts for a slight ly greater proportion of the Navajas collection. As seen in other sites within the Tequila Valleys, Teotihuacn's strategy of regional distribution of bifacial blanks does not appear to have been shared by the Teuchitln culture. Bifacial blank production was not necessarily expected, but only provided a n indicator of one possible form of trade. For eccentric or ritual items, t he most obvious and intuitive method of simply counting flakes and fragments with laminar characteristics to identify jewelry p roduction resulted in essentially equal proportions between sites. Searching for an uncommon relationship between flake scar densities and flake sizes which suggest reduction that works towards the single scar laminar products did result in potential evide nce for laminar production at Llano Grande, which was not present at Navajas. However, this method does not coincide well with any current proposals for the laminar production

PAGE 247

232 techniques, and requires further testing through experimentation to determine wh ether some form of gradual elimination of flake scars may have been of some benefit. Finally, the presence of casual cores at Llano Grande suggests the production of some expedient tools more suited to internal use rather than trade. However, most of the cores were more than half covered in cortex, and appear very lightly used. The proportion of flakes produced from those cores within the guachimontn is therefore likely very low, but the caching of both expediently produced flakes and secondary flake s from debitage within the circle, combined with the small percentage of retouched edges on flakes, suggests an emphasis on internal use of expedient items. However, Navajas also contains a large proportion of utility items, including three times Llano Gra nde's relative proportion of polyhedral blades (mainly 1st series and macroblades), which are also considered internal utility items. Navajas' greater proportion of the more material efficient polyhedral form (Sheets and Muto 1972) over casual flakes is un derstandable, considering the site's greater distance to its source. However, the emphasis on different utility forms at each site poses a problem for interpreting relative degrees of internal utilization of lithics. A small number of items specifically cr eated for internal use may be more intensively utilized than a cache of secondary flakes which are stored only for convenience as potentially usefulness edges T he current study also does not provide data on the number of secondary and expedient flakes or blades which were actually used at either site beyond the more obvious edge retouch data. But the tool data thus far suggest active internal utilization of lithic resources typically designated for common uses at both sites.

PAGE 248

233 Overall, the currently ava ilable data suggest that Llano Grande elites did embrace more of a trade economy than Navajas, but trade is only substantially evident in prepared cores rather than blanks or eccentrics Llano Grande was also not overwhelmingly trade oriented, as the diffe rences between the two sites in cortex proportions are statistically significant but not strongly divergent. Flake scar densit y distributions do hint at possible jewelry production, but the method of analysis needs further confirmation and the more straig ht forward indication of actual counts of laminar debitage do not indicate any difference between sites. The Origin and Purpose of Guachimontn Obsidian The question of how obsidian lithic material came to be deposited specifically within the Llano Grande and Navajas guachimontones relates directly to the issues with workshop identification discussed in C hapter IV Of particular interest is the assertion posed by Moholy Nagy (1990) that supposed workshop deposits at Colha were actually secondary and therefore Colha was not a workshop but a lithic dumping area. However, as Johnson has argued, regardless of the nature of the deposition, the deposits were evidently the result of workshop activity that had to be nearby, if not directly on the premises (19 96:164). Also, in the context of the current study, an active public ceremonial center would not have been a good candidate for what Moholy Nagy called a "workshop dump" (1990:268). Even though the distribution signature s for both sites suggest an indiscri minant move of all sizes which can be easily transported into each ceremonial center, the lithic deposits there most likely served a specific purpose. As

PAGE 249

234 suggested above, t he guachimontn buildings may have been used as a convenient cache for potential exp edient tools, to sort through at later occasions. The strategy has some precedent, as it is suspected by Whittaker in another enclosed lithic "waste" context near the corner of one room within an American Southwest pueblo. Whittaker reasons that in the Pue blo example, the usefulness of expedient edges is suspected to have delayed final dumping in a midden context (Whittaker 2001). Among the Maya, similar lithic deposits which appear to have indiscriminately included most of the debitage as well as pro ducts have been found in the ritual center of Classic Period Tikal. The deposits at Tikal have been interpreted as ceremonial offerings, but those items appear to have been buried within the public space (Moholy Nagy 1989 ; Johnson 1996:166), whereas the de posits at Navajas and Llano Grande were primarily concentrated just above the living floor and among the building collapse, and were therefore exposed for potential reuse. The guachimontn deposits may have been intended as potential tools for the crafting of ceremonial items, for common everyday use if some degree of secularization of the circle had occurred to allow for more common daily activities within the circle But t he concentration of expedient flakes in the Llano Grande guachimontn suggests a possibility of more intensive use than what may have been required for internal activity Utilitarian lithic items may be used for a number of different purposes beyond staple processing, but ceramic studies at Navajas and previously in other areas inha bited by the Teuchitln culture area have detailed a prominent role for staple related ritual within guachimontones based on vessel forms (Butterwick 1998 ; Johns 2014 ; Tyndall and

PAGE 250

235 Beekman 2007), and the density of ceramics by count is approximately nine pe rcent higher at Llano Grande than at Navajas. It is likely that at least some expedient flakes at Llano Grande were utilized for staple processing, but Llano Grande's high informal flake concentration also suggest s obsidian utilization within the rit ual center that was similar to that of the previously discussed workshop s at Kaminaljuyu ( Anderson and Hirth 2009 ) and Teotihuacn (Spence 19 87 ) where expedient tools were used for craft production. Several additional resources around Llano Grande could h ave been used for this purpose. The wooded area s of the hills surrounding Llano Grande suggests a possibility of carved wood products, and perhaps animal skins from animals that inhabited the forests. Maguey products, including fibers, hearts, leaves, nect ar and pulque are additional potential exports which also likely required cutting edges for processing. All of these examples are more perishable and thus more difficult to detect in archaeological contexts but some may have also been considered elite ite ms in other areas with out these resources (Nassaney 1996) Conclusion As anticipated, the geographically and environmentally diverse Tequila Valley s region provided an excellent opportunity to observe differing economic activity in different enviro nments which are nonetheless connected to the same cultural identity. Different positions within the economic structure also suggest different anticipated social roles within the culture which can further affect expected economic strategies (Algaze 1993; B lanton et al. 1996; Chase Dunn and Hall 1991; Earle 1991; Schneider 1977; Wallerstein

PAGE 251

236 1974) T he degree to which different semi peripheral and peripheral groups fill their expected roles may also depend on each group's degree of independence from the core (Kowalewski et al. 1983; Ohnersorgen and Varien 1996). Given Navajas close connection with the core, it is not entirely surprising that the lithic data from the Circle 5 guachimontn shows an emphasize on production for internal use which would like ly have been primarily applied to feasting related activities such as food preparation, and perhaps other agriculture related rituals However, the lack of any indication of trade production at Navajas is initially surprising, given the clear indication of some Late Formative use of Navajas obsidian within the Sayula Basin. The Sayula obsidian sourced to the Navajas quarry does, however, predate the site at Navajas to Sayula's Early Usmajac phase, ca. 400 200 B.C. (Reveles 2006:389 ). Also, a much stronger cultural influence from the core likely affected Navajas' use of the guachimontn, and how it relates to the local economy. Based on the evidence at both Los Guachimontones and Navajas, the stashing and use of large quantities of lithic material within gu achimontones does not appear to be a typical pattern associated with the core. Navajas may have kept large amounts of lithic material from accumulating within the guachimontn due to social norms regarding appropriate use of the guachimontn space regardl ess of the site's degree of lithic production and trade The evidence at Llano Grande suggests a combination of obsidian staple and possibly additional craft based economies rather than a relative trade off from a prominent staple economic base tow ards one based on wealth. Llano Grande had less convenient access to fertile agricultural land than Navajas but evidently Llano Grande's access was not enough to deter the site from continu ing a staple economy O nly a minor proportion of

PAGE 252

237 the data currentl y available from Llano Grande very strongly suggest s any specific known form of production which w as commonly traded namely prepared cores M ost of the obsidian artifacts at Llano Grande can only be argued as the result of workshop production activity fo r trade on the basis of quantity. The presence of potentially tens of thousands of pieces deposited in a single context, whether primary, secondary or final deposits, is most easily explained by the presence of a workshop in close proximity This is especi ally true when considering that the deposits are likely secondary, and therefore probably only a fraction of the total debitage at the site S pecifics about the exact forms of lithic products which were created are more difficult to discern at Llano Grande based on the morphologies and flake attributes present at the site Llano Grande's very similar concentration of ceramics to Navajas is likely the result of the maintenance of staple related activity associated with the core area. The lack of speci alized staple specific tools at Llano Grande is problematic to this interpretation, but considering only one object in this category was found at Navajas near an activity context, the lack of excavation coverage at Llano Grande is a more likely explanation for this absence than a lack of staple processing. Some proportion of the utilitarian lithics found at both sites were therefore likely applied to staple processing for ritual purposes while a large proportion may have also been applied to other crafts, perhaps for other forms of wealth trade The more intensive utilization of a guachimontn for storage of production byproducts at Llano Grande also likely reflects some level of increased leverage from social institutions more heavily invested in lit hic and/or craft production and greater institutional sanctioning of the industr ies for which these lithics were utilized

PAGE 253

238 Sanctioning may take one of a few forms T he related industries could have been tied to ideology and then incorporated into the ritu al system as some scholars view a ritual economy (Demarest 2013:372 ; Wells 2006 ; Wells and McAnany 2012) Alternatively, they may have been merely tolerated by ritual oriented institutions Finally, some degree of secularization of the circle could have ta ken place resulting in the production or storage of lithics for domestic use. The possibility of secularization appears unlikely, since Llano Grande does not appear to have abandoned or even reduced the intensity of the staple ritual aspects of the core area based on the concentration of ceramics T he ceramic data from Llano Grande requires additional analysis to better confirm its use within the guachimontn but initial observations by Beekman (2001:7) show similar ceramic forms to those found wit hin the Navajas guachimontn (Johns 2014). Specifically, the Llano Grande ceramics are a combination of fine ritual wares and common household vessels which include storage jars and bowls I t is also unlikely that elites would assume a passive role, and si mply allow a ctivity centered around a valued resource like obsidian to increase within the ritual space and ignore the opportunity to leverage th at resource and the industries for which it was utilized in order to increase their own power In Kenneth Hirth 's view ideological ties to production generally appeal to ideas about the benefits of related resources to the group as a whole. The organization of institutions around production of the resource then provides legitimation of unequal resource accum ulation within the social structure (1996:225). Economic ideology can then become a vehicle for bringing about ideological change: "It is in this arena that elites promote the development of new ideologies...to shape belief about both the demand for resour ces and

PAGE 254

239 the specific use for which they are collected" (Hirth 1996:226). But the contention suggested by the concept of the arena does not necessarily lead to simply pitting of two economies against each other, where one economy becomes dominant in the end A lternatively, c ontention may be over economic diversification vs. continued specialization on a more traditional resource. Elite appeal to economic diversific ation is described by Hirth's matrix control principle ( 1996:223 225 ) as a way to control as ma ny economic and human resources as possible, and re duce the risk of an economic failure through diversification of resources Hirth (1996:223 22 4 ) further states that c ontrol over the accumulation of multiple resources is accomplished through his c on text p rinciple whereby elites creat e public spaces for each industry so that all industries can be monitored by a higher level authority in the social structure. Additional differences between Llano Grande and Navajas in the patterns of artifact distribut ion between buildings suggest that the Llano Grande guachimontn may have transformed to accommodate multiple production contexts. Within the Llano Grande guachimontn, 70 percent of the analyzed lithics artifacts by count (n=962) were found in a single bu ilding, 14 6 Also, as mentioned in Chapter III building 14 6 is associated with the eight buildings outside the circle, positioned closely behind it (figure 3.7), which suggest s a special purpose related to lithic or craft production for 14 6 C eramics however, are nearly all concentrated in the adjacent building 14 5, including the only whole serving and storing vessels discovered at the site (Beekman 2001: 7, 11 12 ) Contrastingly, a lthough lithic concentrations at Navajas still varied considerabl y between buildings, t he highest lithics concentration in any one building at Navajas is 48

PAGE 255

240 percent of the total artifact sample by count (n=356) For the Navajas ceramic data, sherds were also variably distributed, but the highest sherd count concentratio n in any one building was 32 percent (n=9 475) a lthough some buildings held sherd frequencies as low 4% of the total ( J ohns 2014:figure V 23). Johns concluded that sherds were distributed evenly enough between buildings to suggest the use of pottery for r i tual activity in every building. These data suggest that groups associated with each building at the Navajas guachimontn may have handled these industries independently of each other, whereas the Llano Grande guachimontn saw more group specialization fo r each industry, likely coordinated by a higher authority. Differences between the core corporate economy and sites which follow Hirth's matrix control model require a replacement of the top level corporate structure with a more vertical structure whi ch allows an over arching institution to control even such diverse economies as obsidian trade and agriculture. This form of control would have been most easily accomplished in the semi periphery, where social distance from the corporate core and outside i nfluences likely brought in new ideas and fostered economic and structural innovation, yet also allowed some continued identification with the core to create a form of hybrid economy. The evidence at Llano Grande suggests that elites appear to have sought to expand their economic basis by adding obsidian trade and perhaps other craft related industries which required obsidian tools without losing their hold on existing staple production organization or the accompanying agriculture related ritual. The presence of the surrounding walls at L lano Grande, however, appear to contradict this interpretation of an independent Llano Grande, since the wall structures appear to have required a labor force beyond that which could have been provided by the

PAGE 256

241 populati on of the site, which suggests some investment by core elites towards control of the site as a territorial boundary. But such attempts at establishing territorial control are risky, and sometimes fail due to the costs of maintaining transportation and comm unication links to remote groups ( Kowalewsi 1983:37 ). Llano Grande's distance from the core and relatively difficult access may have made core control too expensive to maintain ( Ohnersorgen and Varien 1 996) which would have afforded Llano Grande a greater degree of independence than what may have been anticipated by core elites Llano Grande would likely have had little external social pressure to continue the fragile corporate strategy, which would have allowed the more powerful groups to assume a control ling position a nd expand their economy However, given Llano Grande's moderate size (approximately 75 buildings) economic expansion wou ld have been somewhat limited by a s mall labor force. Challenges Limitations and Lessons Learned The main limit ation with this project was the lack of excavation coverage at Llano Grande. Projections of overall quantities are based on extremely variable artifact densities over o nly three of eight buildings and a one to two meter wide trench through the patio. The e conomic implications of unexplored areas of the guachimontn could remove a great deal of ambiguity and change the interpretation the role of the ritual center and the economy of the overall site quite dramatically. The focus on only the guachimontones w i t hin each site is also very narrow. The ritual center is an important area to explore because of its central socio economic role within the site, but the discovery of actual

PAGE 257

242 workshop areas and differences in the use of lithic material within buildings outsi de of the guachimontones can add critical information to the interpretation of the site's economy. However, since my own coverage of the existing excavations was not comp rehensive additional excavation may not have mattered for this project P rogres s for artifact analysis was somewhat slower than it could have been, for a few reasons F or one, I had made s ome variable choi ces that were not actually useful for the current question. Attributes were actually recorded with very different questions in min d that were tied to a large r number of data attributes than the current topic Consequently, although several attributes were also quite useful for the current project, much time was spent recording data that were never utilized. Also, at least one importa nt attribute, maximum thickness, was among several dropped within the first few days of lab work partly due to a reduction in time allotted for laboratory analysis. As previously mentioned in Chapters V and VI, thickness would have been quite useful for t he identification of laminar flakes. Closer analysis of the laminar flakes available, and especially subtle aspects of their surface features, also could have been a tremendous benefit to the general knowledge about lithic production techniques unique to W est Mexico The slow progress of analysis was also exacerbated by the large proportion of items of unknown artifact status (the "chunk" items) to identify and sort through. However, the prioritization of lots and units above activity floors then prove d quite useful. The majority of actual artifacts recovered from both sites (and nearly exhaustive coverage below ground at Navajas) were covered, as were artifacts from the most desired contexts closest to the activity floors

PAGE 258

243 Additional time spent o n consultation for refinement of the research design would also have been very beneficial especially given the lack of published literature on lithics within the region. Finally, since d ebitage analysis has proven to be quite labor intensive and tedious, as one may expect, analysis time is at a premium. A ctive recruitment of undergraduate or graduate lab partners c ould also have been tremendously beneficial as well as a provision of valuable experience to potential future colleagues. Future Opportunitie s As mentioned at the end of Chapter I this thesis was only intended to be a start in a particular direction on data which had previously undergone only brief preliminary analysis The project was therefore very much a first pass at a theory driven exploration of the data and the re is a tremendous potential for additional studies The numbers and nature of the lithics recovered thus far already set Llano Grande apart from the core area, yet the interpretation of much of the lithic data remains ambiguous, largely due to lack of exploration of the surrounding context. The differences seen thus far likely point to future discoveries to be made in the yet unexplored building mounds and patio space. New information will likely add valuable details o r counter the very tentative conclusions drawn in this thesis.

PAGE 259

244 Continued Exploration at Llano Grande The first priority in regard to studies that may specifically follow the current project should be a dditional artifact analysis of many of the r emaining samples at Llano Grande Of highest priority among the unanalyzed items are those from within higher collapse contexts than the items analyzed for this thesis, which were deliberately selected from lots immediately above and below activity floors A dditional excavation should then target more contextually representative samples within the Llano Grande guachimontn within additional buildings and in other areas of the patio. Future excavation, handling and storage protocols should also keep additio nal residue analysis in mind for some of the recovered artifacts and take measures to avoid contamination This line of questioning can then also benefit from additional residue analyses and use wear studies to determine the proportions of used flak es and blades, and the nature of the local use of different tool forms within both sites The frequency of tool use for processing possible food items is of particular interest. A dditionally Rodrigo Esparza and Camilo Mireles are conducting further resear ch on the manufa c ture of laminar flakes which may inform better methods of identifying less obvious places of manufacture for these items. Studies involving very close observation and experimentation for both laminar flake and pike manufacture will allow better identification of all distincti ve West Mexican eccentrics. Following more representative exploration of the guachimontones, other buildings and patio spaces within the site should be selected for further exploration. Of particular interest are additional buildings immediately behind the Llano Grande guachimontn

PAGE 260

245 building 14 6. One of those buildings has already been excavated, and additional buildings within the group may provide additional information about 14 6's nature and purpose. Surveys and Additional Excavations A reasonabl y complete view of the nature of the Llano Grande and Navajas econom ies will also require a much more broad exploration of the site s beyond the guachimontones, and additional surveys are needed for the surroundin g areas on both sides of the Teuchitln culture's "boundary" between the semi periphery and periphery to confirm patterns of settlement and artifact distribution of all forms (pottery, figurines and various elite imports as well as obsidian) Surveys can s pan multiple seasons, and should be seen as a parallel priority to existing studies of excavat ed contexts, since resulting architecture maps and surface artifacts can inform questions driving future excavations, and assist in targeting the best strategic s ites for answering those questions So far, ongoing surveys have concentrated on the semi peripher y areas inside the Tequila V alley s (Heredia 2008). These surveys are targeting a comprehensive view of the Valleys, and certainly should continue before peripheral areas are considered Much of the core also remains undocumented, despite Phil Weigand's ( 1975, 1985 among others ) early survey efforts and should be re explored and the data made available via more detailed published reports Eventually, arch itecture and artifacts w hich may be discovered in peripheral areas adjacent to the valleys and in the known peripheral guachimontones

PAGE 261

246 may point to nearby external trade networks and elemental sourcing techniques of external items can potentially confirm a role for Llano Grande in regional trade. Finally, other guachimontn sites in both semi periphery and core areas need to be explored on the same scale as Llano Grande and Navajas and with similar data to determine commonalities and differences betwe en sites. Some evidence of v ariation in the proportions of lithic forms between different semi peripheral areas has been briefly discussed within this thesis, and patterns of material culture variation within and between economic zones is another potential topic for exploration. Other sites within the hills west of the Magdalena Basin may have been the most frequent contacts with Llano Grande, and may be high on the priority list of sites to explore. New Questions with Expanding Exploration E xpan ding our scale of exploration into the periphery and into other semi peripheral sites can also inform additional questions about the nature of the relationships between core, peripheral, and semi peripheral groups. What economic bases and strategies might be found in peripheral sites, and how do these areas interact with semi peripheral and core sites? Do peripheral sites appear to benefit from their relationships with the core, or do they only become dependent as a result? Considering Stein's view of a variab le periphery, how much variability do we see between different peripheral groups and their relationships with the core? What are the natures and purposes of peripheral guachimontones? Are corporate ideals of the core adopted by peripheries to some extent,

PAGE 262

247 or do they partly mimic and reinterpret the corporate institutions of the core to fit their own economic circumstances? Beyond the Tequila Valleys, this study may also have implications for other emergent complex cultures both within Mesoamerica, and throughout the world D id other semi peripheral areas challenge the core, both economically and ideologically through changing economic strategies? Variations in how economic strategies may have been altered in different areas may relate to such factors a s differing environments, available resources the economies of other peripheral groups and core ideologies. Other Paradigms The Tequila Valleys provide opportunities for the exploration of questions in other theoretical paradigms as well Niche con struction theorists are also interested in the impact of external environments on people groups, but may further ask the reciprocal question of what effects groups had on their surroundings. Together, these questions reveal the overall reciprocal effect of human environment interaction within the Tequila Valleys. The effects of agriculture, human population expansion into the passes, and other architectural features such as terraces and walls are all important aspects of human niche construction that direct ly impact available resources. Cultural macroevolutionary theorists may explore the expansion of groups into the semi periphery and periphery to determine whether patterns of material culture variation seen in culture expansions from other parts of the wor ld (e.g. Chatters and Prentiss 2005) are seen among the sites established during the Early Classic expansion of the Teuchitln culture and how the

PAGE 263

248 variation may be generated via cultural mechanisms. These are only a couple of examples of different theoret ical directions that may find new ground in the relatively unexplored area of the Tequila Valleys. The work required to address each of these questions is far reaching in terms of labor and time investment, and require s a long term view of research i n the area. Each season will certainly bring us new and valuable insight into West Mexican prehistory, but g oal driven archaeological research in West Mexico can no longer afford the short term, season by season view of investment and return, as the implic ations of economic patterns are far more complex than what can be derived from discrete views of ritual centers, and even whole sites.

PAGE 264

249 REFERENCES Adams, Robert M. 1974 Anthropological Perspectives on Ancient Trade C urrent Anthropology 15: 211 239 Adams, Ron L. 2004 An ethnoarchaeological study of feasting in Sulawesi, Indonesia. Journal of AnthropologicalArchaeology 23: 56 78. Ahler, Stanley A. 1989 Experimental Knapping with KRF and Midcontinent Cherts: Overview and applications. In Expe riments in Lithic Technolog y edited by Daniel S. Amick and Raymond P. Mauldin pp. 199 234 B AR International Series 528 British Archaeological Reports Oxford. Alden, J ohn R. 1979 A Reconstruction of Toltec Period Political Units in the Valley of M exico. In Transformations: Mathematical Approaches to Culture Change, edited by Colin Renfrew and Kenneth L. Cooke, pp. 169 200. Academic Press, New York. Algaze, Guillermo 1993 Expansionary dynamics of some early pristine states. American Anthropologi st 95(2): 304 333. Anawalt, Patricia 1998 They Came to Trade Exquisite Things: Ancient West Mexican Equadorian Contacts. In Ancient West Mexico: Art and Archaeology of the Unknown Past edited by Richard F Townsend, pp. 233 249. Thames and Hudson, N ew York. Anderson, J. Heath and Kenneth G. Hirth 2009 Obsidian Blade Production for Craft Consumption at Kaminaljuyu A ncient Mesoamerica 20(1): 163 172. Anderson, Kirk, Christopher S. Beekman, and Verenice Yunuen Heredia Espinoza. 2013 The Ex Lagu na de Magdalena And Pre Columbian Settlement in Jalisco, Mexico: The Integration of Archaeological and Geomorphological Datasets. Paper prepared for presentation at the session environment interactions in the Neotropics: historical impact to current Sarah Metcalfe, Elizabeth Rushton, and John Carson for the Royal Geographical Society (with the Institute of British Geographers), London. Andrefsky, W. 2005 Lithics: Macroscopic Approaches to Analysis Second Edition. Cambr idge University Press Cambridge.

PAGE 265

250 Aoyama, Kazuo. 2005 Classic Maya Warfare and Weapons: Spear, dart, and arrow points of Aguateca and Copan. Ancient Mesoamerica 16: 291 304. Barret t, Jason W. 2004 Constructive hierarchy through entitlement: inequal ity in lithic resource access among the ancient Maya of Blue Creek, Belize. Ph.D. dissertation, Department of Anthropology, Texas A&M University College Station Electronic document accessed online from http://txspace.tamu.edu/handle/1969.1/1338 on February 10, 2010. Becker, Marshall J. 1992 Burials as Caches, Caches as Burials: A New Interpretation of the Meaning of Ritual Deposits Among the Classic Period Lowland Maya. In New Theories on the An cient Maya edited by Elin C. Danien and Robert J. Sharer, pp. 185 196. The University Museum, University of Pennsylvania, Philadelphia. Beekman, Christopher S. 1996a The Long Term Evolution of a Political Boundary: Archaeological Research in Jalisco, Mexico Ph.D. dissertation, Department of Anthropology, Vanderbilt University, Nashville. 1996b Political boundaries and political structure: The limits of the Teuchitln tradition. Ancient Mesoamerica 7: 135 147. 2000 The correspondence of region al patterns and local strategies in Formative to Classic period West Mexico. Journal of Anthropological Archaeology 19: 385 412. 2001 Proyecto Arquelogico Valles de Tequila, Temporada de 2000: Mapeo y excavacin en Llano Grande Manuscript on file, A nthropology Department, University of Colorado Denver. 2003a Agricultural Pole Rituals and Rulership in Late Formative Central Jalisco. Ancient Mesoamerica 14 (2): 299 318 2003b Fruitful symmetry: Corn and cosmology in the public architecture of L ate Formative and Early Classic Jalisco. Mesoamerican Voices 1: 5 22. 2007 Introduction to the 2003 Season. Tequila Valley Regional Archaeological Project Season Report 2003 edited by Christopher Beekman, pp. 6 21. Manuscript on file, Anthropology Dep artment, University of Colorado Denver. 2008 Corporate power strategies in the Late Formative to Early Classic Tequila Valleys of central Jalisco. Latin American Antiquity 19: 414 434.

PAGE 266

251 Beekman, Christopher S. 2010 Recent Research in Western Mexican Archaeology. Journal of Archaeological Research 18(1):41 109. I n P ress a Built Space as Political Fields: Community vs. Lineage Strategies in the Tequila Valleys. In Alternative Pathways to Complexity: Households, Markets, World Systems, and P olitical Economy: Essays Honoring the Legacy of Richard E. Blanton, edited by Lane F. Fargher and Verenice Y. Heredia Espinoza, pp. 57 73 University Press of Colorado, Boulder. I n P ress b Conflicting Political Strategies in Late Formative Early Cla ssic Central Jalisco. In Political Strategies in Pre Columbian Mesoamerica edited by Sarah Kurnick and Joanne Baron, University Press of Colorado, Boulder Beekman, C hristopher S. and A. F. Christensen 2003 Controlling for doubt and uncertainty throu gh multiple lines of evidence: A new look at the Mesoamerican Nahua migrations. Journal of Archaeological Method and Theory 10: 111 164. Beekman, Christopher S. and Phil C. Weigand 2008 Conclusiones, Cronologa, y un Intento a Sntesis. In Tradicin Te uchitln edited by Phil C. Weigand, Christopher Beekman, and Rodrigo Esparza, pp. 303 337. Colegio de Michoacn, Zamora, Mxico. Beekman, Christo pher S., and Stephen D. Houston 1993 Political Boundaries in Ancient Mesoamerica Paper presented at the 58th Annual Meeting of the Society for American Archaeology, St. Louis. Beekman, Chris topher S. and William W. Baden 2011 El cultivo del maz y su impacto regional: Agotamiento de los suelos en el corredor de La Venta, Jalisco. In Patrones de asentamie nto y actividades de subsistencia en el Occidente de Mxico. Reconocimiento a la Dra. Helen P. Pollard edited by Eduardo Williams and Phil C. Weigand, pp. 351 382. Colegio de Michoacn, Zamora Mxico Begun, Eric Martel 2013 Detecting Ethnicity at T eotihuacan through Archaeology: the West Mexican Presence at Structure N1W5:19 Ph.D. dissertation, Department of Anthropology, University of Iowa Iowa City. Electronic document retrieved from http://ir.uiowa.edu/cgi/viewcontent.cgi?article=4821&context=e td, accessed April 12, 2014.

PAGE 267

252 Benz, Bruce F. 2007 Preliminary Analysis of Archaeological Material from Bulk Sediment Samples. In Tequila Valley Regional Archaeological Project Season Report 2003 edited by Christopher S. Beekman, pp. 248 255. Manus cript on file, Anthropology Department, University of Colorado Denver. Blanton, Richard E. and Gary M. Feinman 1 984 The Mesoamerican World System. American Anthropologist 86(3): 673 682. Blanton Richard E., Gary M. Feinman, Stephen A. Kowalewski and Peter N.Peregrine 1996 A Dual Processual Theory for the Evolution of Mesoamerican Civilization. Current Anthropology 37(1): 1 14. Bleed, Peter 2001 Trees or Chains, Links or Branches : Conceptual Alternatives for Consideration of Stone Tool Productio n and Other Sequential Activities. Journal of Archaeological Method and Theory 8(1):101 127. Botanicals Online 2014 Holm Oak (Holm oak characteristics) Electronic source, accessed online from http://www.botanical online.com/english/holmoak.htm on Oct ober 4, 2014 B o da, E. 1995 Levallois: A volumetric construction, methods, a technique. In The Definition and Interpretation of Levallois Technology edited by H. Dibble and O. Bar Yosef, pp. 41 68 P rehistory Press, Madison. Bourdieu, Pierre 2002 Social Space and Symbolic Space. In Contemporary Sociological Theory edited by Craig Calhoun, Joseph Gerteis, James Moody, Steven Pfaff, and Indermohan Virk, pp. 267 275. Blackwell Publishing, Oxford. Braswell, Geoffrey E. E. Wyllys Andrews V, and Mic hael D. Glascock. 1994 The Obsidian Artifacts of Quelepa, El Salvador. Ancient Mesoamerica 5: 173 192 Braswell, Geoffrey E., Iken Paap, and Michael D. Glascock 2011 The Obsidian and Ceramics of the Puuc Region: Chronology, Lithic Procurement, and P roduction at Xkipche, Yucatan, Mexico. Ancient Mesoamerica 22(1): 135 154. Breton, Adela C. 1903 Some Mexican Portrait Clay Figures. Man Volume 3, pp. 130 133. The Anthropological Institute, London.

PAGE 268

253 Breton, Adela C. 1905 Some Obsidian Workings in Mexico. International Congress of Americanists: Thirteenth Session Held in New York pp. 265 268. Butterwick, Kristi 1998 Food for the Dead: the West Mexican Art of Feasting. In Ancient West Mexico: Art and Archaeology of the Unknown Past edited by R ichard F Townsend, pp. 89 105. Thames and Hudson, New York. Callaghan, R. T. 2003 Prehistoric trade between Ecuador and West Mexico: A computer simulation of coastal voyages. Antiquity 77: 796 804. Cabrero Garca, Mara Teresa 1989 Civilizacin en el Norte de Mxico: Arqueologa de la Caada del Ro Bolaos (Zacatecas y Jalisco). Universidad Nacional Autnoma de Mxico, Mxico, D.F. Carballo, David M. 2009 Household and Status in Formative Central Mexico: Domestic Structures, Assemblages, an d Practices at La Laguna, Tlaxcala. Latin American Antiquity 20(3): 473 501. Carr, Philip J. and Andrew P. Bradbury 2001 Flake Debris Analysis, Levels of Production, and the Organization of Technology I n Lithic Debitage: Context, form, meaning, edited by Willliam Andrefsky, Jr pp. 126 146. University of Utah Press Salt Lake City. Chadwick, Ian 2011 Tequila: In Search of the Blue Agave Electronic source, accessed online from http://www.ianchadwick.com/tequila/agave_growing.htm and http://www.ian chadwick.com/tequila/cultivation.htm on July 23, 2014. Chase Dunn, Christopher, and Thomas D. Hall 1991 Conceptualizing core/periphery hierarchies for comparative study. In Core/periphery relations in precapitalist worlds edited by Christopher Chase D unn and Thomas Hall, pp. 5 44. Westview Press, Boulder. Chatters, James C. and William C. Prentiss 2005 A Darwinian macro evolutionary perspective on the development of hunter gatherer systems in Northwestern North America. World Archaeology 37(1): 46 65.

PAGE 269

254 Clark, John E. 1981 Towards a definition of workshops Paper presented at the Obsidian in Mesoamerica Symposium, January 26 to 31, edited by Henry Pratt. Fairchild, Pachuca, Hidalgo Mexico 1982 Manufacture of Mesoamerican Prismatic Blad es: An Alternative Technique. American Antiquity 47(2) : 355 376. 1987 Politics, Prismatic Blades, and Mesoamerican Civilization. In The Organization of Core Technology edited by Jay K. Johnson and Carol A. Morrow, pp. 259 284. Westview Press, Boulder. 1988 The Lithic Artifacts of La Libertad, Chiapas, Mexico: An Economic Perspective. Papers No. 52, The New World Archaeological Foundation, Brigham Young University, Provo UT 1997 Prismatic Blademaking, Craftmanship, and Production: An analysis of Obsidian Refuse from Ojo de Agua, Chiapas, Mexico. Ancient Mesoamerica 8(1): 137 159. Clark John E. and Douglas Donne Bryant 1997 A Technological Typology of Prismatic Blades and Debitage from Ojo de Agua, Chiapas, Mexico. Ancient Mesoamerica 8(1): 111 136. Clark, John E. and Phil C. Weigand 2009 Obsidian lapidary without polishing. In Investigaciones Recientes sobre la Ltica Arqueolgica en Mxico edited by L. Mirambell and L. Gonzalez Arratia, pp. 79 93. Coleccin Cientfica no. 561. INAH, Mexico City. Clarkson Chris 2008 Changing Reduction Intensity, Settlement, and Subsistence in Wardaman Country, Northern Australia. In Lithic Technology: Measures of Production, Use, and Curation, edited by William Andrefsky, Jr., pp. 286 316. Cambri dge University Press Cowgill, George 2000 The Central Mexican Highlands f rom t he Rise o f Teotihuacan t o t he Decline of Tula In The Cambridge History of the Native Peoples of the Americas Volume II: Mesoamerica, Part 1, edited by Richard E.VI. Adams and Murdo J. MacLeod, pp, 250 317. Cambridge University Press Curtin, Philip D. 2000 The World and the West: The European Challenge and the Overseas Response in the Age of Empire. Cambridge University Press.

PAGE 270

255 Darling, J. Andrew 1993 Notes on obs idian sources of the Southern Sierra Madre Occidental. Ancient Mesoamerica 4:245 253. 1995 Acquisition and Distribution of Obsidian in the North Central Frontier of Mesoamerica Paper presented at the III Coloquio Pedro Bosch Gimpera: Rutas de Intercam bio en Mesoamerica, Mexico, D.F. Darras, Veronique 2012 Development of Pressure Blade Technology in North Central and West Mexico I n The Emergence of Pressure Blade Making: From Origin to Modern Experimentation edited by Pierre M. Desrosiers, p p. 417 462 S pringer Publications New York Demarest, Arthur A. 2013 Ideological Pathways to Economic Exchange: Religion, Economy, and Legitimation at the Classic Maya Royal Capital of Cancun. Latin American Antiquity 24(4): 371 402. De Len, Jason P. an d David M. Carballo 2003 Technology and Access. Chipped Stone from Middle Formative Tlaxcala, Mexico P aper presented at the 68th Annual Meetings of the Society for American Archaeology, Milwaukee. De Len, Jason P., Kenneth G. Hirth and David M. Carb allo 2009 Exploring Formative Period Obsidian Blade Trade: Three Distribution Models. Ancient Mesoamerica 20: 113 128. DeLucia, Kristin 2008 Looking Beyond Gender Hierarchy: Rethinking Gender at Teotihuacan, Mexico. Archeological Papers of the Ameri can Anthropological Association 18(1): 17 36. DeMarrais, E., L. J. Castillo, and T.K. Earle 1996 Ideology, Materialization, and power strategies. Current Anthropology 37: 15 31. Dockall, John E. and Harry J. Shafer 1993 Testing the Producer Consume r Model for Santa Rita Corozal, Belize. Latin American Antiquity 4(2): 158 179.

PAGE 271

256 Earl e Timothy, editor 1991 The evolution of chiefdoms. In Chiefdoms: Power, Economy, and Ideology, edited by Timothy Earle, pp. 1 15. Cambridge University Press. Esparz a Lpez Rodrigo 2003 La Obsidiana en el Contexto Arqueolgico de Los Guachimontones : Un recurso estratgico en el desarrollo de sociedades estatales. Revista del Seminario de Historia Mexicana IV : 73 95 Fernandez, Rodolfo, and Daria Deraga 1988 Reflexiones sobre las sociedades prehispanicas del occidente de Mexico. Revista Mexicana de Estudios Antropologicos 34: 343 353. Galvn Villegas, Luis Javier 199 1 Las Tumbas de Tiro del Valle de Atemajac. Instituto Nacional de Antropologa e Historia Mxico, D.F. Glascock, Michael D., Phil C. Weigand, Rodrigo Esparza Lpez Michael A. Ohnersorgen, Mauricio Garduo Ambriz, Joseph B. Mountjoy, and J. Andrew Darling 2 010 Geochemical Characterization of Obsidian in Western Mexico: The Sources in Jal isco, Nayarit, and Zacatecas. In Crossing the Straits: Prehistoric Obsidian Source Exploitation in the North Pacific Rim edited by Yaroslav Kuzmin and Michael D. Glascock pp. 201 217 Archeopress, Oxford. Guderjan, Thomas H. 2007 The Nature of an a ncient Maya City: Resources, Interaction, and Power at Blue Creek, Belize The University of Alabama Press, Tuscaloosa. Hammond Norman, Sara Donaghey, Colleen Gleason, J.C. Staneko, Dirk Van Tuerenhout, and Laura J. Kosakowski 1987 Excavations at Noh mul, Belize, 1985 Journal of Field Archaeology 14(3): 257 2 81. Haines, Helen R. 2008 Stingray Spine Use and Maya Bloodletting Rituals: A Cautionary Tale. Latin American Antiquity 19(1):83 98 Hayden, Brian 1995 Pathways to Power. In Foundations of Social Inequality, ed ited by T. Douglas Price and Gary M. Feinman, pp. 15 86. Plenum Press New York. 2009 Funerals as Feasts: Why Are They So Important? Cambridge Archaeological Journal 19(1): 21 52.

PAGE 272

257 Healan, Dan 2009 Ground Platform Preparation and the "Banalization" of Prismatic Blades. Ancient Mesoamerica 20: 103 111. Hendon, Julia A. 1991 Status and Power in Classic Maya Society: An Archeological Study. American Anthropologist 93(4): 894 918. Heredia Espinoza, Verenice Y. 2008 The aga ve landscape and its archaeological context in the Tequila Volcano area Report submitted to the Foundation for Ancient Mesoamerican Studies, Inc. Electronic document, accessed online from http://www.famsi.org/reports/07012/ on August 18, 2013. Hirth, Ken neth G. 1996 Political Economy and Archaeology: Perspectives on Exchange and Production. Journal of Archaeological Research 4(3): 203 239. 1998 The Distributional Approach: A New Way to Identify Marketplace Exchange in the Archaeological Record Cu rrent Anthropology 39(4): 451 476. Hirth, Kenneth and Bradford Andrews 2002 Pathways to Prismatic Blades: Sources of Variation in Mesoamerican Lithic Technology I n Pathways to Prismatic Blades: A Study in Mesoamerican Obsidian Core Blade Technology e dited by Kenneth Hirth and Bradford Andrews, pp. 1 14 T he Cotsen Institute of Archaeology, UCLA, Los Angeles Hirth, Kenneth and J. Jeffrey Flenniken 2002 Core Blade Technology in Mesoamerican Prehistory I n Pathways to Prismatic Blades: A Study in Me soamerican Obsidian Core Blade Technology edited by Kenneth Hirth and Bradford Andrews, pp. 121 129 T he Cotsen Institute of Archaeology, UCLA, Los Angeles. Hodder, Ian 1982 Theoretical Archaeology: a Reactionary View. In Symbolic and Structural Archa eology edited by I. Hodder, pp. 1 16. Cambridge University Press. H oedl, Lucas 2013 Shades of Black: Obsidian Distribution and Social Organization at the Teuchitln Tradition Site of Navajas During the Late Formative. Master's Thesis, Department of An thropology, University of Colorado, Denver. Electronic source, Auraria Library Internet Archive, accessed from https://ia601004.us.archive.org/26/items/shades_of_obsidian/etd_10243_sip1_m.p df on March 24, 2014

PAGE 273

258 1903 The Chichimecs and their ancient culture, with notes on the Tepecano and the ruin of La Quemada, Mexico. American Anthropologist 5:385 440. Hughes, R.E. and R.L. Smith 1993 Archaeology, geology, and geochemistry in obsidian prove nance studies. Geological Society of America Special Paper 283 Geological Society of America Boulder. Jackson, John B. 1984 Discovering the Vernacular Landscape. Yale University Press New Haven. Jac kson, Thomas, and Michael Love 1991 Blade Runni ng: Middle Preclassic Obsidian Exchange and the Introduction of Prismatic Blades at la Blanca, Guatemala. Ancient Mesoamerica 2: 47 59. Johns, Catherine. 2014 Spatial Analysis of Ceramics in Ritual Architecture of the Teuchitln Tradition sis, Department of Anthropology, University of Colorado Denver. Johnson, Jay K. 1996 Lithic analysis and questions of cultural complexity. In Stone Tools: Theoretical Insights into Human Prehistory, edited by George H. Odell, pp. 159 228. Plenum, New Y ork Joyce, Arthur A. 2004 Sacred Space and Social Relations in the Valley of Oaxaca. In Mesoamerican Archaeology: Theory and Practice edited by Julia A. Hendon and Rosemary A Joyce B lackwell, Malden, MA. Julien, C. K., and Julien, M. 1994 Preh istoric technology: A cognitive science? In The Ancient Mind, Elements of Cognitive Archaeology edited by C. Renfrew and E. Zubrow, pp.152 163 Cambridge University Press. Kelley, J. Charles 1971 Archaeology of the northern frontier: Zacatecas and Du rango. In Handbook of Middle American Indians, vol. 11, archaeology of Northern Mesoamerica, part two edited by Robert Wauchope Gordon Ekholm and Ignacio Bernal, pp. 768 801. University of Texas Press, Austin.

PAGE 274

259 Kelley, J. Charles 1974 Speculations on the Culture History of Northwestern Mesoamerica. In The Archaeology of West Mexico, edited by Betty Bell, pp. 19 39. Sociedad de Estudios Avanzados del Occidente de Mxico, Ajijic, Mexico. Kidder, A.V. 1945 Excavations at Kaminaljuyu, Guatemala American Antiquity 11(2):65 75. Knox, Gary W. 2013 Agave and Yucca: Tough Plants for Tough Times Electronic document University of Florida IFAS Extension EDIS, accessed online from http://edis.ifas.ufl.edu/ep419 on July 23, 2014 Kowalewski, Stev en A., Richard E. Blanton, Gary Feinman and Laura Finsten 1983 Boundaries, Scale and Internal Organization. Journal of Anthropological Archaeology 2: 31 36. Kristiansen, Kristian 1991 Chiefdoms, states, and systems of social evolution. In Chiefdom s: Power, Economy, and Ideology edited by Timothy Earle, pp. 16 43. Cambridge University Press. Lemonnier, P ierre 1986 The study of material culture today: Toward an anthropology of technical systems. Journal of Anthropological Archaeology 5: 147 186. Liot, Catherine, Susana Ramrez, Javier Raveles and Otto Schndube 2006 Transformaciones Socioculturales e Tecnolgicas en La Pea. In Transformaciones socioculturales y tecnolgicas en el sitio de La Pea, Cuenca de Sayula, Jalisco e dited by Catherine Liot, Susana Ramrez Javier Reveles and Otto Schndube, pp. 407 417. Universidad de Guadalajara/INAH, Guadalajara. Long, Stanley V. 1966 Archaeology of the Municipio de Etzatlan, Jalisco. Ph.D. d issertation, Department of Anthropology, University of California at Los Angeles. Lpez, Lorenza and Jorge Ramos de la Vega 1998 Excavating the Tomb at Huitzilapa. In Ancient West Mexico: Art and Archaeology of the Unknown Past edited by Richard F Townsend, pp. 89 105. Thames and Hudson New York. 2 006a Some Interpretations of the Huitzilapa Shaft Tomb. Ancient Mesoamerica 17(2) : 271 281. 2006b Organic Offerings, Paper, and Fibers from the Huitzilapa Shaft tomb, Jalisco, Mexico A ncient Mesoamerica 17(2) : 283 296.

PAGE 275

260 Lumholtz, Carl 1903 Unk nown Mexico: Volume II. Macmillan and Co., London. Magne, Martin P.R. 1989 Lithic Reduction Stages and Assemblage Formation Processes In Experiments in Lithic Technology edited by Daniel S. Amick and Raymond P. Mauldin pp. 15 31 B AR International Series 528 British Archaeological Reports, Oxford. 2001 Debitage Analysis as a Scientific Tool I n Lithic Debitage: Context, form, meaning, edited by Willliam Andrefsky, Jr pp. 21 30 U niversity of Utah Press Salt Lake City. Mallory, John K. 198 6 "Workshops" and "Specialized Production" in the Production of Maya Chert Tools: A Response to Schafer and Hester. American Antiquity 51(1): 152 158. Martin, Simon and Nikolai Grube 2000 Chronicle of the Maya Kings and Queens: Deciphering the Dynas ties of the Ancient Maya. Thames and Hudson, London. McAnany, Patricia A. 1989 Stone Tool Production and Exchange in the Eastern Maya Lowlands: The Consumer Perspective from Pulltrouser Swamp, Belize. American Antiquity 54: 332 346. 1995 Living wi th the Ancestors University of Texas Press Austin Moholy Nagy, Hattula 1990 The Misidentification of Mesoamerican Lithic Workshops. Latin American Antiquity 1(3): 268 279. 1999 Mexican Obsidian at Tikal, Guatemala. Latin American Antiquity 10( 3): 300 313. Moholy Na gy, Hattula and Fred W. Nelson 1990 New Data on Sources of Obsidian Artifacts from Tikal, Guatemala. Ancient Mesoamerica 1(1): 71 80 Moholy Nagy, Hattula, James Meierhoff, Mark Golitko, and Caleb Kestle 2013 An Analysis of pX RF Obsidian Source Attributions from Tikal, Guatemala. Latin American Antiquity 24(1): 72 97.

PAGE 276

261 Morell Hart, Shanti, Rosemary A. Joyce and John S. Henderson 2014 Multi Proxy Analysis of Plant Use at Formative Period Los Naranjos, Honduras. Latin Americ an Antiquity 25(1): 65 81. Nassaney, Michael S. 1996 The role of chipped stone in the Political Economy of Social Ranking. In Stone Tools: Theoretical Insights into Human Prehistory, edited by George H. Odell, pp. 181 224. Plenum, New York. Newman, Ja y R. 1994 The Effects of Distance on Lithic Material Reduction Technology Journal of Field Archaeology 21(4): 491 501. Odell, George H., ed. 1996 Introduction to Stone Tools: Theoretical Insights into Human Prehistory pp. 1 5. Plenum, New York. Ohnersorgan, Michael A. and Mark D. Varien. 1996 Formal Architecture and Settlement Organization in Ancient West Mexico. Ancient Mesoamerica 7: 103 120. Parr, Robert E. 2007 Protein Residue Analysis of Twelve Obsidian Artifacts from the Site of Nav ajas, State of Jalisco, Mexico. In Tequila Valley Regional Archaeological Project Season Report 2003 edited by Christopher S. Beekman, pp. 223 233 M anuscript on file, Anthropology Department, University of Colorado Denver. Parry, William J. 1987 Chip ped Stone Tools in Formative Period Oaxaca, Mexico: their Procurement, Production, and Use. Memoirs of the Museum of Anthropology No. 20 University of Michigan, Ann Arbor. Pendergast, David M. 1971 Evidence of Early Teotihucan Lowland Maya Contact at Altun Ha. American Antiquity 36(4): 455 460. Phillips, Stephen C. 2011 Networked Glass: Lithic Raw Material Consumption and Social Networks in the Kuril Islands, Far Eastern Russia Ph.D. d issertation, Department of Anthropology, University of Washingt on Seattle Electronic source, accessed online from https://digital.lib.washington.edu/researchworks/bitstrx`eam/handle/1773/16354/P hillips_Diss_2011.pdf?sequence=1 on April 25, 2014.

PAGE 277

262 Pollard, Helen P. 2003 The Tarascan Empire. In The Postclassic M esoamerican World, edited by Michael E. Smith and Frances F. Berdan, pp. 78 86. University of Utah Press Salt Lake City. Pollard, Helen P., and L. Cahue 1999 Mortuary patterns of regional elites in the Lake Patzcuaro Basin of western Mexico. Latin Ame rican Antiquity 10: 259 280. Reveles, Javier 2005 La Litica de la Cuenca de Sayula I n Arqueologia de la Cuenca de Sayula edited by F. Valdez, Otto Shndube and J.P. Emphoux, pp. 349 368 UDG IRD, Guadalajara. 2006 Analisis Preliminar de los Mate riales Liticos. In Transformaciones socioculturales y tecnolgicas en el sitio de La Pea, Cuenca de Sayula, Jalisco e dited by Catherine Liot, Susana Ramires, Javier Reveles and Otto Schndube, pp. 373 389. Universidad de Guadalajara/INAH, Guadalajara. R eilly, F. Kent III 1996 Art, Ritual, and Rulership in the Olmec World. In The Olmec World: Ritual and Rulership P rinceton University Art Museum, Princeton. Root, Matthew J. 2004 Technological Analysis of Flake Debris and the Limitations of Size Gra de Techniques I n Aggregate Analysis in Chipped Stone, edited by Christopher T. Hall and Mary Lou Larson, pp. 65 94 U niversity of Utah Press Salt Lake City. Schneider, Jane 1977 Was There a Precapitalist W orld System? Peasant Studies VI(1): 20 29. Schndube, Otto 1998 Natural Resources and Human Settlements in Ancient West Mexico I n Ancient West Mexico: Art and Archaeology of the Unknown Past edited by Richard F. Townsend pp. 205 215. Thames and Hudson, New York. Schortman, Edward M. and P atricia A. Urban 1987 Modeling interregional interaction in prehistory. In Advances in Archeological Method and Theory V. 11, pp. 37 95 Academic Press New York. Secretaria de la Presidencia (SP) 1974a Carta Etzatln F 13 D 53 e Soils map, 1:50,0 00. SP. Mexico, D.F. 1974b Carta Etzatln F 13 D 53 g Geology map, 1:50,000. SP.Mexico, D.F.

PAGE 278

263 Secretaria de la Presidencia (SP) 1974c Carta Etzatln F 13 D 53 u Land usage map, 1:50,000. SP. Mexico, D.F. 1974d Carta Tala F 13 D 64 e Soils map, 1:50,000. SP. Mexico, D.F. 1974e Carta Tala F 13 D 64 g Geology map, 1:50,000. SP. Mexico, D.F. 1974f Carta Tala F 13 D 64 u Land usage map, 1:50,000. SP. Mexico, D.F. Sellet, Frdrick 1993 Chaine Operatoire; The Concept a nd i ts App lications. Lithic Technology 18:106 112. Sempowski, Martha L. 1992 Economic and Social Implications of Variations in Mortuary Practices at Teotihuacan. In Art, Ideology, and the City of Teotihuacan: A Symposium at Dumbarton Oaks, 8th and 9th October, 1 988 edited by J. C. Berlo, p p. 27 58. Dumbarton Oaks Washington, D.C. Sempowski, Martha L. and Michael W. Spence 1994 Mortuary Practices and Skeletal Remains at Teotihuacan University of Utah Press Salt Lake City. Shafer, H.J. and T.R. Hester. 198 6 Maya Stone Tool Craft Specialization and Production at Colha, Belize: Reply to Mallory. American Antiquity 51: 158 166. Shaw, Justine M. 2003 Climate Change and Deforestation: Implications for the Mayan Collapse. Ancient Mesoameric a 14: 157 167. Sheehy, James J. 1991 Structure and Change in a Late Classic Maya Domestic Group at Copan, Honduras. Ancient Mesoamerica 2(1): 1 20. Sheets, P ayson D 1975 Behavioral analysis and the structure of a prehistoric industry. Current Anthropology 16:369 391. S heets, Payson D. and G.R. Muto 1972 Pressure Blades and Total Cutting Edge: An Experiment In Lithic Technology. Science 175(4022): 632 634.

PAGE 279

264 Shott, Michael J. 1994 Size and Form in the Analysis of Flake Debris: Review and Recent Approaches. Journal of Archaeological Method and Theory 1(1): 69 110. 2003 Chane Opratoire and Reduction Sequence. Lithic Technology 28(2):95 105. Smith, Michael E. 1997 Life in the Provinces of the Aztec Empire. Scientific American 277(3 ): 76 83. Soto de Arechavaleta Mara de los Dolores 1982 Anlisis de la Tecnologa de Produccin del Taller de Obsidiana de Guachimontn, Teuchitln, Jalisco. Tesis de Licenciatura en Arqueologa, Escuela Nacional de Antropologa y Historia, Mxico, D.F. 1990 Areas de Actividad en un Taller de Manufactura de Implementos de Piedra Tallada I n Nuevos Enfoques en el Estudio de la Ltica edited by Mara de los Dolores Soto de Arechavaleta pp. 215 242. UNAM, Mexico City Southall, Aidan 1988 The Segmentary State in Africa and Asia. Comparative Studies In Society and History 30: 52 82. Spence, Michael W. 1981 Obsidian Production and the State in Teotihuacn. American Antiquity 46(4): 769 788. 1987 The Scale and Structure of Obsidian Produ ction in Teotihuacn. In Teotihuacn: nuevos datos, nuevas sntesis, nuevos problemas edited by Emily McClung de Tapia and Evelyn Rattray, pp. 429 450. Universidad Nacional Autnoma de Mxico, Mxico City 1996 Commodity or Gift: Teotihuacan Obsidian in the Maya Region. Latin American Antiquity 7(1): 21 39. Spence, Michael, Phil Weigand, and Dolores Soto de Arechavaleta 1980 Obsidian exchange in W est Mexico. In Rutas de Intercambio en Mesoamrica y el Norte de Mxico, XVI Mesa Redonda Tomba 1 p p. 357 361. Sociedad Mexicana de Antropologa, Mxico City D.F. 2002 Production and Distribution of Obsidian Artifacts in Western Jalisco. In Pathways to Prismatic Blades: A Study in Mesoamerican Obsidian Core Blade Technology edited by Kenneth Hirt h and Bradford Andrews, pp. 61 80 T he Cotsen Institute of Archaeology, UCLA, Los Angeles.

PAGE 280

265 Stein, Gil 1999 Rethinking World Systems : Diasporas, Colonies, and Interaction in Uruk Mesopotamia. University of Arizona Press, Tucson. 2014 Ec onomic Dominance, Conquest, or Interaction among Equals? Theoretical Models for Understanding Culture Contact in Early Near Eastern Complex Societies. In Proceedings of The International Congress of Young Archaeologists 2014 edited by Mohammad HosseinAziz iKharanaghi MortezaKhanipour and Reza Naseri pp. 55 67. University of Tehran Press. Stemp, W. James, Christophe G. B. Helmke, Jaime J. Awe, Tristan Carter, and Sarah Grant 2012 A Green Obsidian Eccentric from Actun Uayazba Kab, Belize. In Heart of E arth: Studies in Maya Ritual Cave Use AMCS Bulletin 23 pp. 111 124 Association for Mexican Cave Studies, Austin. Stemp, W. James, Mason D. Andruskiewicz, Matthew A. Gleason and Yusuf H. Rashid 2014 Experiments in ancient Maya bloodletting: quantific ation of surface wear on obsidian blades. Archaeological and Anthropological Sciences. Electronic publication, Springer, accessed online from http://link.springer.com/article/10.1007/s12520 014 0204 5 in September 2014 Stuart, G. 2003 Pre Hispanic Soc iopolitical Development and Wetland Agriculture in the Tequila Valleys of West Mexico Ph.D. dissertation, Department of Anthropology, Arizona State University, Tempe. Thomas, Jobin, Sabu Joseph, K.P. Thrivikramji, and K.S. Arunkumar in press c orrected proof S ensitivity of digital elevation models: The scenario from two tropical mountain river basins of the Western Ghats, India. Geoscience Frontiers Electronic document, accessed online from http://www.sciencedirect.com/science/article/pii/S16749871140 00036 on May 27, 2014. Tighe, M. Lorraine and Drew Chamberlain 2009 Accuracy Comparison of the SRTM, NED, NEXTMap USA Digital Terrain Model over Seveal USA Study Sites. Paper submitted to the ASPRS/MAPPS 2009 Fall Conference, November 16 19, 2009 San Antonio E lectronic document accessed online from http://www.asprs.org/a/publications/proceedings/sanantonio09/Tighe_2.pdf, on May 27, 2014.

PAGE 281

266 Tomka, Steven A. 1989 Differentiation Lithic Reduction Techniques: An experimental approach I n Experiments i n Lithic Technology edited by Daniel S. Amick and Raymond P. Mauldin pp. 137 161. B AR International Series 528 British Archaeological Reports Oxford. Townsend, Richard F 1 992 The Aztecs. Thames and Hudson London 1998a Before Gods, Before Kings In Ancient West Mexico: Art and Archaeology of the Unknown Past edited by Richard F. Townsend, pp. 107 135. Thames and Hudson, New York. 1998b Introduction: Renewing the Inquiry in Ancient West Mexico In Ancient West Mexico: Art and Archaeolog y of the Unknown Past edited by Richard F. Townsend, pp. 15 31. Thames and Hudson, New York. Turner, V.W. 1974 Dramas, Fields and Metaphors: Symbolic Action in Human Society Cornell University Press Ithaca, New York Tyndall, Gregory and Christoph er S. Beekman 2007 Ceramic Analysis. In Tequila Valley Regional Archaeological Project Season Report 2003 edited by Christopher S. Beekman, pp. 154 183 M anuscript on file, Anthropology Department, University of Colorado Denver. Valdez, Francisco 199 8 The Sayula Basin: Ancient Settlements and Resources. In Ancient West Mexico : art and archaeology of the unknown pas t edited by Richard F. Townsend, pp. 217 231. Art Institute of Chicago. Wacquant, L.J.D. 1989 Towards a Reflexive Sociology: A W orkshop with Pierre Bourdieu. Sociological Theory 7: 26 63. Wallerstein, Immanuel 1974 The Rise and Future Demise of the World Capitalist System: Concepts for Comparative Analysis. Comparative Studies in Society and History 16(4): 387 415. Weigand, P hil C. 1975 Circular Ceremonial Structure Complexes in the Highlands of Western Mexico. In Archaeological Frontiers: Papers on New World High Cultures in Honor of J. Charles Kelley edited by Robert Pickering, pp. 183 227. Southern Illinois University, Carbondale.

PAGE 282

267 Weigand, Phil C. 1985 Evidence for Complex Societies during the Western Mesoamerican Classic Period. In The Archaeology of West and Northwest Mesoamerica, edited by Michael S. Foster and Phil C. Weigand, pp. 47 91. Westview Press, Boulder 1993 Large Scale Hydraulic Works in Prehistoric Western Mesoamerica. In Economic Aspects of Water Management in the Prehispanic New World, edited by Vernon L. Scarborough and Barry L. Isaac, pp. 223 262. Research in Economic Anthropology, Supplemen t no. 7. JAI Press, Greenwich, Connecticut 1996 La evolucin y ocaso de un ncleo de civilizacin: la Tradicin Teuchitln y la arqueologa de Jalisco. In Las Cuencas del Occidente de Mxico ( poca Prehispnica), edited by Eduardo Williams and Phil C. Weigand, pp. 185 245. Colegio de Michoacn Zamora Mxico 2000 The Evolution and Decline of a Core of Civilization: The Teuchitln Tradition and the Archaeology of Jalisco. In Greater Mesoamerica: The Archaeology of West and Northwest Mexico, edited by Michael S. Foster and Shirley Gorenstein, pp. 43 58 University of Utah Press, Salt Lake City. 2008 Turquoise: Formal economic interrelationships between Mesoamerica and the North American Southwest. In Archaeology without Borders: Contact, Commerc e, and Change in the U.S. Southwest and Northwestern Mexico edited by L.D. Webster and M.E. McBrinn pp. 343 353 University of Colorado Press, Boulder Weigand, Phil C. and Christopher S. Beekman 1998 The Teuchitlan Tradition: Rise of a Statelike So ciety. In Ancient West Mexico: Art and Archaeology of the Unknown Past edited by Richard F. Townsend pp. 35 51. Thames and Hudson, New York. Wells, E. Christian 2006 Recent Trends in Theorizing Prehispanic Mesoamerican Economies. Journal of Archaeolo gical Research 14: 265 312. Wells, E. Christian and Patricia A. McAnany, editors. 20 12 Dimensions of Ritual Economy. JAI Press, Emerald Group Publishing, Bingley, United Kingdom. Whittaker, John C. 1994 Flintknapping: Making & Understanding Stone T ools. University of Texas Press, Austin. Whittaker, John C., Douglas Caulkins and Kathryn A. Kamp 1998 Evaluating Consistency in Typology and Classification. Journal of Archaeological Method and Theory 5(2): 129 164.

PAGE 283

268 Whittaker, John C. and Eric J. Kaldahl 2001 Where the Waste Went: A Knappers' Dump at Grasshopper Pueblo I n Lithic Debitage: Context, form, meaning, edited by Willliam Andrefsky, Jr pp. 32 60 U niversity of Utah Press Salt Lake City. Witmore, Christopher L. 1998 Sacred Sun Centers. In Ancient West Mexico: Art and Archaeology of the Unknown Past edited by Richard F Townsend, pp. 137 149. Thames and Hudson, New York. Wolf, Eric 1982 Europe and the people without history. University of California Press, Berkeley

PAGE 284

269 APPENDIX A CARTA LAND USE AND SOIL MAP SECTIONS. This appendix contains excerpts from the official Mexican government CARTA maps for soils, land use and geography. Soil unit definitions are from the Food an d Agriculture Organization, and are described at http://www.britannica.com/search?query=FAO SOIL Figure A.1: Soil types around Llano Grande. From Carta Etzatln F 13 D 53 e soil map.

PAGE 285

270 Figure A.2: Geology around Llano Grande. From Carta Etzatln F 13 D 53 g geology map.

PAGE 286

271 Figure A.3: Land use surrounding Llano Grande. From Carta Etzatln F 13 D 53 u land use map.

PAGE 287

272 Figure A.4: Soil types around Navajas. From Carta Tala F 13 D 64 e soil map.

PAGE 288

273 Figure A.5: Geology around Navajas. From Carta Tala F 13 D 53 g geology map.

PAGE 289

274 Figure A.6: Land use around Navajas. From Carta Tala F 13 D 64 u land use map.

PAGE 290

275 Figure A.7: Soil types around Los Guachimontones. From Carta Tala F 13 D 64 u soil map.

PAGE 291

276 Figure A.8: Geology around Los Guachimontones. From Carta Tala F 13 D 64 g geology map

PAGE 292

277 Figure A.9: Current land use around Los Guachimontones. From Carta Tala F 13 D 64 g geol ogy map

PAGE 293

278 APPENDIX B SCATTERPLOTS OF SCAR COUNTS AND SIZE MEASUREMENTS All metric units are centimeters. Chart 1: Length x width vs. scar count at Llano Grande Chart 2: Length x width vs. scar count at Navajas

PAGE 294

279 Chart 3: Length vs. sc ar count at Llano Grande Chart 4: Length vs. scar count at Navajas

PAGE 295

280 Chart 5: Width vs. scar count at Llano Grande Chart 6: Width vs. scar count at Navajas